Cross-linguistic Influences in the Second Language Lexicon 9781853598579

Citation preview

Cross-linguistic Influences in the Second Language Lexicon

SECOND LANGUAGE ACQUISITION Series Editor: Professor David Singleton, Trinity College, Dublin, Ireland This new series will bring together titles dealing with a variety of aspects of language acquisition and processing in situations where a language or languages other than the native language is involved. Second language will thus be interpreted in its broadest possible sense. The volumes included in the series will all in their different ways offer, on the one hand, exposition and discussion of empirical findings and, on the other, some degree of theoretical reflection. In this latter connection, no particular theoretical stance will be privileged in the series; nor will any relevant perspective – sociolinguistic, psycholinguistic, neurolinguistic, etc. – be deemed out of place. The intended readership of the series will be final-year undergraduates working on second language acquisition projects, postgraduate students involved in second language acquisition research, and researchers and teachers in general whose interests include a second language acquisition component. Other Books in the Series Effects of Second Language on the First Vivian Cook (ed.) Age and the Acquisition of English as a Foreign Language María del Pilar García Mayo and Maria Luisa García Lecumberri (eds) Fossilization in Adult Second Language Acquisition ZhaoHong Han Silence in Second Language Learning: A Psychoanalytic Reading Colette A. Granger Age, Accent and Experience in Second Language Acquisition Alene Moyer Studying Speaking to Inform Second Language Learning Diana Boxer and Andrew D. Cohen (eds) Language Acquisition: The Age Factor (2nd edn) David Singleton and Lisa Ryan Focus on French as a Foreign Language: Multidisciplinary Approaches Jean-Marc Dewaele (ed.) Second Language Writing Systems Vivian Cook and Benedetta Bassetti (eds) Third Language Learners: Pragmatic Production and Awareness Maria Pilar Safont Jordà Artificial Intelligence in Second Language Learning: Raising Error Awareness Marina Dodigovic Studies of Fossilization in Second Language Acquisition ZhaoHong Han and Terence Odlin (eds) Language Learners in Study Abroad Contexts Margaret A. DuFon and Eton Churchill (eds) Early Trilingualism: a Focus on Questions Julia D. Barnes

For more details of these or any other of our publications, please contact: Multilingual Matters, Frankfurt Lodge, Clevedon Hall, Victoria Road, Clevedon, BS21 7HH, England http://www.multilingual-matters.com

SECOND LANGUAGE ACQUISITION 17 Series Editor: David Singleton, Trinity College, Dublin, Ireland

Cross-linguistic Influences in the Second Language Lexicon Edited by

Janusz Arabski

MULTILINGUAL MATTERS LTD Clevedon • Buffalo • Toronto

Library of Congress Cataloging in Publication Data Cross-linguistic Influences in the Second Language Lexicon Edited by Janusz Arabski. Second Language Acquisition: 17 Includes bibliographical references and index. 1. Language transfer (Language learning). 2. Interlanguage (Language learning). 3. Languages in contact. 4. Second language acquisition. 5. Intercultural communication. 6. Vocabulary–Study and teaching. I. Arabski, Janusz. II. Second Language Acquisition (Clevedon, England): 17. P118.25.C76 2006 418–dc22 2005014689 British Library Cataloguing in Publication Data A catalogue entry for this book is available from the British Library. ISBN 1-85359-856-9 / EAN 978-1-85359-856-2 (hbk) Multilingual Matters Ltd UK: Frankfurt Lodge, Clevedon Hall, Victoria Road, Clevedon BS21 7HH. USA: UTP, 2250 Military Road, Tonawanda, NY 14150, USA. Canada: UTP, 5201 Dufferin Street, North York, Ontario M3H 5T8, Canada. Copyright © 2006 Janusz Arabski and the authors of individual chapters. All rights reserved. No part of this work may be reproduced in any form or by any means without permission in writing from the publisher. Typeset by Saxon Graphics. Printed and bound in Great Britain by the Cromwell Press Ltd.

Contents Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Part 1: Language Contact and Language Transfer Revisited 1 On the Ambiguity of the Notion ‘Transfer’ Hans W. Dechert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Language Transfer in Language Learning and Language Contact Janusz Arabski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3 Could a Contrastive Analysis Ever be Complete? Terence Odlin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 4 The Importance of Different Types of Similarity in Transfer Studies Håkan Ringbom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 5 Language Contact vs. Foreign and Second Language Acquisition ElČbieta MaÚczak-Wohlfeld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Part 2: Language Contact Observed 6 Genre: Language Contact and Culture Transfer Andrzej Ryda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 7 Is Cross-Linguistic Influence a Factor in Advanced EFL Learners’ Use of Collocations? Justyna Leïniewska . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 8 International Terms and Profile Transfer: On Discussion Krystyna WarchaÙ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 9 The Influence of English on Polish Drug-related Slang Magdalena BartÙomiejczyk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Part 3: Lexical Transfer in Language Processing 10 Why Money Can’t Buy You Anything in German: A Functional-Typological Approach to the Mapping of Semantic Roles to Syntactic Functions in SLA Marcus Callies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 v

9612Arabski.indb v

13/01/2006 08:39:36

vi

Contents

11 Lexical Transfer: Interlexical or Intralexical? David Singleton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 12 The Interaction of Languages in the Lexical Search of Multilingual Language Users Danuta Gabryï-Barker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 13 Assessing L2 Lexical Development in Early L2 Learning: A Case Study Anna NiČegorodcew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 14 Code-mixing in Early L2 Lexical Acquisition Joanna Rokita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Part 4: Lexical Transfer in Fixed Expressions 15 Metaphorical Transferability Rüdiger Zimmermann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 16 On the Use of Translation in Studies of Language Contact Jolanta Latkowska. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 17 On Building Castles on the Sand, or Exploring the Issue of Transfer in the Interpretation and Production of L2 Fixed Expressions Anna Cieïlicka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 18 ‘Don’t Lose Your Head’ or How Polish Learners of English Cope with L2 Idiomatic Expressions Liliana Piasecka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 19 Phrasal Verb Idioms and the Normative Concept of the Interlanguage Hypothesis PrzemysÙaw Olejniczak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

9612Arabski.indb vi

13/01/2006 08:42:34

Contributors Janusz Arabski, University of Silesia, 41-205 Sosnowiec, ul. …ytnia 10 Poland. Magdalena BartÙomiejczyk, University of Silesia, ul. Zytnia 10, 41-205 Sosnowiec, Poland. Marcus Callies, Philipps-Universität Marburg, Institut für Anglistik und Amerikanistik, Englische Sprachwissenschaft, Wilhelm-Röpke-Str. 6 D, 35032 Marburg, Germany. Anna Cieïlicka, School of English, Adam Mickiewicz University, Al. NiepodlegÙoïci 4, 61-874 PoznaÚ, Poland. Hans W. Dechert, University of Kassel, Germany. Danuta Gabryï-Barker, University of Silesia, Institute of English, 41-205 Sosnowiec, ul. …ytnia 10, Poland. Jolanta Latkowska, University of Silesia, Ul. …ytnia 10, 41-205 Sosnowiec, Poland. Justyna Leïniewska, English Department, Jagiellonian University, Al. Mickiewicza 9, 31-120 Kraków, Poland. Andrzej Ryda, University of Silesia, Institute of English, ul. …ytnia 10 41205 Sosnowiec, Poland. ElČbieta MaÚczak-Wohlfeld, Jagiellonian University, Home address:, Robzowska 9/3, PL 31-139 Kraków, Poland. Anna NiČegorodcew, English Department, Jagiellonian University, Al. Mickiewicza 9, 31-120 Kraków, Poland. Terence Odlin, Ohio State University, USA. vii

9612Arabski.indb vii

13/01/2006 08:42:35

viii

Contributors

PrzemysÙaw Olejniczak, I Nauczycielskie Kolegium J¿zyków Obcych w Sosnowcu, (English Teacher Training College), ul. ZegadÙowicza 1, 41200 Sosnowiec, Poland. Liliana Piasecka, Opole University, Opole, ul. Oleska 48, Poland. Håkan Ringbom, Åbo Akademi University, Finland. Joanna Rokita, Akademia Pedagogiczna – NKJA, ul. Podchor­Čych 2, 30084 Kraków, Poland. David Singleton, CLCS, Arts Building, Trinity College, Dublin 2, Ireland. Krystyna WarchaÙ, Institute of English, University of Silesia, …ytnia 10, 41205 Sosnowiec, Poland. Rüdiger Zimmermann, Institut für Anglistik und Amerikanistik, PhilippsUniversität Marburg, D-35032 Marburg, Germany.

9612Arabski.indb viii

13/01/2006 08:42:36

Preface This volume consists of a selection of papers analysing the phenomenon of language transfer which results from language contact in bilingual and multilingual language acquisition and learning contexts. The major focus of the volume is on the lexical level of language transfer. In the first instance, it returns to the debate surrounding the definition of language transfer and contact in Part 1 entitled Language Contact and Language Transfer Revisited – where some of the more theoretical aspects of the subject are discussed. The first part includes a discussion on the ambiguity of the term ‘transfer’ itself, beyond its meaning as a psycholinguistic concept. Then the relations between language contact and language learning are discussed. This part also includes an original view of contrastive analysis and a discussion on the importance of cross-linguistic similarity in foreign language learning. The volume then seeks to illuminate the theoretical discussion by presenting a corpus of data and its analysis at various levels of lexical competence (e.g. formulaic language, collocations, idioms, etc.), which are the product of research findings in various contexts of language use. Part 2, entitled Language Contact Observed, consists of papers that are corpus based. They present the transfer processes in discourse organisation, the transfer of genre and the use of collocations by language learners as a result of language contact. The theoretical issues of language contact are also present in connection with a study on drug-related loans in Polish and a semantic equivalence study of the international term ‘discussion’. Part 3 is Lexical Transfer in Language Processing. It consists of five papers. Two empirical works use translation tasks to study interference as a result of L2 acquisition and interference during lexical search processes in L1, L2 and L3. Two others report early L2 acquisition of lexis, concentrating on code-mixing and the role of instructional input in early L2 lexical development. The main theoretical element of the part is the article on lexical transfer which argues that the traditional concept of cross-lexical transfer still remains valid. The last part, Lexical Transfer in Fixed Expressions, is the methodological continuation of the previous part. It deals with the acquisition and ix

9612Arabski.indb ix

13/01/2006 08:42:36

x

Preface

processing of L2 idioms, phrasal verbs, fixed phrases, collocations, and figurative expressions. The elicitation techniques used by the authors were translation and comprehension tasks. One study deals with translating L1 idioms into L2. Most of the data shows the variety of techniques employed to translate the fixed expressions, including transfer strategies. Janusz Arabski

9612Arabski.indb x

13/01/2006 08:42:37

Part 1

Language Contact and Language Transfer Revisited

9612Arabski.indb 1

13/01/2006 08:42:38

9612Arabski.indb 2

13/01/2006 08:42:38

Chapter 1

On the Ambiguity of the Notion ‘Transfer’ HANS W. DECHERT Introduction In the ‘Practical Information’ provided by the organisers of the conference on ‘Language Contact and Language Transfer’ a portion of the text reads like this: The fee [for this conference] is all-inclusive and will be paid at the conference desk. There is a possibility, however, of transferring the payment to our bank account earlier, for which the participants will be given an “early bird bonus” […] Earlier transfers must be paid to our bank account at [name of Polish Bank Institute], with additional annotation SZCYRK-ARABSKI, by 30 April 2003. The annotation is very important, as this ensures the allocation of the money in the right sub-account. The participants are requested to fax their certificates of account payment to the Institute. (Emphasis through the italicised words is mine) Since I have not only been interested in being attributed the privileged status of ‘early bird’, but in the use and meaning of the verb ‘to transfer’ (transferring) and the noun ‘transfer’ relevant to this information, I have got involved with this problem relating to the topic of my paper. The following considerations are the result of my investigation through looking at ordinary and etymological dictionaries and G. Fauconnier’s (1997) seminal work Mappings in Thought and Language. According to the Oxford English Dictionary (1961), Vol. XI, 257: – the verb to transfer means to convey or take from one place, person, etc. to another, to transmit, transport, to give or hand from one to another, – to convey or make over (title, right, or property) by deed or legal process. (No question, the status ‘early bird’ , according to this entry, could be transferred on to me, but that is a different question) 3

9612Arabski.indb 3

13/01/2006 08:42:38

4

Cross-linguistic Influences in the Second Language Lexicon

– the noun transfer in Law stands for conveyance from one person to another, spec. of shares or stock. (I have been wondering, of course, whether this legal connotation of the term suggesting an explicit person-to-person exchange of financial property would permit the holder of the annotated sub-account to claim legal possession of the money) – the act of transferring or fact of being transferred; conveyance or removal from one place, person, etc. to another; transference, transmission. According to the New Oxford American Dictionary (2001), p. 1797 the connotation between transfer and transference of money is more evident: – the verb to transfer (to trans’fer; to ‘transfer: note the transfer of stress shift in modern English) means to move (someone or something) from one place to another. Ex: he would have to transfer to his own account. – the noun stands for an act of moving something or someone to another place. Ex: a transfer of wealth to the poorer nations. These entries, to come back to the first possibility of paying the conference fee mentioned in the information, seem to indicate that, appearing at the congress desk in a one-to-one person context I might have said: ‘May I transfer the conference fee to you?’ The answer most likely would have been: ‘When may we expect to receive the money?’ My invented statement, in other words, would have indicated an inappropriate linguistic transfer caused by my dictionary studies totally neglecting the inappropriate meaning of the term blended with a different analogical mental space, one of the sources of the ambiguity of the term transfer. Or to put it differently, my question would have been an intralingual pragmatic error, not caused by a crosslinguistic interaction with my primary language, but by a lack of expanding the term’s standard meaning to its analogical or figurative meaning. The second case of payment referred to in the Introduction is much more complicated in that there is no actual or very little person-to-person ‘transference’ involved, but on both ends and countries of the implicit communication process a large number of bureaucratic business activities between persons and banks, different currencies and dynamic exchange modalities, paper work as well as electronic information exchange, etc. This short outline of the complexity and shortcomings of the recommended kind of payment, quite in accordance with national and international financial vocabulary labelled ‘transfer’, referring to the early bird solution, may suffice to justify the expansion of the given dictionary meaning of transfer and the underlying ambiguity of the notion.

9612Arabski.indb 4

13/01/2006 08:42:40

On the Ambiguity of the Notion ‘Transfer’

5

According to the Oxford Dictionary of English Etymology (1966: 936) the English verb to transfer is related to French transférer and to the Latin transferre. The basis of the Latin verb is the Greek verb µεταφερειν. The English noun metaphor according to the same dictionary (p. 572) means a figure of speech involving the transference of a name to something analogous. The figurative meaning of transfer, in other words, depends on a relationship or mapping with analogy and metaphor in a blended mental space. This mapping, according to the Oxford English Dictionary (1961: Vol. XI: 257) is already documented in an entry of 1586, referring to Day: Metaphora, which is when a word from the proper or right signification is transferred to another neere vnto the meaning, or another one of 1883 referring to Murray: As the primitive sense (of words) has been … transferred boldly to figurative and analogical uses. Transfer, analogy, and metaphor, historically and etymologically speaking, are candidates to be related to and mapped onto each other in a common mental space of cognition.

Analogical Transfer Much of the perennial discussion and disagreement concerning the effects of the processes of intralingual and interlingual interaction in the acquisition of languages, called transfer or interference, seems to be caused by the ambiguity of these terms. This ambiguity is the result of the theoretical, methodological and empirical complexity and the controversial foundation and goals of linguistically oriented transfer research. A short look at the experimental treatment and discussion of the famous radiation problem within the framework of the problem-solving paradigm, introduced by Duncker (1945) and its later application to transfer research in general may by of particular interest. In his dealing with problem solution phenomena Duncker confronted his subjects with the following problem, among others: Problem no. 2. Given a human being with an inoperable stomach tumour, and rays which destroy organic tissue at sufficient intensity, by what procedure can one free him of the tumour by these rays and at the same time avoid destroying the healthy tissue which surrounds it? (Quoted from Johnson-Laird & Wason, 1977: 15)

9612Arabski.indb 5

13/01/2006 08:42:41

6

Cross-linguistic Influences in the Second Language Lexicon

The solution to this and other similar problems through the invocation of past experience follows a principle which Duncker calls ‘resonance’. The failure to do so Duncker attributes to what he calls ‘functional fixity’ (Johnson-Laird & Wason, 1977: 15–17). Needless to say, Duncker’s concept of ‘resonance’ of past experience in the solution of a new analogical problem is what we now conceptualise as analogical transfer, based on knowledge of the world. Holyoak’s (1984) and Gick & Holyoak’s (1980, 1983) experimental work follows Duncker’s early research closely in that it relates his radiation problem to similar analogical problems such as the Attack-Dispersion Story (cf. Figure1.1). The Attack-Dispersion Story A fortress was located in the center of the country. Many roads radiated out from the fortress. A general wanted to capture the fortress with his army. The general wanted to prevent mines on the roads from destroying his army and neighboring villages. As a result the entire army could not attack the fortress along on the road. However, the entire army was needed to capture the fortress. So an attack by one small group would not succeed. The general therefore divided his army into several small groups. He positioned the small groups at the heads of different roads. The small groups simultaneously converged on the fortress. In this way the army captured the fortress. The Radiation Problem A tumor was located in the interior of a patient’s body. A doctor wanted to destroy the tumor with rays. The doctor wanted to prevent the rays from destroying healthy tissue. As a result the high-intensity rays could not be applied to the tumor along one path. However, high-intensity rays were needed to destroy the tumor. So applying one low-intensity ray would not succeed. The doctor therefore divided the rays into several low-intensity rays. He positioned the low-intensity rays at multiple locations around the patient’s body. The low-intensity rays simultaneously converged on the tumor.

(Quoted from Gick & Holyoak, 1980: 311)

Figure 1.1 The prepositional structures of the Attack-Dispersion Story and the Radiation Problem

9612Arabski.indb 6

13/01/2006 08:42:42

On the Ambiguity of the Notion ‘Transfer’

7

In a large number of experiments they engage their subjects in various problem solution tasks through the presentation of story analogs dealing with the radiation problem and structurally similar but thematically distant stories, as the Attack-Dispersion Story. The design of the experiments had the subjects read one story analog, for instance the attackdispersion story. Then they were asked to solve the problem of the analogous story, for instance the radiation story. Although the resulting individual solutions showed a large variety, a general tendency was evident. Only some students succeeded in transferring the knowledge acquired from reading the first story to the solution of the second problem, most likely due to the ill-defined problem. Analogical transfer, in other words, if it is to contribute to the solution of complex problems as the radiation problem, rarely occurs without additional instruction or hints. Or, to put it differently, if the mapping of two problem domains, because of the diverging and distant content, presents an additional problem, however the task as such may be analogous, structurally speaking, the transfer of the solution suggested by one of the analogs causes remarkable difficulties. Another interesting problem is raised in the article ‘Schema induction and analogical transfer’ (1983) and ‘Mental models in problem solving’ (1984). If analogical transfer means the solution of an unknown problem in terms of (an) old one(s) such a transfer depends on what has been called ‘mapping’. The question then arises what is it that is ‘mapped’. Is it some particular element(s), one or the other proposition of the base analog, or is it the overall macrostructure of the base analog that is mapped onto the target analog, its ‘schema’ or its underlying ‘mental model’? Or to put it another way, is analogical transfer likely to occur more effectively or easily through the mapping of the macrostructural schemata or mental models underlying the problem analogs on a comparatively abstract level of modelling? And does perhaps competence in analogical transfer depend on the activation of such a schema or mental model? The research referred to above does not give a conclusive answer to this question. What is of particular interest in the context of this paper is the obvious preference in the problem-solving literature for the application of the medical radiation domain in relation to the military domain as expressed in the attack-dispersion story. Or could it perhaps be that historically, culturally and conceptually the medical mental space and the military mental space have something in common and are thus likely to be mapped onto each other and blended into each other?

9612Arabski.indb 7

13/01/2006 08:42:43

8

Cross-linguistic Influences in the Second Language Lexicon

A Metaphor that Avoided War In a provocative article with the title ‘Metaphor and war: The metaphor system used to justify the war in the gulf’ (1992) George Lakoff has analysed the metaphors and metaphoric systems which in American political discourse have been applied to justify the first Gulf War. In the light of the lines of argumentation and course of events that led to the first Gulf War and the present absence of any vision of a political solution to the problems resulting from the victory over Saddam Hussein after the second Gulf War, this article is as enlightening now as it ought to have been in the political discussion during the years following the first war and the propaganda leading to the second one. Such metaphors and the analogical mapping of metaphors may lead to wars, as Lakoff’s argumentation clearly demonstrates, so may wars be avoided by the proper choice and mapping of metaphors as well. An example of truly historical dimensions is the final solution to the Cuban Crisis in 1962 to be discussed later. After the discovery of the construction of launch sites for intermediate range nuclear missiles and of the assembling of jet bombers capable of carrying nuclear weapons in the island of Cuba by an American high-flying U-2 reconnaissance airplane on Sunday, 14 October, 1962, CIA specialists in the National Photographic Center in Washington mapped the photos to top Russian secret material provided by the agent Oleg Penkovsky. The result of this careful investigation practically meant an immediate menace of atomic destruction for the total area of the US, only excluding the Northwest. On 16 October at 8.45 President Kennedy was informed of this alarming state of affairs by McGeorge Bundy, National Security Agent of the United States. A secret meeting with top advisers, including Robert Kennedy, was called immediately. On the same day the Executive Committee of the National Security Council, later known under its acronym ExComm, was summoned. For the coming 13 days between 16 October and 29 October, 1962 its task was to discuss and prepare political and military solutions to the crisis. It consisted of about 15 standing members who met regularly on a daily basis, if necessary, twice. Among them was Theodore Sorensen, what has some significance for this paper. On certain occasions specially invited experts participated as well. The so-called Kennedy Tapes are the result of a secret recording of these meetings initiated by JFK. According to Dean Rusk during the first meeting in the morning of 16 October, there seemed to be two courses of action to be taken; a quick surprise strike against the installations or an invasion, or a combination of things (May & Zelikow, 1997: 54). General Taylor later added another

9612Arabski.indb 8

13/01/2006 08:42:44

On the Ambiguity of the Notion ‘Transfer’

9

alternative: ‘I would also mention among the military actions we should take, that once we have destroyed as many of these offensive weapons as possible (through an air strike) we should prevent any more coming in, which means a naval blockade’ (May & Zelikow, 1997: 58). These three military measures, frequently ranked in the same sequence, sometimes in a different line of execution, for the coming days become the continuous focus of the debate. Their goal must be the destruction of the missiles and airplanes already stationed in Cuba and the prevention of the strengthening of these weapons through those on their way to Cuba through a blockade. From the beginning the participants were aware of the potential effect of any of these measures on the Berlin question, which Khrushchev had mentioned during his meeting with Kennedy in Vienna. Whether such military actions should be preceded or accompanied by a declaration of war in the course of the proposals became a constant topic in the light of the Pearl Harbor Trauma. It was George W. Ball who in a passionate memo wrote: ‘ … we tried Japanese as war criminals because of the sneak attack on Pearl Harbor’ and argued in the meeting of October, 18 that a surprise strike, ‘far from establishing our moral strength … would, in fact, alienate a great part of the civilized world by behaving in a manner wholly contrary to our traditions, by pursuing a course of action that would cut directly athwart everything we have stood for during our national history, and condemn us as hypocrites in the opinion of the world’ (May & Zelikow, 1997: 121). Whether the President who during the course of argumentation more and more favoured the blockade solution has ever considered a ‘surgical strike’ – the military solution blended and concealed with a medical metaphor – to me remains an open question. To the best of my knowledge I have not been able to find evidence for this in any of his taped statements. I am aware, though, of Theodore Sorensen’s remark in his famous volume (Sorensen, 1965), according to an entry in the Supplement to the Oxford English Dictionary (1986 Vol. IV, 651). Under Section e. it says, surgical: Designating swift and precise military attack, esp. from the air. orig. U. S. ‘1965 T. C. Sorensen Kennedy XXIV. 684 The idea of … a so-called “surgical” strike … had appeal to almost everyone first considering the matter, including President Kennedy.’ This question as such only seems to indicate a minor problem. Looking at it, however, from a larger perspective, it is of great importance, in that Kennedy’s final decision for the ‘quarantine solution’ most likely has been the decisive motivation for the activation of the medical mental space and mapping it with a relatively constrained military one, which finally allowed Khrushchev to give in. After all, such an analogical transfer and blending of the military and medical spaces may have saved us from the catastrophe of an atomic war.

9612Arabski.indb 9

13/01/2006 08:42:44

10

Cross-linguistic Influences in the Second Language Lexicon

On Thursday, 18 October Kennedy met with the Soviet ambassador. This meeting lasted until 7.15. Gromykow reminded Kennedy of what he knew anyhow from his discussion with Krushchev in Vienna, i.e. that he planned to visit the United States after the election in November and that the ‘rotten tooth’ of the Western military presence in Berlin had to ’be pulled out’ (May & Zelikow, 1997: 168). Although Kennedy gave Gromykow the opportunity to give him some information or at least a hint about the Russian activities in Cuba he denied that any missiles had been stationed there or were on their way to the island. In Robert Lovett’s report of the afternoon events of that day he stated that a tight blockade should precede air strikes and a possible invasion. ‘The blockade would allow a demonstration of national will to persuade the Russians to withdraw their missiles without great bloodshed, without appearing “trigger-happy”’ (May & Zelikow, 1997: 170). For the first time in the ExComm debates it is Lovett in an answer to President Kennedy who in this context introduces the term ‘quarantine’. No reference is made why this is done, nor does Lovett refer to Roosevelt’s famous earlier use of this term. In the Oxford English Dictionary, Second Edition, Vol. II, (1989: 299) Section 2, quarantine is: A period (orig. of forty days) during which persons who might serve to spread a contagious disease are kept isolated from the rest of the community; esp. A period of detention imposed on travellers or voyagers before they are allowed to enter a country or town and mix with the inhabitants; commonly, a period during which a ship, capable of carrying contagion is kept isolated on its arrival at a port. (…) b. fig. Any period, instance, etc. of detention or seclusion, compared to the above; esp. in international politics, a blockade, boycott, or severance of diplomatic relations intended to isolate a nation, or the isolation caused by such action. The specific use arose from a speech by F. D. Roosevelt, President of the U.S. According to this speech of F. D. Roosevelt of October 1937, three months after Japan’s attack on China, Without a declaration of war and without warning or justification of any kind, civilians, including women and children, are being ruthlessly murdered with bombs from the air. […] Innocent peoples, innocent nations, are being cruelly sacrificed to a greed for power and supremacy which is devoid of all sense of justice and humane considerations. […] It seems to be unfortunately true that the epidemic of world lawlessness is spreading. When an epidemic of physical disease starts to spread, the community approves and joins

9612Arabski.indb 10

13/01/2006 08:42:45

On the Ambiguity of the Notion ‘Transfer’

11

in a quarantine of the patients in order to protect the health of the community against the spread of the disease. […] War is a contagion, whether it be declared or undeclared. It can engulf states and peoples remote from the original scene of hostilities. We are determined to keep out war, yet we cannot insure ourselves against the disastrous effects of war and the dangers of involvement. We are adopting such measures as will minimize our risk of involvement, but we cannot have complete protection in a world of disorder in which confidence and security have broken down. (Hofstadter, 1958: 391) The quarantine metaphor in Roosevelt’s speech to which President Kennedy and the participants of the ExComm recur stands for the protection of the health of the people against the dangers caused by an epidemic. War is clearly denounced as a dangerous contagion. A quarantine may re-establish law and justice and protect from the effects of war. The medical source domain taken from Roosevelt’s speech is mapped onto the present crisis in world affairs with the goal to prevent war as the target domain. It is the analogical blending of the two mental spaces that lays the ground for the avoidance of war. The ExComm’s preceding discussion of the alternatives: a surgical strike vs. a quarantine leading to the final solution of the Cuban crisis may shed some light on the ambiguity of the notion of transfer, its potentially disastrous effects, and, at the same time, the role analogical transfer may play in the mapping of mental spaces in the solution of problems. References Fauconnier, G. (1997) Mappings in Thought and Language. Cambridge, UK: Cambridge University Press. Gick, M.L. and Holyoak, K.J. (1980) Analogical problem solving. Cognitive Psychology 12, 306–55. Gick, M.L. and Holyoak, K.J. (1983) Schema induction and analogical transfer. Cognitive Psychology 15, 1–38. Hofstadter, M. (ed.) (1958) Great Issues in American history: A Documentary Record. Vol. 2, (1864–1957). New York: Vintage Books. Holyoak, P.W. (1984) Mental models in problem solving. In J.R. Anderson and S.M. Kosslyn (eds) Tutorials in Learning and Memory: Essays in Honor of Gordon Bower (pp. 193–218). San Francisco: Freeman. Johnson-Laird, P.N. and Wason, P.C. (eds) (1977) Thinking: Readings in Cognitive Science. Cambridge, UK: Cambridge University Press. Lakoff, G. (1992) Metaphor and war: The metaphor system used to justify war in the gulf. In M. Pütz (ed.) Thirty Years of Linguistic Evolution: Studies in Honour of René Dirven on the Occasion of his Sixtieth Birthday (pp. 464–81). Philadelphia: John Benjamins. May, E.R. and Zelikow, P.D. (eds) (1997) The Kennedy Tapes: Inside the White House During the Cuban Missile crisis. Cambridge, MA: The Belknap Press. Sorensen, T.C. (1965) Kennedy. New York: Harper & Row.

9612Arabski.indb 11

13/01/2006 08:42:46

Chapter 2

Language Transfer in Language Learning and Language Contact JANUSZ ARABSKI Language transfer as a linguistic concept has always been considered and seen as a phenomenon which occurs in language learning situations. The two definitions below represent the most common views. Behaviorist psychologists, who first defined ‘transfer’ technically, used it to refer to a process described as the automatic, uncontrolled, and subconscious use of past learner behaviours in the attempt to produce new responses. In this sense, transfer may be of two types: ‘negative’ and ‘positive’. ‘Negative transfer’ refers to those instances of transfer which result in error because old, habitual behaviour is different from the new behaviour that is being learned. For example, if one has regularly driven a car where the gear shift is on the floor, one will invariably reach for the floor when first attempting to drive a new car whose gear shift is on the steering column. In contrast, ‘positive transfer’ results in correct performance because the new behaviour is the same as the old. In our gear shift example above, positive transfer would operate if the new car also had its gear shift on the floor – the old and new gear shifting would be the same. Both types of transfer refer to the automatic and subconscious use of old behaviour in new learning situations. (Dulay et al., 1982: 100–1) transfer A term used in applied linguistcs to refer to a process in foreign language learning whereby learners carry over what they already know about their first language to their performance in their new language. This tendency may be an advantage, if the two languages have features in correspondence, as there will be ‘positive transfer’ (or ‘facilitation’). Rather more noticeable, however, are the cases of ‘negative transfer’ (or interference), where the patterns of the two languages do not coincide. (See Corder, 1973: Ch. 6; Crystal , 1980: 62) 12

9612Arabski.indb 12

13/01/2006 08:42:47

Language Transfer in Language Learning and Language Contact

13

Language transfer, however, takes place also in other language situations. In this discussion we would like to present two other language situations where language transfer is present. They are language contact (LC) and dialect contact (DC) phenomena. These two are going to be compared with language learning situations (LL) in the discussion below. The phenomenon of language contact concerns two languages and the influence of one on the other. In the discussion below we are then going to discuss three situations in which different language systems are put into contact with each other. (1) Foreign language learning (LL). (2) Language contact (LC). (3) Dialects in contact (DC). These three situations are going to be discussed from the point of view of the processes which occur in them but especially from the point of view of language transfer: (1) How language transfer operates (functions) in these three situations and with what intensity. (2) In what way these types of contact differ from each other. The following are some basic characteristics of language transfer in language learning. In a foreign language learning situation language transfer depends on the kind of language contact. Its intensity and type depends on which L1 is in contact with which L2, and how distant genetically are L1 and L2. Language transfer appears with greater intensity when the two systems are genetically closer and thus when there are more points of reference for the transfer to occur. There is more language transfer between Polish and Russian, both positive and negative, than between Polish and English in the foreign language learning process. Poles learning Russian rely more on Polish to help them acquire Russian but they also make a lot of errors in structures which are different in Russian. The genetic proximity of Russian makes Poles transfer Polish endings and add them to Russian roots. The transfer of Polish morphological endings hardly ever happens in Polish-English contact, in Polish-English interlanguage (IL). Genetic proximity as a criterion of transfer occurrence has been described by Ringbom (1976) and others. The intensity of language transfer also depends on the stage of interlanguage development. In the case of Polish-English interlanguage, negative transfer does not occur at the very beginning of the English learning process, at the stage of imitation. It then becomes more and more frequent

9612Arabski.indb 13

13/01/2006 08:42:47

14

Cross-linguistic Influences in the Second Language Lexicon

until at the advanced level it starts to decrease. It decreases when L2 structures are well established and have become resistant to L1 influence. Some structures are especially susceptible to transfer. In the case of Polish-English interlanguage (IL) these are e.g. word order, tense system and lexis. Pronounciation L1 habits are most obvious since they are universally present in every spoken IL. The age of learners is another factor determining transfer occurrence and frequency. Transfer occurs less frequently among children whose L1 system is not yet strong enough to influence the new L2 structures. Young learners, especially children before puberty, naturally acquire a second language without much influence from L1. Structures which are deeply rooted in the system of L1 are more likely to be transferred into interlanguage than those whose philogenetic status in the system of L1 is declining (cf Krashen, 1983). The Polish Pluperfect tense (przybyÙem byÙ) is already an archaic form and Polish learners of English do not transfer it into Polish English IL by applying positive transfer. They make errors as if the Pluperfect did not exist in Polish (L1). Structures that are marked in L1 are less likely to be transferred into IL. For example, idiomatic expressions are not transferred from L1 as often as unmarked universal structures (cf Kellerman, 1983). Likewise, in a situation when a learner concentrates on a grammatical rule and applies a monitor he/ she does not transfer L1 habits as often as in a conversation situation when he/she is concentrating on semantic content and not how to apply an L2 rule. The most important and universal feature of negative transfer is its simplifying character. It is used in order to simplify target language structures under the influence of L1 and these are very unlikely to be complicated by L1. The English Vowel System, for example, is simplified under the influence of the Polish system. The English system consists of nine vowels and the Polish one of five vowels (see Figure 2.1).

i: i

i ε e

æ a

a

Figure 2.1 English and Polish vowel systems

9612Arabski.indb 14

13/01/2006 08:42:48

Language Transfer in Language Learning and Language Contact

15

Two English sounds /Ή/ and /æ/ are usually simplified under the influence of the Polish system to /e/ and the English Ή/æ contrast disappears in Polish English IL. Bet and bat sound the same to a Polish learner of English and they are produced by Poles in the same manner. The same process of underdifferentiation applies to many grammatical and lexical structures and, for example to prepositions. Polish w has many English counterparts. Among them are in, on, and at: in March on Sunday at home

– – –

w Marcu w niedziel¿ w domu.

As a result, one of them is selected to simplify the English system – in is used instead of many other counterparts resulting in interlanguage forms: *in Sunday *in home. One of the earliest descriptions of language transfer is by Lado (1957: 2) in his classic Linguistics Across Cultures: Individuals tend to transfer the forms and meanings, and the distribution of forms and meanings of their native language and culture to the foreign language and culture – both productively when attempting to speak the language and to act in the culture, and receptively when attempting to grasp and understand the language and the culture as practiced by natives. (Lado, 1957: 2) It is the culture which is transferred, language structures being part of it. The transference of culture occurs typically in language contact situations. This can be illustrated by present Polish-English language contact. This contact is the result of globalisation and of English having become the language of international communication. Contact between the two languages has over 100 years of tradition but recently, after political changes in Poland, has been radically intensified. Currently Polish borrows from English several new lexical items every week. There is some evidence for the borrowing of pragmatic and syntactic structures as well. Within the last 10 years over 1000 English words have been borrowed by the lexical system of Polish. At the beginning of the 20th century there were around 250 English loans in Polish. By 1995 the estimated number of anglicisms had grown to 1700 (cf. MaÚczakWohlfeld, 1994: 8). Lexical items are borrowed for at least three reasons. The new items are needed because of:

9612Arabski.indb 15

13/01/2006 08:42:51

16

Cross-linguistic Influences in the Second Language Lexicon • •

new political and economic situations; technological developments; new styles of life.

•

The character of the loans reflects the character of Polish cultural contact with Anglo-Saxon culture in the past. Years ago it was sport (e.g. boks, derby, tenis). More recently it has become contact with English as the language of international technology and communication (offset, tranzystor, lunch, T-shirt, workoholic). Cultural contact and the new borrowed concepts need new terms, a new lexis to convey new meanings. Lately, however, there have been more and more English loans which are used in Polish, not to reduce the lexical deficit of vocabulary but to identify the speaker with English or American culture; e.g. solarium inside, bilety do wszystkich destynacji kupisz u nas (tickets for every destination available here), sale, reservation, exit, Nokia connecting people, be inspired, and expressions used by young people: hello, wow, sorry, O.K, cool. The influence of English does not go much beyond lexis. However, there is some evidence that Polish word order has recently changed under the influence of English. According to Rusiecki (2000) the new Polish term wirtualna rzeczywistoï° (virtual reality) instead of traditional Polish rzeczywistoï° wirtualna (N+Adj) may be evidence of the English influence on Polish word order and thus syntax. This influence, however, does not change Polish rules because Polish word order rules here are not fixed and in several instances Adj+N is possible. wirtualna rzeczywistoï° (virtual reality) Čeglarskie obozy (sailing camps) komediowy serial (comedy serial) are used instead of rzeczywistoï° wirtualna obozy Čeglarskie serial komediowy. The rule to use adjectives in the attributive instead of postpositive position in Polish is not obligatory since in the following examples both attributive and postpositive use is possible: wiklinowy kosz – kosz wiklinowy (wicker basket) powszechna opinia – opinia powszechna (public opinion) letniskowa miejscowoï° – miejscowoï° letniskowa (summer resort) miejscowa elita – elita miejscowa (local gentry) d¿bowy las – las d¿bowy (oak wood).

9612Arabski.indb 16

13/01/2006 08:42:52

Language Transfer in Language Learning and Language Contact

17

Only the attributive position, however, is permissible with certain structures: wiek ïredni (middle age) ptaki ïpiewaj­ce (songbirds) pan mÙody (groom) Changing the word order of the above phrases changes their meanings: ïredni wiek (average age) ïpiewaj­ce ptaki (singing birds) mÙody pan (young gentleman). Since the rule is not clear-cut and both word orders are possible (with some restrictions), the construction is susceptible to transfer from English. The results of my study on compliment responses in Polish among young Poles conducted recently (Arabski, 2004) show another aspect of language transfer. Young Poles use much more thanking in compliment responses than the older generation of Poles, probably under the influence of American English. Compliments are accepted, e.g. • •

You have a nice sweater. Yes, isn’t it.

where traditionally in Polish it was more common to reject a compliment. The situation is completely different in the case of dialect contact. Here the two systems are close typologically and often geographically. These days, however, because of the media, physical proximity is difficult to describe and is perhaps less relevant. The contact can take place with the help of TV, press and the radio. Dialects are ‘varieties of language that are mutually intelligible at least to some degree’ (Trudgill, 1986: 1). The main processes which are characteristic of dialects in contact are ‘accent convergence’ and ‘accent divergence’ (Trudgill, 1986: 2). Accent convergence takes place when a speaker wants to gain the approval of his message receiver and adopts his accent patterns to the receiver’s patterns. When he speaks he uses the words or accent of his interlocutor, e.g. a British speaker may use sidewalk instead of pavement when talking to an American. Accent divergence is the opposite. The speaker wants to disassociate himself from or show disapproval of others. Accent and behavioural convergence are studied by social psychologists of language who have developed accommodation theory. ‘This theory focuses on speech, and discusses and attempts to explain why speakers modify their language in the presence of others in the way and to the extent that they do. It also examines the effects and costs of this type of

9612Arabski.indb 17

13/01/2006 08:42:53

18

Cross-linguistic Influences in the Second Language Lexicon

modification’ (Trudgill, 1986: 2). The decision of which element of a different dialect should undergo convergence and be accommodated depends on its degree of salience. Salient elements are selected to be modified to sound like the dialect which one wants to imitate. An example of this kind of selection is the set of features of American English used by British pop singers to sound American (Trudgill, 1986: 13). (1) Words such as life, my tend to be sung with a monophthongal vowel of the type [a·], although in spoken English English they are most usually pronounced with a diphthong of the type [ai~ i~̎i], etc. (2) Words such as girl, more tend to be pronounced with an /r/ even by those English English speakers (the majority) who do not have nonprevocalic /r/ in their speech. (3) Words such as body, top may be pronounced with unrounded [ ] instead of the more usual British [d]. (4) It is not usual to pronounce words such as dance, last with the /a:/ that is normal in the speech of south-eastern England. Instead they are pronounced with the /æ/ of cat (as in the north of England, although the pronunciation is usually [æ] rather than northern [a]). In addition, words such as half and can’t, which are pronounced with /a:/ by most northern English speakers, must also be pronounced with /æ/. Thus: d

d

cat

Dance

half

south-eastern England

/æ/ = [æ]

/a:/

/a:/

northern England

/æ/ = [a]

/æ/

/a:/

pop-song style

/æ/ = [æ]

/æ/

/æ/

(5) The pronunciation of intervocalic /t/ in words like better as [t] or [ ], which are the pronunciations most often used by most British speakers, is generally not used. In pop singing, a pronunciation of the type [ ~d] – a voiced alveolar flap – has to be employed. Some lexical borrowings from American English are the following from the 1930s: bakery for baker’s shop grocery for grocer’s shop raincoat for mackintosh sweater for pullover toilet for lavatory.

9612Arabski.indb 18

13/01/2006 08:42:54

Language Transfer in Language Learning and Language Contact

19

The more recent borrowing is radio for wireless (the latter used until c. 1960). The accommodation process in dialect contact is the acquisition of salient features of the dialect which one accommodates to. In the above examples the use of selected American lexical items and selected features of American pronunciation is intended to make the speaker sound like an American. In LC, as in the case of typologically distant Polish-English contact discussed above, the lexical contact is most obvious. Lexis in general floats from language to language without too many obstacles. Other structures like pragmatics (A), syntax (B), morphology (C) and phonology (D) are (in this order) more resistant to foreign influence. Phonology being the most resistant (see Figure 2.2). In Polish-English language contact almost nothing but lexis can be borrowed. Distant from Polish both culturally and also in its typology, English does not allow anything other than lexical transfer, lexical borrowings or lexical influence. In English-American DC the influence gets as far as phonology and includes lexis. The reasons to acquire foreign structures, to borrow foreign words or to accommodate other dialect forms, are the deficit of L2 structures which makes us unable to: • • •

communicate in a foreign language (LL) name new products, institutions, etc. (LC) sound like another dialect speaker (DC)

D

C

B

A

Figure 2.2 Resistance to foreign influence in LC

9612Arabski.indb 19

13/01/2006 08:42:55

20

Cross-linguistic Influences in the Second Language Lexicon

In order to satisfy these needs and eliminate the deficit of structures we acquire, borrow and accommodate. These three situations are the situations of acquiring new language structures for different purposes and for three different motives. The only real language contact situation is the LL situation, in which two systems are put in contact with each other when theoretically the entire L1 system is confronted with the entire L2 target system. In this respect the LC situation differs because it represents not the contact of language systems but the contact of cultures. Over 94% of the borrowed English lexical items are nouns (MaÚczak-Wohlfeld, 1994: 151). So it is a contact of concepts, ideas, gadgets and institutions. Identification with L2 culture is another factor to be considered in the three situations. DC serves to provide identification with different dialect speakers. Linguistically and culturally the differences here are not as great as in the case of LL and LC. In order to identify with the speakers of a different dialect one selects a form which is salient and thus marked. The counterpart of salient forms in DC are the ones which are needed for communication in LL. Identification is also present in LC when one uses a foreign lexical item only to demonstrate his/her identifying with a foreign culture (using newly coined lexical loans for which common counterparts exist in Polish like reservation, destynacja, wow). In LL identification is also present since integrative motivation makes learners acquire L2 in order to identify with a target culture. Integrative motivation occurs in many LL situations. Motivation is one of the key factors responsible for successful language learning. L2 structures which are filtered in by a language learner are acquired. The counterpart of selected structures for acquisition in L2 are those which are salient in DC situations. Not every L2 structure is acquired in a LL situation. There is an order of acquisition which determines which structures are acquired before and after other structures in spite of the fact that the entire systems are in contact. Not every structure is accommodated in a DC situation. The selection is common for both processes and situations. The selection is also present in LC. Some borrowings are necessary because they are names of products, institutions or gadgets which do not exist in L1. Others are acquired to associate the speaker with L2 culture, as was mentioned above. In this respect borrowings are like DC lexical elements. They have the status of mentioned above bakery, grocery and sidewalk. Identification, motivation, communication even selection are the domain of social psychologists of language. Social psychology studies the reasons for borrowings and accommodations. It studies who accommodates to whom and why. These four phenomena are present in the three types of language contact we have presented above.

9612Arabski.indb 20

13/01/2006 08:42:57

Language Transfer in Language Learning and Language Contact

21

Language transfer understood as language change is a different matter. It is the domain of linguists, like e.g. historical change. Linguistically speaking, language transfer functions in the same way in the two abovementioned types of language contact. When the same two structures in the same kind of language contact are juxtaposed in a situation of either language learning or language contact, they result in the same interlanguage structure. English /Ή/ and /æ/ will be simplified by Polish learners to Polish /e/. The same phonetic form will be found in English borrowings. Dialect contact is a different phenomenon in this respect. British English speakers adopt native American forms of pronunciation because of the typological proximity of the two systems. It is a case of accommodation. The presence of transfer from the source dialect is not as obvious but it does exist. The source dialect influence differs from individual speaker to individual speaker but it is there. Not every speaker fully accommodates the target form and distorts it to some degree. The distortion is the result of negative transfer. LL, LC and DC have many features in common. These features appear with different intensity in the three types of language contact. Their priorities are also different for each type of language contact. References Arabski, J. (ed.) (2004) Pragmatics and Language Learning. Kraków: Universitas. Corder, S.P. (1973) Introducing Applied Linguistics. London: Penguin. Crystal, D. (1980) A First Dictionary of Linguistics and Phonetics. Boulder, CO: Westview Press. Dulay, H., Burt, M. and Krashen, S. (1982) Language Two. New York: Oxford University Press. Kellerman, E. (1983) Now you see it, now you don’t. In S. Gass and L. Selinker (eds) Language Transfer in Language Learning. Rowley, MA: Newbury House Publishers. Krashen, S.D. (1983) Newmark’s ‘Ignorance Hypothesis’ and current second language acquisition theory. In S. Gass and L. Selinker (eds) Language Transfer in Language Learning. Rowley, MA: Newbury House Publishers. Lado, R. (1957) Linguistics Across Cultures. Ann Arbor: University of Michigan Press. MaÚczak-Wohlfeld, E. (1994) Angielskie Elementy Leksykalne w J¿zyku Polskim. Kraków: Universitas. Ringbom, H. (1976) What differences are there between Finns and Swedishspeaking Finns learning English. In H. Ringbom and R. Palmberg (eds) Errors Made by Finns and English Speaking Finns in Learning of English. AFTIL. Rusiecki, J. (2000) Ja mówi° po anglo-polski – czyli o polszczyĊnie przeÙomu tysi­cleci. J¿zyki Obce w Szkole 5 (5). Trudgill, P. (1986) Dialects in Contact. Oxford: Basil Blackwell.

9612Arabski.indb 21

13/01/2006 08:42:57

Chapter 3

Could a Contrastive Analysis Ever be Complete? TERENCE ODLIN For many linguists the research problems connected with transfer can be summed up in the phrase the Contrastive Analysis Hypothesis. Other problems in second language acquisition (SLA) have likewise been conferred with the moniker ‘hypothesis’, for example, the Critical Age Hypothesis. Even so, we will do well to remember that the way we talk about any phenomenon can affect what we expect to find. The label ‘hypothesis’ has the advantage of reminding us that we should view problems such as transfer and age of acquisition as research areas accountable to actual data. However, the term ‘hypothesis’ can also be misleading. Birdsong (1999) has noted, for instance, that there is actually more than one possible hypothesis connected with the notion of age-related differences in SLA. By similar logic in the case of transfer research, the so-called ‘Contrastive Analysis Hypothesis’ should not imply a monolithic hypothesis – or a litmus test that could show the hypothesis to be true or false. An influential article by Ronald Wardhaugh (1970) tacitly acknowledged that the Contrastive Analysis Hypothesis is less than monolithic. Wardhaugh distinguished between a priori and a posteriori predictions, which he considered to entail two very different hypotheses, the former termed ‘strong’ and the latter ‘weak’. The strong version consists of predictions based simply on similarities and differences between two languages. The weak version consists of predictions based on actual learner performances and correlated with cross-linguistic differences. For Wardhaugh, only a weak version of the Contrastive Analysis Hypothesis could succeed. Later in this paper I will show why a dichotomy between strong and weak predictions risks oversimplifying the issue.

Difficulties of Contrastive Predictions Although transfer researchers have good reason to question the distinction Wardhaugh made, the difficulties of contrastive analysis should not 22

9612Arabski.indb 22

13/01/2006 08:42:58

Could a Contrastive Analysis Ever be Complete?

23

be underestimated. For one thing, a great deal of SLA research indicates that transfer plays little if any role in some cases. Moreover, even when transfer does play a role, it can take different forms, some easier to detect than others. In an important challenge to Wardhaugh, Jacquelyn Schachter (1974) argued that learners’ avoidance patterns sometimes arise from differences between the native and target language: thus, speakers of Chinese and Japanese may have an especially hard time with the relative clauses of English because relativisation in these languages is so different. Although there have been empirical challenges to Schachter’s analysis (e.g. Kamimoto et al., 1992), there is other evidence both for avoidance and for its interaction with cross-linguistic differences (Odlin, 1989, 2003). In effect, then, avoidance can sometimes be considered a covert kind of language transfer. A somewhat similar case of covert transfer is evident in a study by Jarvis and Odlin (2000) of prepositions – and the absence of obligatory prepositions – in the acquisition of English in Finland. Finland has two official languages, Finnish and Swedish. The latter is, like English, a prepositional language, whereas Finnish uses a mixed set of structures for the same purposes. The primary type of structure is the nominal case inflection: Finnish nouns can inflect for 15 different cases, often translatable into Swedish and English as prepositions. Along with noun inflections, there are postpositions and even a handful of prepositions in Finnish. Both Finnish and Swedish speakers often choose an English preposition that reflects the influence of their native language. In addition, however, Finnish but not Swedish speakers often omit prepositions which are obligatory in the target language, as in C[harlie] C[haplin] and the woman go to sit the grass. Because cases such as the absence of on before the noun phrase the grass occurred only among the Finnish learners and never among Swedish learners with comparable backgrounds, zero-prepositions should be viewed as an interaction of two processes often thought to be mutually exclusive: simplification and transfer. We have thus seen that covert behaviours involving the absence rather than the presence of something can be implicated in cross-linguistic influence, and so contrastive analysts face the very formidable problem of predicting when they will find a covert as well as an overt behaviour. Two other problems in making predictions about transfer will also be noted here, but only in passing. First, individual variation matters in the study of transfer, as seen in the fact that different speakers of Finnish showed different patterns of transfer, as did the Swedish speakers (Jarvis & Odlin, 2000). Second, a highly significant concomitant of individual variation is the fact that all cross-linguistic influence arises through the subjective

9612Arabski.indb 23

13/01/2006 08:42:59

24

Cross-linguistic Influences in the Second Language Lexicon

assessments of individual learners. Indeed, many in the transfer research community have taken subjective assessment into account as seen in the work of Kellerman (1977, 1995) and others (cf. Odlin, 1991, 2003; Ringbom, 1987; Abdullah & Jackson, 1998). Later in the paper I will consider the problem of subjective assessment more.

Constraints as Predictions Turning from the difficulties of contrastive predictions, we should consider the notion of constraints on transfer. Critics of a priori contrastive analysis have often sought to posit constraints on language transfer yet have apparently ignored the parallel nature of predictions and constraints. If it is hard to predict when language transfer will occur, it is no easier to predict when it will not occur. In a recent lecture, I argued that a constraint is not very different from a contrastive prediction but is, rather, its ‘evil twin’ (Odlin, 2002a). At least four constraints on syntactic and/or lexical transfer have been proposed, yet all four predictions have turned out to be empirically unsound. From at least as early as the 19th century some linguists have viewed bound morphology as a structural subsystem that is immune to cross-linguistic influence. However, a wealth of evidence contradicts any blanket claim about the non-transferability of bound morphology (Weinreich, 1953/1968; Dusková, 1984; Orr, 1987; Thomason & Kaufmann, 1988; Jarvis & Odlin, 2000). Another constraint sometimes proposed would have it that basic word order is never transferable (e.g. the use of an SOV sentence pattern in an SVO language). Once again, however, there is considerable evidence suggesting that basic word order is not immune to transfer (Odlin, 1989, 1990). So far I have reviewed two of four so-called constraints on transfer. The third prediction involves idioms. An early claim of Kellerman (1977) about Dutch idioms was that they are not transferable, although he has subsequently scaled back his constraint. In any case, language contact research indicates that second language learners sometimes take idioms from their native language and create translation equivalents in the target language (Odlin, 1991). The most recent constraint on transfer that I have seen proposed involves the construct known in Universal Grammar research as the ‘functional projection’. This construct includes a wide variety of grammatical morphemes including articles, and Vainikka and Young-Scholten (1998) have argued that functional projections are not transferable. However, a great deal of evidence indicates that this supposed constraint on transfer will not hold for articles (Odlin, 2003).

9612Arabski.indb 24

13/01/2006 08:42:59

Could a Contrastive Analysis Ever be Complete?

25

To point out the inadequacy of the constraints just discussed does not, however, mean that there are no constraints on transfer. True, we would do well to avoid sweeping claims such as the one holding that bound morphology is not transferable. However, some kinds of morphological transfer rarely if ever occur. For example, it is probably safe to predict that learners of a synthetic language will not often try to use nominal case inflections or person/number markers when trying to speak or write a target language that is analytic (and this prediction is consonant with conditions suggested by Weinreich (1953/1968)). For instance, I would be surprised if many native speakers of Polish used the first-person plural suffix –my on English verbs to produce interlanguage forms such as *readmy for we read or gomy for we go. Yet though these errors seem very improbable, it may well be true that Polish learners of an East or South Slavic language would be tempted to use some bound morphology from their native language. The discussion by Dusková (1984) of Czech learners of Russian suggests that whatever constraints there may be on bound morphology, they depend very much on both of the languages in a contact situation, and on learners’ perceptions of the cross-linguistic distances.

Predictions and Transferability Just as some predictions about when transfer will not occur seem reasonable to make, so do some predictions about when transfer will occur. At the beginning of this paper, I noted the distinction made by Wardhaugh between a priori and a posteriori contrastive analysis. Although it might seem that these categories exhaust the types of possible contrastive predictions, there is in fact a kind of prediction that combines elements of the a priori and a posteriori approaches. For example, there now exist several studies comparing the use of articles in English and other languages by learners whose native language has articles and by learners whose native language does not have articles. Despite the differing native and target languages, the results have been quite consistent in pointing to the advantage learners generally enjoy when their native language has articles (Odlin, 1989, 2003). Accordingly, we could confidently predict that, for example, speakers of French, a language with articles, would have an easier time than would speakers of Russian with the articles of Swedish. Even though the empirical basis for the prediction comes from a posteriori studies, it does not seem risky to make an a priori prediction in such a case. Elsewhere I have made similar suggestions about other constructions including relative clauses (Odlin, 2001) and absolute constructions (Odlin, 1992). Such predictions are among the desiderata of a more general theory

9612Arabski.indb 25

13/01/2006 08:43:00

26

Cross-linguistic Influences in the Second Language Lexicon

of transferability, a concept long discussed in the literature but still not achieved (cf. Kellerman, 1995; Selinker, 1992; White, 2000; Odlin, 1992, 2003). Even so, the remarkable consistency of findings such as those on articles gives real hope for ever-better contrastive predictions.

Completeness in Contrastive Analysis Affirming the robustness of some contrastive predictions does not, however, address the question that is the title of this paper: Could a contrastive analysis ever be complete? In asking the question, I do not wish to give the impression that existing contrastive analyses lack value. On the contrary, I have found quite helpful many cross-linguistic comparisons (e.g. Stockwell et al., 1965; Fisiak et al., 1978). I am convinced, moreover, that teachers who can carefully compare the native and target languages are especially qualified to help their students, as seen, for example, in a recent study by Han (2001). Yet however confident we may be about the theoretical and practical value of contrastive research, much remains to be done. Some of the necessary work will involve a search for evidence to support or refute the following two claims: •

No contrastive analysis can be comprehensive without a viable theory of transferability. No theory of transferability can be entirely satisfactory without a viable account of relevant affective factors.

•

There are other challenges for contrastive analysis besides: for instance, the sheer volume of detail necessary, as seen in the fact that even a massive descriptive grammar of English such as Quirk et al. (1985) is incomplete. Yet another area in need of exploration is the role of affect in language transfer. In the remaining sections of this paper I will first give an example of how affect and transfer can intersect even in the domain of syntax. I will then discuss wider ramifications of affective transfer including the problem of linguistic relativity.

Transfer of a Focus Construction The target language in the example is English, and the learner’s native language is Swedish. It comes from student writing in a database developed by my friend and colleague Scott Jarvis (1998). The writing samples consisted of descriptions of the Charlie Chaplin film Modern Times including the following sentence: He [Chaplin] say it were he som take the bred.

9612Arabski.indb 26

13/01/2006 08:43:01

Could a Contrastive Analysis Ever be Complete?

27

The word som may seem strange in this sentence until one realises that in Swedish it can function as a relative pronoun, which would correspond here to the English pronoun who. Jarvis also had students write in their native languages, and the database has very similar sentences in Swedish such as the following: Charlie säger då att det var han som tog brödet = Charlie says then that it was he who took bread-the The transfer in the English sentence is not just the pronoun som but rather an entire construction most commonly called a cleft sentence: it were he som take the bred. (Other characteristics of this sentence also suggest transfer, e.g. the choice of were instead of was, but the cleft is the point of interest for this paper.) Linguists usually describe cleft sentences with terms such as ‘focusing constructions’ and ‘emphatic devices’. Yet as Celce-Murcia et al. (1999: 623) note, notions such as focus and emphasis are hard to define ‘because they overlap with other topics such as exclamation and intensification’. The functions of exclamation and intensification clearly involve affect, but it is an open question whether or not we should consider all cleft sentences and other focusing devices to be equally relevant to the study of language and emotion. At the very least, however, such structures are attention-getting devices: they indicate what a speaker or writer ‘feels’ the audience should note and, presumably, remember for at least a little while. However much the emotional content may vary in different focus constructions and other emphatic devices, all of them probably involve what psychologists have long referred to as arousal (e.g. Ortony et al., 1988). In any case, linguistic work on right and left dislocation as well as on cleft constructions has occasionally noted the affective dimensions of such constructions even where the languages studied are as different as Wolof, English, Czech and Chinese (cf. Irvine, 1990; Firbas, 1992; Guo, 1999).

Transferability of Focus Constructions There was both positive and negative transfer from Swedish in the cleft sentence it were he som take the bred. The pronoun som is clearly a case of a mistaken interlingual identification. On the other hand, the use of a cleft pattern similar to one in Swedish shows positive transfer. This result invites an inference that when both the native and target language have cleft patterns, this type of focus construction will be highly transferable. Indeed, there are several other examples in the database of transfer of cleft patterns. Moreover, other language contact situations also indicate the

9612Arabski.indb 27

13/01/2006 08:43:02

28

Cross-linguistic Influences in the Second Language Lexicon

transferability of cleft and related focus constructions. Work by Trèvise (1986) on the English of French learners supports this inference, as does work on the historical development of cleft patterns in the distinctive forms of English spoken in some parts of Ireland and Scotland (Filppula, 1986, 1999; Odlin, 1997). Earlier in this paper, I discussed a kind of prediction we can make about the advantage generally enjoyed when the native language has articles. A similar kind of prediction seems viable for cleft sentences: we can expect positive transfer of cleft patterns in other language contact situations (even while some negative transfer may be a concomitant). A more difficult prediction to make is whether focus constructions are always transferable. If they are, then notions such as focus and emphasis play an especially important role in this and perhaps other kinds of syntactic transfer. In fact, evidence from pidgins and creoles (e.g. Faraclas, 1988; Sankoff, 1993) suggests that the cross-linguistic similarities need not be as great as they are between focus constructions in, say, French and English. It remains an open question, however, as to whether or not such cases can occur in many other language contact situations.

Formal versus Functional Criteria in Transfer The formal similarity of the cleft patterns in Swedish and English no doubt strengthens the conviction of learners that they are using a transferable construction. However, the factors promoting transfer are also communicative. The meaning of the English sentence was quite similar to the Swedish cleft, with both focusing on who it was that stole the bread. The issue of focus constructions illustrates a more general problem for any viable theory of transferability: the relative importance of formal versus functional similarity. Formal similarity can certainly promote acquisition, as seen in a recent dissertation by Nakahama (2003). In this study the target language was Japanese and the native languages were Korean and English. The target language is well known for two particles, ga and wa, which can mark the categories of subject and discourse topic. Korean has particles quite like ga and wa (ka and ul respectively), and Nakahama’s results show that the Koreans found it relatively easy to use these particles even though some students occasionally diverged from the norms of native speakers of Japanese. In contrast to the Koreans, the native language of the English speakers offered no formal similarities that would aid in acquisition even though the functional categories of topic and subject also exist in English. Yet despite cases such as the relative success of the Koreans in Nakahama’s study, formal similarity alone does not drive the acquisition process.

9612Arabski.indb 28

13/01/2006 08:43:03

Could a Contrastive Analysis Ever be Complete?

29

If it did, we would not find learners making interlingual identifications when the formal differences are large. In fact, however, the study by Jarvis and Odlin (2000) already mentioned indicates that speakers of Finnish use the semantic information in nominal cases to choose an English preposition when writing spatial descriptions. For example, some native speakers of Finnish used the predicate sit to the grass, as part of a description of one scene in the film Modern Times, with the preposition to reflecting an allative case inflection in Finnish. (The Finnish influence on this construction is quite clear, partly from the fact that no speakers of Swedish in the same study ever used the preposition to in the same context.) What mattered most for the Finnish learners was their apparent belief that the semantics of Finnish and English were similar enough. This and other kinds of evidence suggest that the domains of language related to function, i.e. semantics and pragmatics, provide the primary basis for interlingual identifications.

Affect and Crucial Similarity The judgements about meanings that learners make have a somewhat different status from the semantic and pragmatic descriptions that underlie contrastive analyses. About 25 years ago, studies that successfully countered the scepticism of Wardhaugh and others began to emphasise learners’ actual assessments (e.g. Schachter, 1974; Kellerman, 1977, 1978; Wode, 1983). Wode, for example, stressed the need for learners to notice (whether consciously or not) a crucial similarity between the native and target language. Despite the intuitive soundness of Wode’s analysis, researchers have not yet been able to provide a full answer to the question of just what can make a similarity crucial. Among the necessary elements of a satisfactory answer is the relation between meaning and affect. Learner judgements must be quite complex, due in no small measure to the complicated intersection of emotion and meaning which has been discussed by many linguists, as in the classic study of semantics by Ogden and Richards (1923) appropriately titled The Meaning of Meaning. The authors argue that emotional meanings permeate human language, and to support their argument they append a discussion of phatic communication by the anthropologist Bronislw Malinowski. At the same time, Ogden and Richards explicitly follow in the footsteps of the English empiricists Locke and Hume in insisting on the need to distinguish emotional from referential language, since the latter entails a set of norms different from those in phatic communication. When using their native language, people have many communicative needs, and since at least some of those needs will arise when using a second

9612Arabski.indb 29

13/01/2006 08:43:03

30

Cross-linguistic Influences in the Second Language Lexicon

language, any interlingual identification by learners may well involve interactions of referential and phatic meanings found in the native language. Learner judgements in such cases will accordingly constitute a kind of stimulus appraisal, a notion, which has been discussed at length by Schumann (1997). This notion characterises many kinds of affective behaviour as purposive and practical: humans along with other living creatures assess stimuli in the world to judge if any particular stimulus will enhance or hinder the fulfilling of needs and desires. With specific reference to interlingual identifications, we can surmise that stimulus appraisal entails, inter alia, a judgement about communicative utility. Thus when speakers of Swedish use cleft sentences based on their native language, they have implicitly conducted a stimulus appraisal. In the case of the writer of the sentence it were he som take the bred, there seems to be an implicit assumption that the syntactic merger of affective and referential information in the Swedish cleft pattern will help in communicating affect and reference in English.

Grammatical Patterns and Habitual Thought The main example of affective transfer in this paper was a cleft sentence, which is one member of a large family of focusing constructions. The notion of stimulus appraisal seems applicable not only to clefts but to other kinds of focus as in the case of the Tok Pisin constructions described by Sankoff. Other areas of grammar likewise lend themselves to affective transfer, and again stimulus appraisal seems applicable. For instance, a look at the Spanish of Galicia in northwest Spain shows apparent instances of affective transfer of what is called the dative of solidarity (Goyanes et al., 1996: 31). Another area of grammar where some kind of affective transfer seems probable is negation (Odlin, 1998), and still other areas will likely show similar cases of affect and transfer. Some linguists may wonder why anyone would look at syntax or morphology for examples of affective transfer when, after all, intuition suggests that lexis, pragmatics, or phonology would be more obvious areas. Indeed the literature on language contact and second language acquisition shows some excellent examples in those areas such as a study of African idioms in English and French vernaculars in the Americas (Rickford & Rickford, 1976), work on apologies in Hebrew (Olshtain, 1983), and a phonological study of prestige variants of Thai speakers pronouncing the /r/ of American English (Beebe, 1980). Despite the clear importance of studies such as those just cited, the relation between affect and grammatical structure seems especially important to study. Examples such as cleft sentences and the dative of

9612Arabski.indb 30

13/01/2006 08:43:04

Could a Contrastive Analysis Ever be Complete?

31

solidarity strengthen the argument of Ochs and Schiefflin (1989) that affect has pan-linguistic dimensions or, as the title of their article puts it, that ‘language has a heart’. Moreover, the examples given show that grammar involves more than just formal properties reducible to an axiomatic system. Even if some formalists (e.g. Gregg, 1989) have difficulty in accepting functionalist approaches to grammar (e.g. Givón, 1984), expressions of affect are among the meanings of meaning that syntax and morphology serve to express. Yet another reason to look at affect in grammatical structure seems the most compelling of all, especially for transfer researchers. John Lucy, a leading proponent of relativist analyses of language, has argued for a relation between typological diversity in grammatical structures and what he calls ‘general cognitive dispositions’ (1992: 133). For Lucy, grammatical structure can and does offer clues to ‘habitual thought’, a notion speculated on by Whorf (1956) and other relativists but empirically tested by Lucy in a series of remarkable experiments (1992). Although Lucy does not discuss the question of emotions, the fact that grammatical structure interacts with human affect suggests that habitual thought will have an affective dimension. Language transfer likewise has an affective dimension as seen in the transferability of cleft and other focus structures. Accordingly, it becomes natural to wonder if transfer ever involves affective dimensions of habitual thought. To probe this question, second language researchers will need to consider the work of other linguists as well as other specialists who have been trying to sort out the universal and particular of human affect. All of the following questions thus seem relevant to a sound theory of transferability: (1) Are there emotions common to the human race? (2) Are such emotions expressed similarly in all human languages? (3) Does every language have its own distinctive emotional repertory? (4) Does such a repertory cause cognition to vary from one language community to the next? (5) Does the distinctive repertory operate as a constraint on second language acquisition? For contrastive analysis, the key research problem is threefold: first, to verify the reality of language-specific emotions; second, to verify that such emotions influence habitual thought as coded in language; and third to verify that such coding is transferable. Although not yet widely pursued, these three verification issues can be studied empirically. To solve the threefold problem, however, linguists will have to address the wider issues raised in the five questions. Accordingly the debate over

9612Arabski.indb 31

13/01/2006 08:43:05

32

Cross-linguistic Influences in the Second Language Lexicon

the universality of some human emotions becomes one that we cannot afford to ignore. Charles Darwin (1872 /1998) considered a wide range of evidence supporting such universality, whereas Margaret Mead (1975) emphasised the importance of culturally specific patterns of affect. The traditions fostered by Darwin and Mead have continued, and there are many insightful recent studies in the universalist and relativist veins (e.g. Ekman, 1994; Lutz & White, 1986; Lutz & Abu-Lughod, 1990; Wierzbicka, 1999). It is beyond the scope of this paper to say much about the debate, and I have in fact written more about it in other work (Odlin, 1998, 2002b, 2003). In any case, I hope I have made clear just why it matters so much for second language research. If contrastive analysts must take seriously the complex questions of linguistic relativity and human affect, good research may help to shed light on the question of how much the acquisition of a second language can change one’s identity and one’s personality. In a memoir of her immigration from Poland to North America, Eva Hoffman (1988: 146) recalls that her mother worried that her daughter was becoming ‘English’, which was a maternal way of saying emotionally cold. Hoffman herself had a sense that her paralinguistic repertory was changing and that her gestures were more subdued as a result of using English. However true this personal conviction, some research on paralanguage by Hearn (1989) suggests that the facial expressions of Poles and Americans are more mutually intelligible than what Hoffman’s memoir might lead us to believe. Good second language investigations may never be able to resolve such discrepancies in our views of human differences and similarities, but research can surely help us to look at the complexities with ever more discerning eyes. References Abdullah, K. and Jackson, H. (1998) Idioms and the language learner: Contrasting English and Syrian Arabic. Languages in Contrast 1, 83–107. Beebe, L. (1980) Sociolinguistic variation and style shifting in second language acquisition. Language Learning 30, 433–47. Birdsong, D. (1999) Introduction: Whys and why nots for the Critical Period Hypothesis in second language acquisition. In David Birdsong (ed.) Second Language Acquisition and the Critical Period Hypothesis (pp. 1–22). Mahwah, NJ: Lawrence Erlbaum. Celce-Murcia, M., Larsen-Freeman, D. and Williams H. (1999) The Grammar Book: An ESL/EFL Teacher’s Course (2nd edn). Boston: Heinle & Heinle. Darwin, Ch. (1872/1998) The Expression of Emotion in Man and Animals. Oxford: Oxford University Press. Dusková, L. (1984) Similarity: An aid or hindrance in foreign language learning? Folia Linguistica 18, 103–15.

9612Arabski.indb 32

13/01/2006 08:43:06

Could a Contrastive Analysis Ever be Complete?

33

Ekman, P. (1994) All emotions are basic. In P. Ekman, J. Richard and J. Davidson (eds) The Nature of Emotion: Fundamental Questions (pp. 15–19). New York: Oxford University Press. Faraclas, N. (1988) Nigerian Pidgin and the languages of southern Nigeria. Journal of Pidgin and Creole Languages 3, 177–97. Filppula, M. (1986) Some Aspects of Hiberno-English in a Functional Sentence Perspective. Joensuu, Finland: Joensuu Publications in the Humanities. Filppula, M. (1999) The Grammar of Irish English. London: Routledge. Firbas, J. (1992) Functional Sentence Perspective in Written and Spoken Communication. Cambridge: Cambridge University Press. Fisiak, J., Lipinska-Grzegorek, M. and Zabrocki, T. (1978) An Introductory EnglishPolish Contrastive Grammar. Warsaw: Panstwowe Wydawnictwo Naukowe. Givón, T. (1984) Syntax, Volume I. Amsterdam: John Benjamins. Goyanes, H., Núñez, C., Patricia R. and Túñez, M. (1996) Información en Galego. Santiago de Campostella: Lea. Gregg, K. (1989) Second language acquisition theory: A generativist perspective. In Susan Gass and Jacquelyn Schachter (eds) Linguistic Perspectives on Second Language Acquisition (pp. 15–40). Cambridge: Cambridge University Press. Guo, J.S. (1999) From information to emotion: The affective function of rightdislocation in Mandarin Chinese. Journal of Pragmatics 31, 1103–28. Han, Zhao-Hong (2001) Fine-tuning corrective feedback. Foreign Language Annals 34, 582–99. Hearn, V. (1989) Interpretations of and attitudes toward emotional expression: A Polish-American comparison. Unpublished PhD dissertation. The Wright Institute. Hoffman, E. (1988) Lost in the Translation. New York: Dutton. Irvine, J. (1990) Registering affect: Heteroglossia in the linguistic expression of emotion. In Catherine Lutz and Lila Abu-Lughod (eds) Language and the Politics of Emotion (pp. 126–61). Cambridge: Cambridge University Press. Jarvis, S. (1998) Conceptual Transfer in the Interlanguage Lexicon. Bloomington: Indiana University Linguistics Club. Jarvis and Odlin, T. (2000) Morphological type, spatial reference, and language transfer. Studies in Second Language Acquisition 22, 535–56. Kamimoto, T., Shimura, A. and Kellerman, E. (1992) A second language classic reconsidered: The case of Schachter’s avoidance. Second Language Research 8, 251–77. Kellerman, E. (1977) Towards a characterisation of the strategy of transfer in second language learning. Interlanguage Studies Bulletin 2, 58–145. Kellerman, E. (1978) Giving learners a break: Native language intuitions about transferability. Working Papers in Bilingualism 15, 59–92. Kellerman, E. (1995) Crosslinguistic influence: Transfer to nowhere? In W. Grabe et al. (eds) Annual Review of Applied Linguistics, 15 (pp. 125–50). New York: Cambridge University Press. Lucy, J. (1992) Grammatical Categories and Cognition. Cambridge: Cambridge University Press. Lutz, C. and White, G. (1986) The anthropology of emotions. Annual Review of Anthropology 15, 405–36. Lutz, C.A. and Abu-Lughod, L. (eds) (1990) Language and the Politics of Emotion. Cambridge: Cambridge University Press.

9612Arabski.indb 33

13/01/2006 08:43:06

34

Cross-linguistic Influences in the Second Language Lexicon

Mead, M. (1975) Review of Darwin and Facial Expression: A Century of Research in Review Paul Ekman (ed.) Journal of Communication 25, 209–13. Nakahama, Y. (2003) Cross-linguistic influence on the development of referential topic management in L2 Japanese oral narratives. Unpublished PhD dissertation. Georgetown University. Ochs, E. and Schiefflin B. (1989) Language has a heart. Text 9, 1, 7–25. Odlin, T. (1989) Language Transfer. Cambridge: Cambridge University Press. Odlin, T. (1990) Word order transfer, metalinguistic awareness, and constraints on foreign language learning. In B. VanPatten and J.F. Lee (eds) Second Language Acquisition/Foreign Language Learning (pp. 95–117). Clevedon: Multilingual Matters. Odlin, T. (1991) Irish English idioms and language transfer. English World-Wide 12, 175–93. Odlin, T. (1992) Transferablility and linguistic substrates. Second Language Research 8, 171–202. Odlin, T. (1997) Bilingualism and substrate influence: A look at clefts and reflexives. In J. Kallen (ed.) Focus on Ireland (pp. 35–50). Amsterdam: John Benjamins. Odlin, T. (1998) On the affective and cognitive bases for language transfer. In R. Cooper (ed.) Compare or Contrast? (pp. 81–106). Tampere, Finland: University of Tampere Studies in English. Odlin, T. (2001) Language transfer and substrate influence. In R. Mesthrie (ed.) Concise Encyclopedia of Sociolinguistics (pp. 499–503). Amsterdam: Elsevier. Odlin, T. (2002a) The contrastive prediction in second language acquisition and its evil twin. University of London, 21 February 2002. Odlin, T. (2002b) Language transfer and cross-linguistic studies: Relativism, universalism, and the native language. In R. Kaplan (ed.) The Oxford Handbook of Applied Linguistics (pp. 253–61). New York: Oxford University Press. Odlin, T. (2003) Cross-linguistic influence. In C. Doughty and M. Long (eds) Handbook on Second Language Acquisition (pp. 436–86). Oxford: Blackwell. Ogden, C.K. and Richards, I.A. (1923) The Meaning of Meaning: A Study of the Influence of Language upon Thought and of the Science of Symbolism (with supplementary essays by B. Malinowski and F. G. Crookshank). New York: Harcourt, Brace & World. Olshtain, E. (1983) Sociocultural competence and language transfer: The case of apology. In Susan Gass and Larry Selinker (eds) Language Transfer in Language Learning (pp. 232–49). Rowley, MA: Newbury House. Orr, G. (1987) Aspects of the second language acquisition of Chichewa noun class morphology. Unpublished PhD dissertation. University of California, Los Angeles. Ortony, A., Clore, G.L. and Collins, A. (1988) The Cognitive Structure of Emotions. Cambridge: Cambridge University Press. Quirk, R., Greenbaum, S., Leech, G. and Svartvik, J. (1985) Comprehensive Grammar of the English Language. London: Longman. Rickford, J. and Rickford, A. (1976) Cut-eye and suck-teeth: African words and gestures in New World guise. Journal of American Folklore 89, 294–309 Ringbom, H. (1987) The Role of the First Language in Foreign Language Learning. Clevedon: Multilingual Matters. Sankoff, G. (1993) Focus in Tok Pisin. In F. Byrne and D. Winford (eds) Focus and Grammatical Relations in Creole Languages (pp 117–40). Amsterdam: John Benjamins.

9612Arabski.indb 34

13/01/2006 08:43:07

Could a Contrastive Analysis Ever be Complete?

35

Schachter, J. (1974) An error in error analysis. Language Learning 24, 205–14. Schumann, J. (1997) The Neurobiology of Affect in Language. Malden, MA: Blackwell. Selinker, L. (1992) Rediscovering Interlanguage. London: Longman. Stockwell, R., Bowen, J.D. and Martin, J. (1965) The Grammatical Structures of English and Spanish. Chicago: University of Chicago Press. Thomason, S. and Kaufmann, T. (1988) Language Contact, Creolization, and Genetic Linguistics. Berkeley: University of California Press. Trèvise, A. (1986) Is It transferable topicalization? In Eric Kellerman and Michael Sharwood Smith (eds) Crosslinguistic Influence and Second Language Acquisition (pp. 186–206). New York: Pergamon. Vainikka, A. and Young-Scholten, M. (1998) The initial state in the L2 acquisition of phrase structure. In S. Flynn, M. Gita and W. O’Neil (eds) The Generative Study of Second Language Acquisition (pp. 17–34). Mahwah, NJ: Lawrence Erlbaum. Wardhaugh, R. (1970) The contrastive analysis hypothesis. TESOL Quarterly 4, 123–30. Weinreich, U. (1953/1968) Languages in Contact. The Hague: Mouton. White, L. (2000) Second language acquisition: From initial to final state. In J. Archibald (ed.) Second Language Acquisition and Linguistic Theory (pp. 130–55). Oxford: Blackwell. Whorf, B.L. (1956) Language, Thought, and Reality: Selected Writings. Cambridge: MIT Press. Wierzbicka, A. (1999) Emotions across Languages and Cultures: Diversity and Universals. Cambridge: Cambridge University Press. Wode, H. (1983) (ed.) On the systematicity of L1 transfer in L2 acquisition. Papers on Language Acquisition, Language Learning and Language Teaching (pp. 144–49). Heidelberg: J. Groos Verlag.

9612Arabski.indb 35

13/01/2006 08:43:08

Chapter 4

The Importance of Different Types of Similarity in Transfer Studies HÅKAN RINGBOM Linguists are fascinated with differences between languages, and crosslinguistic differences have, at least since the advent of Contrastive Analysis, been central for a long time in SLA research, while crosslinguistic similarity has been in the background, with only occasional side references. In some recent works on transfer the importance of both similarity and difference are noted, e.g. Kellerman (1995); Jarvis (2000). But it is similarity that is the more important concept, especially to the learner. We can, in fact, not really talk about differences unless there is a basic underlying similarity. Pit Corder was one of those who already stated this in passing 30 years ago: ‘In order to compare anything the dimensions or categories used must be applicable to both objects’. (1973: 234). Or, in the words of Carl James, ‘it is only against a background of sameness that differences are significant’ (1980: 169; see also Ringbom, 1987: 33ff.). Transfer depends, as Kellerman pointed out more than 25 years ago (1977), on cross-linguistic similarity, perceived similarity. Perceived similarity is not the same as ‘objective’ cross-linguistic similarity, which is a concept that might perhaps be defined theoretically, although there are problems here. Ellegård (1978) made a brave attempt at a definition, but it has attracted no attention whatever that I am aware of. If we start looking closely at the concept of cross-linguistic similarity in SLA, we will find that the concept is far from clear-cut. Similarity works differently at different stages of learning, for different modes of learning, and with different individual learners. While most perceived similarities will facilitate learning, there are also instances where similarity can lead to errors, as in false friends. And, of course, we also have to reckon with intra-linguistic similarities, not only crosslinguistic ones. If we go into the problem of cross-linguistic similarity relations, however, we realise that the main types are, in fact, not two, but 36

9612Arabski.indb 36

13/01/2006 08:43:09

Different Types of Similarity in Transfer Studies

37

three in number. We should distinguish between the following different types of cross-linguistic similarity relations: (1) A similarity relation where the learner is able to establish a one-toone relationship with another unit, usually in the L1. Across related languages there will be cognates that facilitate both comprehension and learning, especially at the early stages, even if some of them may later turn out to be false, or partially false, cognates. Also, grammatical structures across Germanic languages are largely similar, with the same basic categories. (2) A difference relation, where the learner can perceive both similarity and difference. For grammar, Swedish learners of German will find that German has three grammatical genders, not two which they are used to from L1. This means that there are comprehension and learning problems, but the learners certainly are already aware of the category of grammatical gender. Most native speakers, however, usually do not have much declarative knowledge of their L1, so a thorough mastery of some non-native language achieved in a classroom context, might be more useful for the learning of a new language. (3) A zero relation where the learner cannot relate the TL aspect to previous linguistic knowledge, as when an English learner is confronted with grammatical gender in German. This is a situation confusing the learner, who will have considerable time-consuming organisational problems at the beginning. In Pit Corder’s words, ‘Absence of perceived similarities produces learning conditions where learning is delayed at the important initial stages’ (1983: 66). Only after some mastery of the organisational problems has been achieved can the learner proceed to the next stage of clearly delineated choice problems. Where there are zero relations, TL categories not existing in L1 are commonly perceived as redundant at the early stages of learning. Foreign learners of Finnish tend to see most of the 15 Finnish case endings as redundant, just as Finnish learners of English tend to leave out the articles, especially the definite article. These three relations are manifested differently in different contexts. Particularly important here is the item-system distinction. Learners have different expectations for items than for systems: everybody is aware that the items are different in languages other than the L1, but tends to assume that the procedures and underlying systems are basically the same in the target language as in L1, or some other language known, unless they have been shown to be different.

9612Arabski.indb 37

13/01/2006 08:43:10

38

Cross-linguistic Influences in the Second Language Lexicon

The distinction between items and systems is an important distinction that has not been much present in SLA discussions. It originates from Cruttenden’s paper in 1981, dealing with L1-acquisition, and it can be applied to FL learning also by linking it with the concept of transfer. According to Cruttenden, item learning ‘involves a form which is uniquely bonded with some other form or with a unique referent, whereas system learning involves the possibility of the commutation of forms or referents while some (other) form is held constant. The two-stage process applies to phonology, intonation, morphology and syntax, and semantics (1981: 79)…. Learning takes place initially on an item by item basis at all levels of language.’ If the distinction is applied to the learning of languages other than the L1, the unique bondage characteristic of item learning is predominantly cross-linguistic, especially at early stages of learning and in a formal language learning context, since the relevant concepts already exist in the learner’s mind. In comprehension learners tend to look for similarities, especially formal similarities between items in order to establish a (simplified) equivalence relation between them. Equivalence between individual items is difficult to perceive without an existing underlying functional equivalence between categories. Now, perceived or assumed cross-linguistic similarity is what lies behind the concept of transfer, which, generally speaking, means a process whereby the learner makes use of linguistic resources other than their knowledge of the language in which communication takes place. Transfer, however, has many different shapes and it occurs in all the three similarity relationships. Transfer should be discussed with reference to these similarity relations, to items and systems, to the distinction between similarity of form and similarity of function/meaning, and, not least important, to the different modes of comprehension and production. We must take into account not only transfer in production but also transfer in comprehension, since comprehension precedes learning and at least some comprehension must occur before production. Unfortunately transfer in comprehension has hardly been studied at all. Some analyses within the functionalist framework of the competition model has, though, been done, for example by Kilborn (1989). There are at least five different dimensions of cross-linguistic similarity relevant to transfer studies: • • •

9612Arabski.indb 38

Perceived and assumed similarity vs. ‘objective’ similarity. Similarity, difference and zero relations in items and systems. Similarity of form vs. similarity of function/meaning.

13/01/2006 08:43:10

Different Types of Similarity in Transfer Studies • •

39

The effect of similarity affecting comprehension vs. similarity affecting production. Similarity for L1-transfer vs. L2-transfer in L3-learning.

I believe we can distinguish three levels of transfer, which I will label overall level, item level and system level. The overall level is an umbrella term which depends on how much cross-linguistic similarity the learner can generally perceive between items and systems in the two languages, beginning from a common alphabet, phonemes in common and similar phonotactics, over the division into grammatical categories (case, gender, word classes) to the number of cognates and other lexical similarities. Item similarity is normally based on an underlying similarity of form and an associated, assumed similarity of function or meaning between source language and target language. In system transfer the learner assumes an underlying similarity of function, while formal similarity is not necessarily involved. The distinction between formal and functional/semantic similarity is not hard and fast, the very vagueness of the concept perceived formal similarity makes it a continuum. Error analysis, however, shows clear instances from both ends of the scale. Overall transfer is particularly evident in comprehension and has a general facilitating effect on acquisition: the result is positive transfer. The more closely related the languages are, the more overall transfer we can find. In the Nordic context, Swedish, Norwegian and Danish are, on the whole, mutually comprehensible for native speakers of a standard variety of these languages. Scandinavians do not have to learn to produce another Scandinavian language, they can communicate successfully merely by speaking their own L1. But there are also cases where we can find useful similarities leading to positive transfer across totally unrelated languages, such as Swahili and Finnish. A former colleague of mine, a professor of sociology, who had spent two years in Tanzania, answered a question whether it was difficult to learn Swahili with the words ‘If you know Finnish there will be no major problem learning Swahili’. In this case both languages are highly agglutinative, and Finnish, which was my colleague’s L2, provided much more concrete help than the many Germanic languages he knew well as L1, L3, L4 and L5. Overall transfer relies on individual items and on some functional equivalence perceived between the underlying systems. The more target language items that can be perceived to be similar to items in a language the learner already knows, the more overall transfer there is. Also, if the systems in the two languages are congruent or nearly congruent, overall transfer will occur. Where, on the other hand, there is a basic zero similarity relation, there is

9612Arabski.indb 39

13/01/2006 08:43:11

40

Cross-linguistic Influences in the Second Language Lexicon

also no overall transfer. Overall transfer illustrates a principle well known in applied linguistics: the whole is more than the sum of the individual parts, i.e. a large number of item similarities put together produce an even larger, substantial facilitating effect on learning. While overall transfer is clearly positive, item transfer generally also has positive effects on learning, occasionally a negative effect. It is worth emphasising that items in item transfer are not only words, lexical items, they can just as well be phonemes, morphemes or phrases. Where the learner is able to establish cross-linguistic similarity between individual items, such identification may or may not lead to errors, though less often errors in comprehension than in production. In comprehension the surrounding context, linguistic and extra-linguistic, helps to eliminate errors that are often made in production. Initially, learners reduce their workload by making use of an oversimplified equivalence hypothesis L2 = L1, mapping the meanings or functions of L2-items directly on to existing L1-items in comprehension and L1-items on to L2-items in production. Such cross-linguistic one-to-one relationships between lexical items assumed by learners at early stages of learning often work well in establishing primary counterparts (to use Arabski’s excellent term (1979) for perceived lexical translation equivalents), but gradually such oversimplified relationships will have to be modified as learning progresses and the learner develops a more realistic understanding of the lexical and conceptual organisation of the target language. Only for most technical and scientific vocabulary the one-to-one relationship need not be modified. At early learning stages learners tend to focus on form rather than the more abstract concepts of meaning or function, and perceived formal similarities aid them in establishing cross-linguistic relations in long-term memory. Whereas item transfer is normally based on an assumed underlying similarity of form as well as function, system transfer assumes functional equivalence without formal similarity (necessarily) being involved. But since functional systems in two languages are hardly ever fully congruent, system transfer often leads to error and is thus mostly negative transfer. At early stages of learning the learner’s tendency is to assume one-to-one relationships, which are in frequent conflict with the actual network of one-to-many or many-to-many relations in the real world. There is a difference between perceived and assumed item similarity, relating to the difference between comprehension and production. Production relies on the learner’s intentions in ways comprehension does not, and the two thus illustrate two different modes of use calling upon different retrieval procedures. In comprehension an approximate L1equivalent can often be retrieved from even a slightly similar L2-stimulus.

9612Arabski.indb 40

13/01/2006 08:43:12

Different Types of Similarity in Transfer Studies

41

Where similarity can be perceived between existing words and structures a general facilitating effect for comprehension and learning occurs, though in some cases, notably deceptive cognates, formal similarity can lead to errors. Assumed similarity, on the other hand, occurs in production, in that the L1-word or structure is merely assumed to exist also in the target language. Assumption can be based on previous perception, but that is not always the case. For two of the three similarity relations, perceived and assumed similarity are very close to each other, partly no doubt because comprehension and production work in constant interaction. When there is a zero similarity relation, however, there is a difference, which is manifested in the use or non-use of transfer. In target languages totally different from the L1 there will be little visible transfer in comprehension and learning, but transfer will occur in production, when the learner does not perceive, but merely assumes that items and systems in the target language will work more or less in the same way as in L1. This, in Kellerman’s (1995) terms, is ‘transfer to nowhere’. Such transfer, based on assumed functional similarity, is clearly manifested in L1-based errors, for example in grammatical errors in English made by Finnish learners. Another manifestation of transfer in production is avoidance. In some circumstances, i.e. the learning of wholly different languages in a classroom context with a restricted TL input, the receptive vocabulary of learners has been found to be no more extensive than their productive vocabulary (Takala, 1984). The main explanation for this, in addition to the limited L2-output in a traditional Finnish classroom in the 1970s, lies in the insignificant role of transfer. The considerable linguistic differences between Finnish and English have concealed the similarities essential for transfer in comprehension. My last point concerns the difference between lexical L1-based errors and L2-based errors in L3-learning, with material from Finland, some of it already published. Table 4.1 shows the different types of transfer-based errors in production. In previous studies (Ringbom, 1987: 118, 2001: 61) the frequency distribution of these categories in an English vocabulary test for Finnish and Finland-Swedish learners has been given. A very similar distribution can be seen in a composition test of Swedish given to Finnish comprehensive school students with English as their L2. Table 4.2 shows that item transfer, usually based on formal similarity, will frequently occur from a related L2, but rarely from an unrelated L1. System transfer, on the other hand, is frequent from L1, but rare from a related L2. The fairly unadvanced stage of the learners in this study (comprehensive school students) and the better knowledge of English than

9612Arabski.indb 41

13/01/2006 08:43:13

42

Cross-linguistic Influences in the Second Language Lexicon

Table 4.1 Lexical item and system transfer errors in production Linguistic process

Underlying cause

Transfer of form/meaning

From what lg.

Example

A+B Language switch & coinage (hybrids)

Insufficient awareness of intended form, instead of which (a modified form of) a SL word is used

Mainly form (results in non-existing TL word). Function words with no formal similarity often switched between nonnative lgs.

L1 or L2 (also from lgs not very well known by learner)

I was much pigger after the walking (Sw. pigg = ‘refreshed, alert’)

C Totally or partially deceptive cognate

Awareness of an existing TL form, but confusion caused by formal similarity (but no or only partial semantic similarity) to a word in another language

Form: results in existing TL word with meaning different from the intended one

L1 or L2

This affects the health of people. I am one offer of this. (Sw. offer = ‘victim’)

D Semantic extension of single lexical units

Awareness of existing TL form, but not of necessary semantic restrictions

Meaning

L1 or, occasionally, very advanced L2proficiency

Football company pro football club (Fi. seura = both ‘club’ and ‘company’)

E Calques – word units (compounds, phrasal verbs, idioms)

Awareness of existing TL units, but not of semantic/ collocational restrictions

Meaning (some similarity of form may also be present)

L1 or possibly very advanced L2proficiency

Tourists are taking place everywhere. (Sw. ta plats = ‘take up space’)

9612Arabski.indb 42

13/01/2006 08:43:14

43

Different Types of Similarity in Transfer Studies

Table 4.2 Lexical transfer errors in Swedish (L3) compositions made by Finnish comprehensive school students with English as their L2 (n=180). Item transfer

System transfer

Deceptive Calques cognates, hybrids, language shifts From Fi. (L1) From Eng. (L2)

Semantic extensions

9

63

51

130

5

2

Based on Laitakari (2001)

of Swedish that they have means that occasional examples of system transfer from L2 also occur. Similarly, Swedish-speaking learners of English, most of whom are fluent in Finnish, occasionally take refuge to lexical L2 -based system transfer from Finnish. Examples of loan translations from Finnish made by Swedish speakers are thus firesticks for ‘matches’, oldboy for ‘bachelor’ and postmarks for ‘stamps’. Though system transfer tends to originate from the L1, not from the L2, it is dangerous to maintain that certain types of error, in this case L2-based calques or semantic extension errors, do not generally occur. Any language teacher knows that details of individual learner production are impossible to predict. Learners often come up with strange words or constructions that teachers or researchers simply could not have foreseen. What is important here is frequencies, we have to talk about trends. Psychotypology is very important, but it is not the only relevant factor for determining the relative strength of different types of transfer. On the basis of a case study, Williams and Hammarberg (1998) argued convincingly that other variables also enter the picture, and this is clearly true. One aspect that some learners seem to make use of is the so-called foreign language effect. This means that learners avoid making use of their L1, and are more prone to be influenced by another foreign language, consciously or not. Other relevant variables are the learner’s stage of L3-learning and the learner’s L2proficiency. At the earliest stages of L3-learning there appears to be much more L2-transfer than later on. And if the learner is a near-native speaker or even a very advanced learner of L2 it will make him prone to make use of such L2-based strategies as are normally borrowed only from L1. For a conclusion, I shall summarise some of the main points I have tried to make in this paper, and some corollaries to them.

9612Arabski.indb 43

13/01/2006 08:43:15

44

Cross-linguistic Influences in the Second Language Lexicon

(1) Perceived or assumed similarity is not the same as ‘objective’ similarity. Perceived similarity is relevant for comprehension and learning while in production similarity is assumed rather than perceived. Such assumptions may or may not be based on previous perception. (2) Though similarities can be established much more easily across related than unrelated languages, they may also be perceived in important respects across totally unrelated languages. (3) Perceived similarity between languages, which is manifested in overall transfer, can be asymmetric. Danes tend to understand Swedish better than Swedes understand Danish. (4) Mutual comprehensibility depends primarily on a large number of lexical cross-linguistic similarities, while ease or difficulty of learning is largely determined by how much structural system similarity can be perceived across languages. This paper has tried to show how important the concept of cross-linguistic similarity is in foreign language learning, particularly where aspects of transfer are discussed. Many of the points made need to be further elaborated and supported by experimental work. The role played by cross-linguistic similarity varies both quantitatively and qualitatively, depending on a number of other, closely interlinked variables. Also, the relations between the variables briefly discussed here need to be further specified. References Arabski, J. (1979) Errors as Indications of the Development of Interlanguage. Katowice: Uniwersytet Slaski. Corder, S.P. (1973) Introducing Applied Linguistics. Harmondsworth: Penguin. Corder, S.P. (1983) A role for the mother tongue. In S. Gass and L. Selinker (eds) Language Transfer in Language Learning. Rowley, MA: Newbury House. Cruttenden, A. (1981) Item-learning and system-learning. Journal of Psycholinguistic Research 10 (1), 79–88. Ellegård, A. (1978) On measuring language similarity. In J. Weinstock (ed.) Nordic Languages and Linguistics. University of Texas. James, C. (1980) Contrastive Analysis. London: Longman. Jarvis, S. (2000) Methodological rigor in the study of transfer. Language Learning 50 (2), 245–309. Kellerman, E. (1977) Towards a characterization of the strategy of transfer in second language learning. Interlanguage Studies Bulletin: Utrecht 2, 58–145. Kellerman, E. (1995) Cross-linguistic influence: Transfer to nowhere. Annual Review of Applied Linguistics 15, 125–50. Kilborn, K. (1989) Sentence processing in a second language: The timing of transfer. Language and Speech 32 (1), 1–23. Laitakari, K. (2001) Skillnaden mellan L1 och L2 vid inlärningen av ett tredje språk – En komparation av finsk och engelsk transfer i inlärarsvenska. Unpublished MA thesis, Åbo Akademi University.

9612Arabski.indb 44

13/01/2006 08:43:16

Different Types of Similarity in Transfer Studies

45

Ringbom, H. (1987) The Role of the First Language in Foreign Language Learning. Clevedon: Multilingual Matters. Ringbom, H. (2001) Lexical transfer in L3 production. In J. Cenoz, B. Hufeisen and U. Jessner (eds) Cross-linguistic Influence in Third Language Acquisition: Psycholinguistic Perspectives (pp. 59–68). Clevedon: Multilingual Matters. Takala, S. (1984) Evaluation of Students’ Knowledge of English Vocabulary in the Finnish Comprehensive School. Reports from the Institute for Educational Research 350. University of Jyväskylä. Williams, S. and Hammarberg, B. (1998) Language switches in L3 production: Implications for a polyglot speaker. Applied Linguistics 19 (3), 295–333.

9612Arabski.indb 45

13/01/2006 08:43:17

Chapter 5

Language Contact vs. Foreign and Second Language Acquisition ˙BIETA MAN ´CZAK-WOHLFELD ELZ Although the title of this paper may sound a little strange, it seems that there is a certain relation between language contact which results in the process of borrowing, and foreign/second language acquisition which is also a demonstration of a different sort but still of language contact. Nemser (1991: 345) even goes so far as to assert that the process of borrowing and the process of foreign/second language acquisition deal with ‘aspects of a unified phenomenon’. However, it is not the purpose of this paper to prove the above claim as it has already been done by Nemser himself. On the contrary, we aim at showing how the result of the process of borrowing may be used in the process of foreign/second language acquisition. Since the readership of this volume may be in their majority specialists in foreign and second language acquisition and not necessarily experts in the theory of language contact, we would first like to present a very brief overview of the most basic assumptions of this theory and then concentrate on its reference to foreign and second language acquisition. It is a well-known fact that language is a system that undergoes constant changes. The most conspicuous ones are seen on the lexical level. Within our life span all of us are able to notice the establishment of new vocabulary items. They are introduced in three different ways: (1) The invention of neologisms; e.g. in Poland after World War II there was a tendency to eliminate germanisms and substitute them for Polish equivalents, as was the case of German szlafrok ‘a dressing gown’ replaced with the Polish neologism podomka. (2) The change of word meanings; e.g. in Polish the word of Latin origin konwencja ‘convention’ was originally used only in the sense of general agreement or a custom or customary practice (SWO, 1999), and relatively recently a new meaning taken from American English was added to it, that is ‘an assembly of the delegates of a political party to select candidates for office’ (COD, 1995; SWO, 1999). 46

9612Arabski.indb 46

13/01/2006 08:43:18

Language Contact vs. Foreign and Second Language Acquisition

47

(3) The addition of borrowings. The last way of enlarging vocabulary will be of our concern. Therefore, we should first define the term. It seems that Haugen’s (1950: 212) suggestion, although a very old one, is most adequate: The heart of our definition of borrowing is then the attempted reproduction in one language of patterns previously found in another. Why do I consider this definition to be most appropriate? This is due to its general character which covers different types of borrowing, not necessarily lexical items but also affixes, structures or even phonemes, since the term patterns encompasses every linguistic unit. We all agree that the term borrowing is controversial because the very word implies the lender’s consent and an obligation to return a borrowed item. Certainly, nothing like that happens in the case of the discussed process. It is worth adding that this terminological inadequacy is present in a number of European languages. It is evident that loans result from contact between populations. Language contact is a fraction of cultural contact, which is the reason why Bloomfield (1933: 458) suggests referring to them as to cultural borrowings, which ‘show us what one nation has taught another’. A similar standpoint was earlier taken by Sapir (1921), who went even further and believed that there was a connection between linguistic borrowings and the expansion of culture. To put it more precisely, Sapir claimed that the more loanwords one nation gives to others the more civilised the society should be considered. Certainly, nowadays nobody would agree with such an attitude. Presumably, Cienkowski’s (1964) reference of culture to borrowing sounds more plausible at present as the Polish linguist suggested that only on the basis of borrowings (their types, the time of their first occurrence, etc.) can we detect contact between different nations. However, most scholars dealing with borrowings limit their considerations to linguistic aspects and we would proceed along these lines in a similar way. As is generally known, loans are introduced by bilingual speakers1 and are first used only by them and then gradually spread around. Therefore, we can speak of a dynamic process which ends in the introduction of the loanword to the language system, or to use de Saussure’s terminology, it becomes part of langue or, as referred to by Chomsky’s more modern term, part of our competence. This implies that the loan becomes a stable element in the language. The briefly discussed description of the process of borrowing is the 20th century approach, as in the 19th century it was believed that there was no

9612Arabski.indb 47

13/01/2006 08:43:18

48

Cross-linguistic Influences in the Second Language Lexicon

question of introducing loans but of mixing languages. At the present state of our linguistic knowledge we all agree that there are no mixed tongues, apart from Creole or Pidgin languages. We also know that the process of borrowing is inevitable as there are no pure natural languages. To quote Jakobson’s famous saying, ‘no language as no economics is self-sufficient’ (Buttler et al., 1973). However, it is worth explaining that there is no correlation between the number of borrowings and efficiency to communicate in a particular language. Still we have to find a certain balance between the introduction of loans and their possible overuse. It is generally accepted that if we take over a good number of loanwords we make our language richer and more modern although, at the same time, we impoverish it by eliminating native items. It is evident that we should not overuse borrowings, especially when there are native equivalents (Deroy, 1956). In the case of overusing borrowings, we speak of luxurious loanwords. Let us only add that they are usually taken over because of fashion or snobbery; however, as we know from the history of languages, in the course of time they are often eliminated; cf. the 17th and 18th century impact of galicisms on Polish, which later disappeared. Luxurious loans are usually contrasted with necessary loans that cover exotics or foreignisms, names of designates and concepts unknown in the borrowing language as well as internationalisms. To come back to the process of borrowing, it is worth stressing that, as has already been stated, new loans are introduced by bilinguals who first try to imitate ‘models’ as closely as possible. That is the reason why we are confronted with so-called citations at first and only later does the process of assimilation start on four linguistic levels of analysis, i.e. the phonetic, graphic, morphological and semantic. It is clear that the adaptation does not always take place on all the levels. Therefore, following Haugen (1950), we distinguish loans that are either imported or substituted on different planes of linguistic description. To give an example, the lexical item dČinsy in Polish exhibits substitution on the phonetic, graphic and morphological levels and importation on the semantic plane. It has to be mentioned that some loanwords never undergo any assimilation in the borrowing language and they remain quotations (often referred to by the German term Fremdwörter); however, most loans become adapted and, sticking to the German terminology, we would call them Lehnwörter. Some of these are so well-assimilated that they are not felt as loanwords, particularly by those speakers who do not know the language of origin of a borrowing, which may be illustrated by the ‘Polish’ word rum < E rum already attested at the turn of the 18th and 19th centuries in Polish. It is

9612Arabski.indb 48

13/01/2006 08:43:19

Language Contact vs. Foreign and Second Language Acquisition

49

obvious that the loans which are well assimilated often undergo derivational processes typical of the borrowing language. The next issue is connected with the classification of loans. The most common and at the same time the most general one is characterised by the following distinctions: (1) loanwords, e.g. P budČet < E budget; (2) loanblends (or hybrids), e.g. P drewland being a combination of the Polish morpheme drew- and English land; (3) loanshifts (or calques), e.g. P nastolatek structured after E teenager (Hockett, 1958). Certainly, very many different divisions have been suggested by other scholars, as is the case with Haugen (1950) who still distinguishes semantic loans. Another classification of loans is not based on their form but on the way they are introduced. Thus, we can differentiate between oral loans, i.e. those that are imported via spoken language or graphic loans which enter the borrowing language by a written medium. Both types of loanwords are introduced either directly from the donor language or via a mediating language. The latter means of importing loanwords is connected with the very definition of the term borrowing since, if we accept the idea of indirect loans, we simultaneously reject the understanding of the loanword suggested by some scholars who claim that the last source of contact indicates the origin of a word. To illustrate, let us quote Kuroczycki and Rzepka’s (1979) example of the lexical item chuligan which did not enter Polish directly from English (hooligan) but via Russian, and therefore the mentioned linguists consider the loan a russianism. The final question that should be addressed concerns the reasons for borrowing. According to Weinreich (1970), there are both linguistic and extra-linguistic causes. Among the linguistic reasons that Weinreich mentions we have: • • • • •

low frequency of native equivalents; the existence of homonyms in the borrowing language; no emotive aspects associated with native equivalents; no differentiation of native words in the same semantic field; wrong or even bad connotations of native words.

On the other hand, extra-linguistic reasons involve the need-filling motive, i.e. the necessity to name new designates or ideas as well as the prestige motive, which is manifested by the tendency to borrow lexical items from a more prestigious language.

9612Arabski.indb 49

13/01/2006 08:43:20

50

Cross-linguistic Influences in the Second Language Lexicon

Having presented a very brief outline of the process of borrowing, which according to Weinreich (1970) may be referred to as the acquisition of new elements, we would like to relate it to the problem of foreign and second language acquisition, which, as stated above, also involves the contact of two languages. I would like to focus my attention on one aspect of borrowing and foreign/second language acquisition which concerns lexical items. It is evident that without words people cannot talk about other people, animals, things, actions, states or relations. Without words there would be no sound structure, no word structure, and no syntax. It only shows that lexicon is central to language as well as foreign and second language acquisition. Therefore it seems that the choice of this aspect of language is convincing for our further discussion. In the case of closely related languages the way of acquiring lexicon is relatively easy because of the presence of cognates. This is, for instance, the case of English and Scandinavian languages (Rusiecki, 1980).2 However, the situation is different in the case of two languages that are unrelated or related distantly as English and Polish happen to be. For obvious reasons these two languages will be used as an illustration of language contact present both in the process of borrowing and in the process of foreign as well as second language acquisition. It is worth mentioning that some of the following remarks will be made from the perspective of the methodology of teaching English to Polish students. Some articles dealing with this problem have already been published, notably Rusiecki (1980); MaÚczak-Wohlfeld (1994, 1995). To summarise their main ideas it has to be stated that the authors have assumed that the principle of introducing easier matters before more complicated ones should be observed. Therefore in the quoted papers it has been suggested that first some English loanwords used in Polish should be taught. It goes without saying that English is a world language and as such it has for some time served as an important source of borrowings in Polish. This is the reason why it is claimed that at the beginning we should introduce such well-adapted anglicisms as: bar. badminton, rum, etc. All of them are well-known to Polish learners of English; their spellings and meanings in Polish agree with the English counterparts, the only difference being in their pronunciation. It is maintained that, later, another group of English loans should be mentioned, namely those that differ not only on the phonetic, but also on the graphic level, e.g. drain vs. dren, jungle vs. dČungla, gin vs. dČyn, essay vs. esej, colt vs. kolt, revolver vs. rewolwer, rumpsteak vs. rumsztyk, stress vs. stres, etc. At the next stage it is thought best to discuss those loanwords that vary from their English models not only phonetically

9612Arabski.indb 50

13/01/2006 08:43:21

Language Contact vs. Foreign and Second Language Acquisition

51

and graphically but also morphologically. These are anglicisms that take the so-called double plural (Fisiak, 1961). This simply means that English nouns are borrowed in their plural forms and are considered by Poles not knowing English well enough to be singular nouns, e.g. bamboo vs. bambus or skunk vs. skunks. We can account for such forms as bryczesy (E breeches) or dČinsy (E jeans) in the same way. It merely means that the Polish plural ending -y has been added to the English plural (the so-called pluralia tanta). Finally, the suggestion is to introduce those anglicisms that differ from their English models not only phonetically, graphically, or morphologically but also semantically; e.g. the noun dress means in English (1) ‘a one piece woman’s garment consisting of a bodice and a skirt’; (2) ‘clothing, esp. a whole outfit, etc.’ (3) ‘formal or ceremonial attire’; (4) ‘an external covering, the outward form’ (COD, 1995), whereas in Polish the anglicism dress refers to ‘a training suit’ (SWJP, 1996). As the next step it is proposed to draw the learner’s attention to so-called internationalisms, i.e. words based on Greek or Latin stems, common to a number of European languages, including English and Polish, e.g. E democracy and P demokracja, E imperialism and P imperializm, E phenomenon and P fenomen, etc. as well as lexical items from other languages which have been borrowed independently both by English and Polish, e.g. E communism and P komunizm (from French), E robot and P robot (from Czech). Needless to say, these items are in most cases spelled and pronounced distinctly and they behave differently on the morphological level in the two languages in question. As Rusiecki (1980) has correctly observed, it is a difficult task to establish a list of English loanwords and internationalisms since the knowledge of lexicon is dependent on sex, age, and the semantic field involved. This means that a person keen on sports would easily recognise English sports terms occurring in Polish. On the other hand, a person interested in politics might find sports terminology unfamiliar but comprehend words used in political journalism. However, Rusiecki (1980: 88–96) has managed to compile such a list of 200 ‘common core’ items. The introduction of the words discussed above at the very early stage of teaching enables the beginner to communicate very soon because the lexical items that the learner recognises as familiar are easy to memorise. Many of them can be useful in everyday communication, some are related to various fields of interest the learner may be willing to discuss. Needless to say, elementary information about English syntax is required but the familiarity of the lexemes and their usefulness in communication will make the process of learning to speak much shorter in comparison with the process of memorising other lexical items. Certainly, it is not asserted

9612Arabski.indb 51

13/01/2006 08:43:22

52

Cross-linguistic Influences in the Second Language Lexicon

that one should only teach anglicisms and other borrowings at a beginners’ course since loanwords basically constitute nouns and do not cover many semantic fields. That is why other parts of speech and words from different semantic fields will have to be introduced as well, but the advantage of pointing to a number of English and non-English borrowings is of psychological nature since we speed up the process of vocabulary acquisition. The next step of teaching English vocabulary would concern more advanced learners of English because the teacher is encouraged to focus on so-called false friends. They are characterised by similar phonetics but both their origin and their meanings differ very much, e.g. the English lexical item eventual means ‘occurring or existing in one course or at last’ (COD, 1995), but its Polish equivalent ewentualny denotes ‘in the case of something or of something urgent’ (SWJP, 1996), cf. MaÚczak-Wohlfeld, 1995. Finally, some attention should be paid to the so-called false anglicisms (Bantas, 1977) or pseudo-anglicisms (Filipovi², 1982) which are characterised by the omission of one English element in the receiving language, e.g. P happy-end < E happy ending, P kemping/camping < E camping site, P smoking < E smoking jacket .3 The above-mentioned suggestions concerning the way of introducing English vocabulary in relation to anglicisms were proposed in the 1980s and early 1990s. Should we follow them at the beginning of the 21st century? It seems to me that the linguistic situation is nowadays quite different and the presented suggestion of making use of anglicisms in acquiring English lexicon, i.e. in applying one type of language contact to a slightly different one, should be revised. At the very beginning it is worth realising that nowadays in Poland there are no true beginners learning English as a foreign language. Certainly, the same claim is relevant to another group of Polish learners who acquire English as a second language. Everybody seems to know some single lexical items of the type hello, sorry and even some phrases like how are you. Besides, at the present moment Poles know far more anglicisms as well as English items than before, especially if they are particularly interested in certain domains, e.g. the most popular one, computer science, and they are more conscious of their origin, which is due to much greater access to English speaking countries because of the mass media, electronic communication, travel opportunities, etc. Therefore it would most probably seem to be unnatural to start with the introduction of anglicisms for a number of reasons. First of all, nowadays many more of them are already known in their original graphic form and phonetic shape. Similarly their meaning is familiar to learners. Due to this, it might seem artificial to learners to select those words that

9612Arabski.indb 52

13/01/2006 08:43:22

Language Contact vs. Foreign and Second Language Acquisition

53

they are acquainted with at the very beginning of the process of teaching, especially that they belong to different semantic fields like music, sports, fashion, computer science, etc. Therefore if we were to follow the guidelines earlier proposed by Rusiecki (1980) and MaÚczak-Wohlfeld (1994, 1995) it might lead students to the false conclusion that only anglicisms and English words they happen to be familiar with constitute the English lexicon. There was no such danger before since the number of English loans or English vocabulary items known to an average Pole was much lower. Secondly, such a false assumption might result in a later disappointment when many totally unknown English lexical items are to be introduced. Thirdly, it seems that it would be quite rewarding to learn English lexical items at random, which means, according to different semantic fields that are under discussion and to discover some similarities due to the impact of English on Polish. Marginally, it should be noted that even advanced students who attend the linguistic seminar at the Institute of English Philology, Jagiellonian University, when asked to enumerate English borrowings present in Polish were able to mention very few anglicisms and, what is even worse, they erroneously attributed English descent to some words, e.g. krokodyl (crocodile), korekta (proof-reading), radio, paszport (passport), dyrektor (director), etc. Therefore in the case of a certain discrepancy in the phonetic, graphic, and possibly the morphological forms between the English word and the English loan, like e.g. the English lexeme budget rendered as budČet [budžt] in Polish, it is worth drawing the students’ attention to their correlation. This would be the case of ‘the meeting point’ of two types of language contact resulting in the introduction of borrowings and one particular type of language contact present in the process of foreign/second language acquisition. In conclusion, we would like to repeat once again the statement presented at the beginning of the paper: there is a correlation between language contact, in the traditional sense of the word, as well as foreign and second language acquisition which should be taken advantage of. Hopefully, one such possibility has been convincingly described in reference to lexicon. However, considering the present state of language relations, we are very much against previously expressed views stressing the importance of introducing in our case anglicisms to Polish learners of English at the very first stage of teaching English. We only maintain the importance of drawing attention to one type of language contact, that is the influence of English on Polish, during another type of contact, that is when teaching English as a foreign or second language to Poles.

9612Arabski.indb 53

13/01/2006 08:43:23

54

Cross-linguistic Influences in the Second Language Lexicon

Notes 1. According to some linguists dealing with the theory of borrowing, the term bilingualism is interpreted in a very general way. Occasionally, it is even claimed that in order to introduce a loan it is sufficient to know a couple of words and thus to be called a bilingual speaker. 2. It has to be noted that Rusiecki is aware of the dangers involved: (a) the learner may wrongly assume that the whole of the vocabulary of the new language contains cognates; (b) the presence of false cognates; cf. some remarks in the discussion to follow. 3. It is worth stressing that false anglicisms occurring in Polish are relatively often confused with English words even by the Poles who have a very good command of English.

References Bantas, A. (1977) A bird’s eye-view on English influences upon the Romanian lexis. Studia Anglica Posnaniensia 9, 119–33. Bloomfield, L. (1933) Language. New York: Holt, Rinehart & Winston. Buttler, D., Kurkowska, H. and Satkiewicz, H. (1973) Kultura j¿zyka. Warszawa: PWN. (COD) The Concise Oxford Dictionary of Current English (1995) (9th edn). Oxford: Oxford University Press. Cienkowski, W. (1964) Ogólne zaÙoČenia metodologiczne badania zapoČyczeÚ leksykalnych. Poradnik J¿zykowy, 417-29. Deroy, L. (1956) L’imprunt linguistique. Paris. Filipovi², K. (ed.) (1982) The English Element in European Languages. (Vol. II). Zagreb: University of Zagreb. Fisiak, J. (1961) ZapoČyczenia angielskie w j¿zyku polskim: Analiza interpretacji leksykalnej. PhD thesis, University of RódĊ. Haugen, E. (1950) The analysis of linguistic borrowing. Language 26, 210–31. Hockett, Ch. F. (1958) A Course in Modern Linguistics. New York: Macmillan. Kuroczycki, T. and Rzepka, W.K. (1979) ZapoČyczenia leksykalne z j¿zyka rosyjskiego we wspóÙczesnej polszczyĊnie pisanej. Studia Rossica Posnaniensia 10, 107–15. MaÚczak-Wohlfeld, E. (1994) Leksykon w procesie nauczania j¿zyka angielskiego. J¿zyki Obce w Szkole 3, 206–8. MaÚczak-Wohlfeld, E. (1995) Should we teach English lexicon? In A. NiČegorodcew and A. Kosiarz-Stolarska (eds) Second and Foreign Language Acquisition (pp. 43– 48). Kraków: Universitas. Nemser, W. (1991) Language contact and foreign language acquisition. In V. Ivir and D. Kalogjera (eds) Languages in Contact and Contrast (pp. 345–64). Berlin: Mouton de Gruyter. Rusiecki, J. (1980) Latent bilingualism. Papers and Studies in Contrastive Linguistices 12, 79-98. Sapir, E. (1921) Language. An Introduction to the Study of Speech. New York: Harcourt, Brace & World. (SWJP) SÙownik wspóÙczesnego j¿zyka polskiego (1996) Warszawa. (SWO) SÙownik wyrazów obcych. Wydanie nowe (1999) Warszawa. Weinreich, U. (1970) Languages in Contact. Findings and Problems (7th edn). New York: Linguistic Circle of New York.

9612Arabski.indb 54

13/01/2006 08:43:24

Part 2

Language Contact Observed

9612Arabski.indb 55

13/01/2006 08:43:25

9612Arabski.indb 56

13/01/2006 08:43:26

Chapter 6

Genre: Language Contact and Culture Transfer ANDRZEJ ŁYDA Introduction This chapter on Polish-English language contact takes its inspiration from an essay by Piotr Kuhiwaczak (1999) on the political dimension of language identification. In the article in which he analyses a rare but not unprecedented situation of a single language splitting or rather said to be splitting into two varieties, Kuhiwczak discusses the role of political mechanisms involved in the process of shaping national and linguistic awareness, which some would rather recognise as a fallacy of national and linguistic distinctiveness. Kuhiwczak’s ‘Translation and language games in the Balkans’ reveals how hostility and animosity, when played upon skilfully by institutionalised entities, can lead to the formation of conception or misconception of the language that we speak: […] Croatia and Serbia manipulate commercial texts to enhance linguistic, political and cultural differences. This happens even at the most trivial level of information on ordinary packaging. The Balkan states have succeeded in involving even multinational companies in producing labels which reflect their constructed identity. Common products such as washing powders, sweets, toys and other consumer goods, which normally carried information in Serbo-Croatian as well as many other languages provide separate Serbian and Croatian entries. (Kuhiwczak, 1999: 222–3) This is illustrated by the following text found with McDonald’s toys: SRPSKI – Igra²ka je testirana za uzrast dece od 3 god i vise. Pažnja: može sadrzati sitnije delove pa se ne preopu²uje za decu spod 3 godine. (This toy has been safety tested for children aged 3 and over. Caution: it may contain small parts and is not intended for children under 3.) 57

9612Arabski.indb 57

13/01/2006 08:43:26

58

Cross-linguistic Influences in the Second Language Lexicon

HRVATSKI – Ova igra²ka je namijena za djecu od 3 i više godina. Upozorenje: Može sadržavati manje dijelove i nije namijenjena za djecu ïladu od 3 godine. Molimo sa²uvajte ovu obavijest. (This toy is intended for children aged 3 and over. Caution: it may contain small parts and is not intended for children under 3. Please retain information for reference). (1999: 223) It does not require any profound studies to notice that the two texts representing allegedly two languages differ inasmuch as their author took the trouble to select synonymous expressions for the same concepts, syntactic modification being of no significance for the message itself. Yet the very fact of placing the semantically similar and functionally identical texts next to each other and labelling them Serbian and Croatian respectively carries the presupposition of their being two separate languages.

Genre as an Institution However, it is not so much the socio-political manipulation which aims at creating the illusion of the irreconcilable duality of the previously homogeneous linguistic organism that I shall focus on. What follows from examples such as the one above is that the effectiveness of the political manœuvres in the Balkans depends on the same factors that in other cultures, ethnic and linguistic, bring about changes that do not lead to splitting but rather to uniting. The case in point is that the language split in the Balkans is being effected through texts of institutional character, in an institutional setting and on a global scale. Institutional as performed by big international companies and global as involving communication forms of great currency, printed in a multitude of copies and read by millions of recipients. As such the linguistic engineering is no longer symbolic but produces its material tangible objects exerting a strong influence on the awareness of the language communities and other communities that become convinced of the separateness of the languages and hence the nations. It is not then a coincidence that Serbian and Croatian can be found on safety instructions, product information labels, city transport tickets, etc. All of them are popular text types and just because of their circulation and readership they are perfect material for ideological manipulation. They are easily recognisable due to their form, structuring, potential content and predominantly their communicative purpose as they exemplify various discourse types, known as genres.

9612Arabski.indb 58

13/01/2006 08:43:27

Genre: Language Contact and Culture Transfer

59

In what follows I assume that genre is a conventionalised form of organisation of thinking and speaking about the reality, shared by language community speakers, conventionalised but not petrified since: Genres can come and go, or change, being cultural constructs which vary with the times, with fashion, and with ideological movements within society. […] Genres have […] change[d] in terms of the register invoked, among other changes, but the genre labels stay the same, since they are descriptors of socially constituted, functional categories of text. (Lee, 2001: 47) With the parallel development of the world and emergence of new communicative tasks genres undergo change. But the question that remains concerns the nature of the modifications. Do the new communicative tasks require that genres change? Do the changes consist in internal self-induced rearrangement of the constitutive elements of a particular genre or is it that genres are borrowed and adopted from other language communities? The literature pertaining to the study of language contact and borrowings is enormous. Equally exhaustive is the list of works on Polish-English language contact and loans. Yet in no work have I been able to find a category of genre loans, although the phenomenon has a long tradition in literature, with e.g. 13th century Sicilian sonnet introduced to England by Wyat and to Poland by Kochanowski three centuries later. A partial explanation is provided by the very confusing term genre and the difficulties in determining its boundaries and value. As Swales (1990) observes, the concept of genre has maintained a central position in folklore studies and literary studies, and not so central in linguistics and rhetoric. First, this should not be surprising given the long tradition of recognising sentence as the ultimate form of linguistic analysis. Secondly, and more importantly, genre is not regarded as a unit of the language system but as Swales (1990: 45–55) puts it they form (1) a class of communicative events, (2) with some shared recognisable communicative purpose, (3) with more or less prototypical exemplars, (4) constrained in terms of content, positioning and form, and (5) usually named. To this it should be added that genres are not fixed but dynamic in their form and function. Buckingham (1993) argues that genre is not simply given the culture, rather, it is in a constant process of negotiation and change. What are then the factors that initiate or contribute to this process of negotiation and change? It seems that as it is the case with other social constructs, the source of the alterations should be sought in the impact of social, economic and technological factors. Genres interact with the media, the unquestioned holder of power and the modern definers of values and ideologies.

9612Arabski.indb 59

13/01/2006 08:43:27

60

Cross-linguistic Influences in the Second Language Lexicon

Genres in Transition: The Case of Poland The values and ideologies of contemporary Poland have been constantly changing for the last 15 years. It is also its social constructs, its modes of communication, and its genres that have assumed a different shape. The period of 15 years coincides with the ongoing process of intensive cultural and linguistic contact with Anglo-Saxon culture, mainly as a result of the global role of English. Research into Polish-English language contact shows how intensive the increase in lexical borrowing has been with several new terms entering the Polish lexical stock each week. Frequent also are borrowings into the morphological layer; minor changes are observed in the syntactic component and in the phonological one (Markowski, 1992). However, if the description of the outcomes of English-Polish language contact is ever to be complete, it is essential to include the phenomenon of importation of larger and more abstract communicative entities of genre. Addressing the questions raised above concerning the mechanism of genre formation once the prevalent conventions are revised, we may first turn to Todorov 1976: 161, after Swales (1990) for whom, ‘a new genre is always the transformation of one or several old genres: by inversion, by displacement, by combination’. Indeed all these transformations processes can be identified while studying the ‘new’ genres that enriched the paradigm of Polish discourse forms after the introduction of a new political system, e.g. the complete replacement of Polish Čyciorys with CV, the gradual development of podanie into motivational letter, and the emergence of job interviews (Ostaszewska, 2000, 2004). However Todorov’s inventory is not exhaustive as no mention is made of genres that are completely new constructs of no affinity with previously existing ones. In the case of PolishEnglish contact, i.e. a situation of linguistic contact between a ‘minor’ culture and the shaping power of a dominant one, we have witnessed the emergence of completely unprecedented forms that filled the sociolinguistic vacuum, the existence of which started to be realised only with the transformation of the socioeconomic system and the system of goals and values. Just one example to illustrate this claim is mission statement, a generic form that appeared only in the late 1990s. Sometimes the modification of an existing genre is subtler and does not affect the structural characteristics of its textualisation but is related to its function. An interesting example of this kind is a genre so often referred to in EFL literature, i.e. horoscope. Horoscopes can be regarded as belonging to the group that Harweg (1980, after Duszak 1998) calls non-obtrusive texts (texts that we wait for, as contrasted with obtrusive ones (texts that wait for us)). Non-obtrusive texts are characterised as those which are determined

9612Arabski.indb 60

13/01/2006 08:43:28

Genre: Language Contact and Culture Transfer

61

by certain expectations of the addressees. It is the addressees who wait for a text and that is why such a text does not need to attract their attention. These expectations, a crucial aspect of horoscopes, are related to their dominant function, which, considering the subject matter and the style of presentation, could be characterised as phatic. The essence of phaticness lies among others in an attempt to maintain non-hostile, if not friendly, communication, a prerequisite of which is to appear trustworthy and honest in the eyes of the recipient. This can be accomplished by a number of linguistic means ranging from vagueness to the projection of possible worlds. At the same time horoscopes constitute ideal material for advertisers. In recent years the influence of the media, especially British popular magazines like Cosmopolitan or Vogue, manifested in Polish editions or Polish translation of the original texts, has been a major factor in the emergence of a new horoscope genre in Polish. The novelty of the genre consists in a transformation of the basic phatic function into a conative one. This has become possible by an almost imperceptible conversion of horoscopes into advertisements, by careful selection of lexis related directly to the text of advertisements placed next to the horoscope or by suggesting solutions offered, ‘somehow accidentally’ in the adjacent advertisements, e.g. A party at the end of the month puts a sparkle in your eye or Don’t miss out this golden opportunity next to a picture presenting the Chinese New Year fireworks with golden sparkles and advertising Hong Kong (see Figure 6.1). Polish magazines have followed suit (see Figure 6.2). It is interesting to note that all the new genres of the present-day Poland have been created or functionally adapted in an institutionalised context (public offices, companies, media). This is exactly what is predicted by Bargiela-Chiappini and Nickerson’s (2002: 39) model of the social context for the operation of multinational corporations viewed as: a series of four hierarchical levels, consisting of the multinational corporation as a discrete social entity at one level […], followed by a set of (typified) situations that occur and re-occur as the corporation pursues its institutional goals. At the next level, within each situation, the corporation evolves a series of typified communication practices, or genres, in order to accomplish the necessary (communicative) action, and finally, these genres are realized as individual textualizations […] Considering the inevitable dynamism of institutional (social and economic) transformations, subsequent development of new generic forms within larger social groups as well as the unifying function of genres on a global scale through their importation from one culture to another, we may ask a question whether SLA research, focusing on the individual,

9612Arabski.indb 61

13/01/2006 08:43:29

62

Cross-linguistic Influences in the Second Language Lexicon

Figure 6.1 Intertextual reference: A horoscope and an advertisement lexical links (English) Cosmopolitan April 2003.

9612Arabski.indb 62

13/01/2006 08:43:30

Genre: Language Contact and Culture Transfer

63

Figure 6.2 Intertextual reference: A horoscope and an advertisement lexical links (Polish). rather than on the speech community (Sankoff, 2001), should pay more attention to the acquisition of competence in the use of L2 generic forms, the influence of new genres on L1 and the interaction between similar, though distinct, generic formations in L1 and L2 speech communities. In the context of media studies Buckingham (1993) observes ‘that children progressively acquire (or at least come to use) a discourse of genre as they mature – that is, a set of terms which facilitate the process of categorisation, or at least make certain kinds of categorisation possible’ (my bold type). If genre acquisition is possible in this domain or visual arts, can it be ever demonstrated that it is equally possible in the case of language acquisition? The target in SLA is a different linguistic system and the learner’s major goal is perfect mastery of that system measured by assessment of the learner’s interlanguage and the target language (Wojtaszek, 1998). However, it has been said above that genre is not a unit of the language

9612Arabski.indb 63

13/01/2006 08:43:33

64

Cross-linguistic Influences in the Second Language Lexicon

system but rather a set of recognisable dynamic conventions of thinking and speaking (also writing) in a language community. As such the problem of genre may seem to lie outside the area of language learning and as a consequence language teaching. To ignore genre in the context of SLA, however, would be a mistaken decision because it would be tantamount to rejecting a whole array of other factors involved in SLA and closely related to genre. As Kramsch (1998) points out, interlanguage is now seen as a social semiotic construct in which style, genre, and textuality all play a role in defining the mediating process of the acquisition of another semiotic code (my emphasis). What remains to be documented is just how significant the role is. References Bargiela-Chiappini, F. and Nickerson, C. (2002) Business discourse: Old debates, new horizons. IRAL 40, 273–86. Buckingham, D. (1993) Children Talking Television: The Making of Television Literacy. London: Falmer Press. Duszak, A. (1998) Tekst, dyskurs, komunikacja mi¿dzyj¿zykowa. Warszawa: Wydawnictwo Naukowe PWN. Harweg, R. (1980) Beginning a text. Discourse Process 3, 313–26. Kramsch, C. (1998) Language and Culture. Oxford: Oxford University Press. Kuhiwaczak, P. (1999) Translation and language games in the Balkans. In G. Anderman and M. Rogers (eds) Word, Text, Translation. Liber Amicorum for Peter Newmark (pp. 164–224). Clevedon: Multilingual Matters. Lee, D.Y.W. (2001) Genres, registers, text types, domains, and styles: Clarifying the concepts and navigating a path through the BNC jungle. Language and Technology 5 (3), 37-72. Markowski, A. (1992) Nowsze zapoČyczenia w polszczyĊnie. Poradnik j¿zykowy 3, 237–41. Ostaszewska, D. (2000) Gatunki mowy i ich ewolucja. Mowy pi¿kno wielorakie. Katowice: Wydawnictwo Uniwersytetu gl­skiego. Ostaszewska, D. (2004) Gatunki mowy i ich ewolucja. Tekst a gatunek. Katowice: Wydawnictwo Uniwersytetu gl­skiego. Sankoff, G. (2001) Linguistic outcomes of language contact. In P. Trudgill, J. Chambers and N. Schilling-Estes (eds) Handbook of Sociolinguistics. Oxford: Basil Blackwell. Swales, J. (1990) Genre Analysis. English in Academic and Research Settings. Cambridge: Cambridge University Press. Todorov, T. (1976) The origin of genres. New Literary History 8, 159–70. Wojtaszek, A. (1998) The process of foreign culture acquisition in present-day Poland. In J. Arabski (ed.) Studies in Foreign Language Learning and Teaching (pp. 49–61). Katowice: Wydawnictwo Uniwersytetu gl­skiego.

9612Arabski.indb 64

13/01/2006 08:43:34

Chapter 7

Is Cross-linguistic Influence a Factor in Advanced EFL Learners’ Use of Collocations? ´NIEWSKA JUSTYNA LES Introduction Phraseological studies centre around one particular linguistic phenomenon, namely, the existence, alongside grammatical rules and semantic restrictions, of a collocational level of restriction on how words can be combined. Positing a separate collocational level is justified by a certain amount of arbitrariness, or unpredictability, of some word combinations, which is seen as lying outside the semantic characteristics of a word. The often idiosyncratic character of collocational restrictions becomes especially visible when collocations are compared across languages. Word combinations in a language are usually seen as ranging from free to fixed ones (although it is by no means easy to provide specific criteria according to which this level of fixity can be specified), with collocations being an intermediate category. The collocational level of restriction on word combinations has recently been attracting more attention in SLA studies; however, in TEFL methodology the interest in word combinations is connected mostly with their pragmatic/discoursal role (Lewis, 1993; Nattinger & DeCarrico, 1992), since many re-occurring word combinations take on a pragmatically specialised meaning which is a property of the word combination as a whole. The fact that certain words tend to co-occur is often explained in psycholinguistic terms, with reference to the fact that some word combinations may be processed as wholes. A distinction between two mechanisms of language production has been made, among others, by Pawley and Syder (1983) and by Sinclair (1991), who calls them the ‘open-choice principle’ and the ‘idiom principle’. The nature of language production is thus accounted for in terms of syntactic and semantic rules, on the one hand, and the retrieval of chunks, which can also be broken down into 65

9612Arabski.indb 65

13/01/2006 08:43:35

66

Cross-linguistic Influences in the Second Language Lexicon

component parts if there is need to do so, on the other. The role of chunks is explained by the economics of language production: they represent a compromise between increased demands on storage space and gains in the speed of production and ease of comprehension (Nation, 2001: 319). The role of chunks is augmented by the fact that they often have a specialized pragmatic/ situational meaning which resides in the chunk as a whole rather than in its component parts. With respect to chunking vs. rule formation, collocations seem to be a borderline case, which can arguably be a source of difficulty in L2 learning and use. The mental lexica of a foreign language learner are bound to be connected at least to some extent (Singleton, 1997). In the mental lexicon, numerous links must hold between the nodes; it is likely that while fixed multi-word units involve very strong links, the use of conventional collocations involves the activation of links of variable strength. Therefore, the former category may be easier to learn and use than the latter for a second language learner. If selectional restrictions are best learned when they are strong, the L2 learners would tend to either operate according to the openchoice principle, or to use the word combinations which are fixed. The subtler kinds of collocational restriction are likely to be more elusive, and more liable to cross-linguistic influence. The available findings from research in the area of the use of collocations by advanced learners of English indicate that collocations are indeed a problem area and a marker of non-nativeness (Arnaud & Savignon, 1997; Bahns & Eldaw, 1993; Channell, 1980; Granger, 1998; Howarth, 1996, 1998a, 1998b; Källkvist, 1998; Lorenz, 1998, 1999; Waller, 1993); they also support the view that partly restricted collocations, which cannot be learned as unitary items nor produced entirely by rule, are in many ways the most problematic ones in L2 acquisition and use, and most likely to reflect crosslinguistic influence. Advanced learners have been found to differ from native speakers in that they produce deviant collocations (‘visible’ errors) as well as in other, subtler ways, which become apparent in studies of the patterns of overuse or underuse of specific types of collocations. Such ‘hidden’ distributional differences become apparent in corpus analyses of word combinations in learner texts. Although L1 influence has been tentatively put forward by some researchers as the potential reason for some of the characteristics of learner language, very little is known about the extent of cross-linguistic influence on the use of collocations, either because of the difficulty in identifying L1induced errors and distributional differences, or because the L1 background was not controlled (as in e.g. Howarth, 1996). Biskup (1992), in one of the very few studies which deal directly with L1 influence on the learners’ use

9612Arabski.indb 66

13/01/2006 08:43:36

Is Cross-linguistic Influence a Factor?

67

of L2 collocations, found cross-linguistic influence to be an important factor, although it might have been augmented by the translation task used for the purpose of the investigation. The collocational competence of foreign language learners can be seen as either essentially the same as that of (monolingual) native speakers, or qualitatively different at any level of proficiency because of the existence of the L1 in the mind, which makes it impossible to use the L2 in a fully ‘monolingual’ mode. The findings of Granger (1998) strongly support the latter possibility, since in her study advanced learners were found to use predominantly those collocations which are congruent with those of the L1. This paper presents some findings from a larger project investigating the collocational aspects of advanced EFL use (Leïniewska, 2003), in an attempt to establish the extent of cross-linguistic influence on the use of collocations by Polish learners of English at an advanced level. The first question to be asked is whether the use of collocations by learners differs from that of native speakers; if so, what the differences are and whether they reflect the influence of the L1.

Method The investigation was carried out by means of a set of two tests, both of which involved adjective intensification. Test 1 elicited the subjects’ collocational choices by asking them to complete gapped sentences, while Test 2 provided the subjects with a choice of collocates, eliciting the subjects’ acceptability and saliency judgements concerning the given collocations. Test 1 consisted of 16 sentences, all of which contained an adjective premodified by the adverb very. Each sentence was printed twice, and in the repeated sentence the adverb very was replaced with a blank space, e.g.: 1. The report was very critical of the railway’s poor safety record. The report was …………… critical of the railway’s poor safety record. The instructions asked the subjects to replace the word ‘very’ with a different word which would intensify the meaning of the underlined adjective, and to try to use a different word in each sentence. Test 2 contained the same sentences as Test 1, in the same order. This time, each sentence was printed on the page only once, with a blank space preceding the adjective. Below each sentence, a choice of 10 more or less suitable intensifiers of varying levels of restrictedness was given, e.g.: 1. The report was …………… critical of the railway’s poor safety record. acutely deeply extremely glaringly highly immensely profusely severely sharply strongly

9612Arabski.indb 67

13/01/2006 08:43:36

68

Cross-linguistic Influences in the Second Language Lexicon

The tests were administered to both advanced learners of English (Group 1) and to native speakers of English (Group 2). The subjects in both groups had a comparable educational background – they were all students of languages at university level. Group 1 consisted of 113 students of English at the English Department of the Jagiellonian University in Krakow, while Group 2 consisted of 61 students of languages at UK universities. The subjects were also roughly comparable with respect to age (most of them in their early twenties). The level of proficiency in English of Group 1 was roughly comparable with the level required to pass the Cambridge Advanced English examination, or higher. While the number of years of studying English and the specific educational experience in this respect were expected to vary among the subjects, the group were relatively homogeneous with respect to their current level of advancement. All of the subjects could be classified as learners of English as a foreign (rather than second) language, in that their experience of learning English had been limited to mostly educational settings, and none of them came from bilingual families or had extensive experience of functioning in an English language environment. Therefore, their L2 acquisition could not be termed naturalistic. The procedure of administering the tests was the same for all the subjects. The tests were completed by the participants in a classroom setting. After copies of Test 1 were collected, copies of Test 2 were handed out to the participants. The time limit was 10 minutes for the first test and 5 minutes for the second one.

Results and Discussion In general, the Polish group found it more difficult to provide suitable intensifiers, as evidenced by the fact that the Polish group left altogether 12.8 % empty gaps in Test 1, while the English group left only 0.05% of the test items without an answer. The ability to offer different intensifiers in each of the 16 test sentences was also different in the two groups: the average number of different lexical items given by one subject was 10.43 in the Polish group and 13.46 in the English group. The lexical variation within groups was not directly comparable since the size of the groups was not the same, but it was mathematically interpolated so as to enable a comparison. The native speaker group produced on the whole 150 different lexical items. If the Polish group produced 172 different lexical items, however, and if it had consisted of the same number of subjects as the native speaker group, it would have produced 127 different answers.

9612Arabski.indb 68

13/01/2006 08:43:37

Is Cross-linguistic Influence a Factor?

69

Apart from the quantity of the different intensifier types produced by the two groups, the distribution of these intensifiers in terms of their frequency is of importance for the present analysis. The most striking characteristic of this distribution is that a majority of the intensifier types occurred very seldom – only once or twice. In the Polish group, 106 answer types (over 60%) occurred only once or twice. Figure 7.1 presents the distribution of Test 1 answers for the Polish group and the native speaker group. Those answer types which individually make up less than 2% of all tokens are labelled ‘rare answers’. As can be seen in Figure 7.1, such rare answers account for 46.3% of all tokens in the Polish group and for 49.5% of all tokens in the native speaker group. Rare answers include 160 types (93% of all types) in the Polish group and 137 types (91% of all types) in the native speaker group. The pie charts presented in Figure 7.1 also include subsets of rare answers which are single occurrences – they make up 5.3% and 6.5% of all tokens in Group 1 and Group 2, respectively. The distribution of answer types is therefore very similar for the two groups with respect to the relative number of rare answer types and the percentage of tokens that they account for. Another similarity between the two groups is the number of types that individually account for more than 2% of all tokens, that is, do not fall into the category of rare answers: 11 in the case of the Polish group and 13 in the case of the native speakers. Also, another shared characteristic of the two groups is the fact that the distribution is dominated by one intensifier (extremely), though this tendency is stronger in the case of the Polish group. Overall, the sets of answers of the two groups (172 types in Group 1 and 150 types in Group 2) have 89 types in common. This means that just over a half (52%) of the intensifiers used by the Polish group appear in the native speaker data. However, most of the discrepancies occur in the groups of intensifiers which were used seldom by either of the groups. Most of the answers which are frequent in the Polish group appear also in the answers of the native speakers. Similarly, the most frequent answers of the native speakers appear in the learner data. The extent of cross-linguistic influence Very few of the answers given by Group 1 in Test 1 can be attributed with any degree of certainty to the influence of Polish, for example crazily popular (Polish szalenie popularny), even too obvious (aČ nazbyt oczywiste), far exaggerated (dalece przesadzony), hellishly talented (piekielnie uzdolniony). Except for these rare cases, neither the deviant forms, the non-existent words, nor the inappropriate extensions of the collocability range made by the Polish learners display the influence of Polish. Most of the unsuitable

9612Arabski.indb 69

13/01/2006 08:43:38

70

Cross-linguistic Influences in the Second Language Lexicon

highly 7.2% extremely 19.0%

single 5.3%

really 5.4% quite 4.1% totally 3.5% terribly 3.4 deeply 2.8% much 2.3% so 2.1% pretty 2.0% enormously 2.0%

rare answers 46.3%

highly 8.8% extremely 11.0%

single 6.5%

hugely 4.2% terribly 4.2% really 3.3% greatly 2.7% deeply 2.6% immensely 2.6% incredibly 2.5% overly 2.3% fiercely 2.2% freezing 2.2% blatantly 2.1%

rare answers 49.5% Figure 7.1 The distribution of intensifier types in the answers to Test 1 given by Group 1 (upper graph) and Group 2 (lower graph) answers produced by the Polish group are items such as much, so, indeed, rather, more than, used by the learners for lack of a proper booster which would conform to the test instructions, or items such as freezingly (in freezingly cold), an incorrect form derived from the adjective freezing in order to obtain a -ly adverb. Another, very small group of unsuitable answers are

9612Arabski.indb 70

13/01/2006 08:43:39

Is Cross-linguistic Influence a Factor?

71

those which are clearly semantically inappropriate (such as densely inaccurate, grudgingly critical). None of the clearly unsuitable answers of the Polish group displays the influence of the L1. Both groups produced some ‘creative’ combinations (e.g. astoundingly offensive, mind-numbingly cold, aggressively protective in the native speaker group; profusely talented, overwhelmingly boring, annoyingly offensive in the Polish group). While the Polish learners are generally less successful than native speakers with such creative combinations, they do not seem to be drawing on their L1 for help with the task of intensifying adjectives. Instead, their attempts at creating an intensifier are often based on forming adverbs of manner from synonyms of the adjectives to be intensified (as in e.g. jadedly tired, rudely offensive). Another way in which cross-linguistic influence could manifest itself is the learners’ preference for those collocations which are congruent in English and in Polish. There is some evidence for such a preference. For example, the fact that in Test 2, many learners chose deeply as the intensifier for religious (similarly to the native speakers) may have been augmented by the fact that, in Polish, the collocations gÙ¿boko religijny is a very frequent one. The collocation wildly inaccurate, which does not have a congruent Polish equivalent, was selected in Test 2 by six native speakers but by none of the learners. In test item 4 (significant) the top choices of the native speakers were hugely, extremely, and enormously, while the learners strongly preferred particularly significant (37 uses, while only two native speakers selected this collocation), which could be related to the fact that the congruent collocation szczególnie waČny / istotny is common in Polish. However, this tendency was not observed in the majority of the test items. On the whole, since the learners use both congruent and incongruent collocations, and do not make errors clearly attributable to L1 transfer, it seems that the cross-linguistic factor does not play a major role in the learners’ use of collocations in the L2. Another factor which could interplay with the influence of the L1 on the learners’ collocational choices is the possibility that the learners may avoid forms which are L1–L2 congruent, even if the L2 forms in question are acceptable, because they are wary of false friends. The present study, however, does not provide sufficient evidence to draw definitive conclusions in this respect. ‘Restricted’ and ‘general’ intensifiers While the learners do not rely on the L1 in their use of collocations, there is another characteristic which makes their answers different from those of the native speakers. It has to do with the use of intensifiers which belong to two different groups:

9612Arabski.indb 71

13/01/2006 08:43:41

72

Cross-linguistic Influences in the Second Language Lexicon

(1) the ‘general’ category, which includes boosters such as highly, extremely, largely, really, immensely or deeply; (2) the ‘restricted’ category, which includes intensifiers such as fiercely, blatantly, glaringly. Such intensifiers are to a greater extent restricted to a certain group of adjectives. Most similarities between the groups result from the use of similar general intensifiers, while most discrepancies occur because of the marked underuse by the learners of intensifiers from the restricted category. The most striking difference between the two groups results from the fact that the very popular intensifiers glaringly and blatantly in the native speaker group are not used at all by the Polish group. The most interesting finding which emerges from the comparison of the answers to the two tests is that the Polish group chose many restricted intensifiers in Test 2, some of which were not used at all in Test 1. A good illustration of this tendency is provided by the answers given to the test item 14 (obvious), presented in Table 7.1. Of the three restricted collocates of obvious – blatantly, glaringly, and blindingly – none was used by the Polish subjects in Test 1. However, all three were selected by some learners in Test 2, most notably the adverb blatantly (11 occurrences). The native speakers used all three intensifiers in both tests, but they also chose them more often in Test 2. Another example of this tendency is provided by the use of the intensifier grossly. The intensifier was offered in Test 2 as one of the options in three sentences (test item 2 – inaccurate, test item 9 – offensive, and test item 15 – exaggerated). The uses of grossly by both groups are presented in Table 7.2. In Test 1, the native speakers used grossly once in test item 2 (inaccurate), and three times in test item 15 (exaggerated). It was not used in any other sentences. The Polish learners used grossly three times in test item 15, and they did not use it in any other sentences of the test. In Test 2, both groups use it more often in all three sentences. In the native speaker group, Table 7.1 The uses of blatantly, glaringly, and blindingly in test item 14 (obvious) Group 1 (Polish students)

Group 2 (native speakers)

Intensifier

Test 1

Test 2

Test 1

Test 2

blatantly

–

11

19

25

glaringly

–

2

10

21

blindingly

–

1

3

12

9612Arabski.indb 72

13/01/2006 08:43:42

73

Is Cross-linguistic Influence a Factor?

Table 7.2 The use of grossly by both groups Group 1 (Polish students) Group 2 (native speakers) Test item

Test 1

Test 2

Test 1

Test 2

2 – Inaccurate

–

12

1

32

9 – Offensive

–

8

–

6

15 – Exaggerated

3

12

3

16

grossly tops the lists of intensifiers used with inaccurate and exaggerated. It is worth noting that the collocation grossly inaccurate was produced by only one native speaker in Test 1, while 32 native speakers (more than half of the group) chose it as the most suitable collocation in the corresponding sentence of Test 2. One more example of the above tendency will be given here – the use of fiercely by the two groups (Table 7.3). Apart from the uses with protective and competitive, the native speakers used fiercely also with disappointed (one occurrence) and religious (three occurrences). Polish learners also produced one more collocation with fiercely (fiercely cold). Both groups selected fiercely in Test 2 in the sentences with protective and competitive more often than they did in Test 1, and the native speaker group displayed a very strong preference for this intensifier (the 46 choices of fiercely in test item 12 account for more than 75% of all the answers for this test item). The above results point to several very interesting characteristics of the subjects’ collocational competence. Most importantly, it appears that the learners display the same tendency as native speakers to choose more restricted collocations in Test 2. When asked to make a judgement on which collocations are most suitable, the native speakers strongly favour restricted collocations, and the learners display a similar preference, only Table 7.3 The uses of fiercely by the two groups in test items 12 and 16 Group 1 (Polish students) Group 2 (native speakers) Test item

Test 1

Test 2

Test 1

Test 2

12 – Protective

1

28

8

46

16 – Competitive

3

36

9

25

9612Arabski.indb 73

13/01/2006 08:43:43

74

Cross-linguistic Influences in the Second Language Lexicon

weaker. This means that language users in general seem to regard restricted collocations as more appropriate, felicitous, stylistically ‘stronger’. Such restricted collocations are not, however, automatised enough to occur to a large number of respondents while completing Test 1, and a wide choice of different intensifiers is offered by the subjects. Since the learners chose many restricted intensifiers in Test 2, even though they did not offer them in Test 1, it is unlikely that the collocational differences between the two groups are only the result of the fact that the learners do not notice transparent collocations in the input they are exposed to. Rather, the lower number of intensifiers from the restricted category in Test 1 in the learner group is probably the result of the same mechanism that made the Polish students leave so many blank gaps, that is, of the difficulty they experience with recalling lexical items. It seems, therefore, that collocational aspects of word knowledge may be particularly liable to differences in terms of receptive and productive knowledge. Another point which needs to be made here is that the collocational competence of Polish learners cannot be described as ‘misguided’ (Granger, 1998). A misguided notion of salience is the explanation given by Granger for the fact that the advanced learners in her study (of French L1) assumed a wider range collocability than the native speakers for some restricted intensifiers. The subjects in Granger’s study not only failed to recognise such collocations as, for example, bitterly cold, but also marked combinations such as bitterly aware and bitterly miserable as acceptable. Such a tendency would show in Test 1 if, for example, the Polish learners had used grossly in other sentences than the native speakers. This would indicate that the learners know the intensifier itself, but are not familiar with the restrictions on its use. This is not the case, however; rather than using restricted intensifiers in inappropriate contexts, the Polish learners display a tendency to use fewer restricted collocations. Their notion of salience is therefore not so much misguided as simply weaker than that of native speakers.

Conclusions To sum up, the collocational competence of advanced learners in this study displays the following characteristics: (1) The learners find it much more difficult than do the native speakers to recall suitable vocabulary items. (2) The learners use fewer restricted collocations than the native speakers, and tend to fall back on the use of ‘general-purpose’ intensifiers.

9612Arabski.indb 74

13/01/2006 08:43:44

Is Cross-linguistic Influence a Factor?

75

(3) The learners generally manage to avoid errors in the use of collocations. (4) The learners’ use of collocations in English gives very little evidence of the cross-linguistic factor at work. The first two characteristics indicate that the collocational links in the mental lexicons of the learners are weaker than those of native speakers. This explains why it is more difficult for the learners to recall vocabulary items which can be used as intensifiers. When they need to intensify an adjective, they do not have an option which ‘springs readily to mind’ (Benson et al., 1986: 253) and resort to using a general-purpose intensifier. However, as evidenced by the results for Test 2, some of the links are strong enough to enable the recognition of suitable intensifiers. When the learners recognise an intensifier which is more restricted, that is, typically used with a given adjective, they show a preference for this item. Moreover, the general qualitative tendencies in the use of intensifiers are similar in the two groups compared in this study. They are visible in the distribution of the intensifiers in terms of their frequency of occurrence in the data and the number of intensifier types occurring with a certain frequency. These findings indicate that the collocational competence of the learners is developing in the direction of native speakers’ competence. It is often suspected that the learners tend to operate more according to what Sinclair calls the ‘open-choice principle’. Such a characteristic of learners would mean that they are more likely to use words as the ‘building bricks’ of language rather than to rely on ready-made combinations. In the case of collocations, this tendency could be manifested in the use of semantically appropriate, but not collocationally restricted collocates. This observation finds strong support in the results of this study, as the Polish subjects show a weaker preference than native speakers for the restricted intensifiers, choosing either ‘general-purpose’ intensifiers, such as extremely, or other semantically appropriate but not restricted ones. All in all, the Polish subjects in this study do not seem to be prone to collocational error. This finding contrasts with the conclusions of many other studies (e.g. Howarth, 1996; Källkvist, 1998) in which the problematic nature of collocations in EFL acquisition is very much emphasised. This discrepancy can be explained in two ways: firstly, the term ‘advanced learner’ is used in many research studies with an apparently wide meaning; secondly, the term ‘collocational error’ can similarly be stretched to encompass most lexical errors in writing or speech. All errors in the use of lexis, except spelling errors and non-words formed by overgeneralising morphological rules, are only identifiable in the context of the sentence in

9612Arabski.indb 75

13/01/2006 08:43:44

76

Cross-linguistic Influences in the Second Language Lexicon

which they are produced, which makes it possible to say what the intended word is. Therefore most lexical errors can be classified as collocational in a very broad understanding of the term. References Arnaud, P.J.L. and Savignon, S.S. (1997) Rare words, complex lexical units and the advanced learner. In J. Coady and T. Huckin (eds) Second Language Vocabulary Acquisition: A Rationale for Pedagogy (pp. 157–173). Cambridge: Cambridge University Press. Bahns, J. and Eldaw, M. (1993) Should we teach EFL students collocations? System 21, 101–14. Benson, M., Benson, E. and Ilson, R. (1986) Lexicographic Description of English. Amsterdam and Philadelphia: John Benjamins. Biskup, D. (1992) L1 influence on learners’ renderings of English collocations: A Polish/English empirical study. In P.J.L. Arnaud and H. Béjoint (eds) Vocabulary and Applied Linguistics (pp. 85–93). Basingstoke and London: Macmillan. Channell, J. (1980) Applying semantic theory to vocabulary teaching. ELT Journal 35, 115–22. Cowie, A.P. (ed.) (1998) Phraseology: Theory, Analysis, and Applications. Oxford: Clarendon Press. Granger, S. (1998) Prefabricated patterns in advanced EFL writing: Collocations and formulae. In A.P. Cowie (ed.) Phraseology: Theory, Analysis, and Applications (pp. 145–60). Oxford: Clarendon Press. Howarth, P. (1996) Phraseology in English Academic Writing: Some Implications for Language Learning and Dictionary Making. Tübingen: Niemeyer. Howarth, P. (1998a) Phraseology and second language proficiency. Applied Linguistics 19, 24–44. Howarth, P. (1998b) The phraseology of learners’ academic writing. In A.P. Cowie (ed.) Phraseology: Theory, Analysis, and Applications. Oxford: Clarendon Press. Källkvist, M. (1998) Lexical infelicity in English: The case of nouns and verbs. In K. Haastrup and A. Viberg (eds) Perspectives on Lexical Acquisition in a Second Language (pp. 149–74). Lund: Lund University Press. Leïniewska, J. (2003) The collocational aspects of advanced EFL learners’ lexical competence. Unpublished PhD thesis. Kraków: Jagiellonian University. Lewis, M. (1993) The Lexical Approach: The State of ELT and a Way Forward. Hove: Language Teaching Publications. Lorenz, G. (1998) Overstatement in advanced learners’ writing: Stylistic aspects of adjective intensification. In S. Granger (ed.) Learner English on Computer. London and New York: Longman. Lorenz, G. (1999) Adjective Intensification – Learners versus Native Speakers: A Corpus Study of Argumentative Writing. Amsterdam/Atlanta, GA: Rodopi. Nation, I.S.P. (2001) Learning Vocabulary in Another Language. Cambridge: Cambridge University Press. Nattinger, J. and DeCarrico, J. (1992) Lexical Phrases and Language Teaching. Oxford: Oxford University Press. Pawley, A. and Syder, F.H. (1983) Two puzzles for linguistic theory: Nativelike selection and nativelike fluency. In J.C. Richards and R.W. Schmidt (eds) Language and Communication (pp. 191–226). London: Longman.

9612Arabski.indb 76

13/01/2006 08:43:45

Is Cross-linguistic Influence a Factor?

77

Sinclair, J. (1991) Corpus, Concordance, Collocation. Oxford: Oxford University Press. Singleton, D. (1997) Cross-linguistic aspects of the mental lexicon. In R. Hickey and S. Puppel (eds) Trends in Linguistics: Language History and Linguistic Modelling. Studies and Monographs 101. Berlin: Mouton de Gruyter. Waller, T. (1993) Characteristics of near-native proficiency in writing. In H. Ringbom (ed.) Near-Native Proficiency in English. Åbo: Åbo Akademi University.

9612Arabski.indb 77

13/01/2006 08:43:46

Chapter 8

International Terms and Profile Transfer: On Discussion KRYSTYNA WARCHAŁ Introduction ‘Every language is a self-contained system and, in a sense, no words or constructions of one language can have absolute equivalents in another,’ writes Wierzbicka (1991: 10) in Cross-Cultural Pragmatics. Still, there is a group of elements which appear more easily identifiable across languages in terms of form and meaning than others and whose formal and semantic convergence is evident enough to be appreciated by language users who are neither linguists nor polyglots. This is not to say that such international terms always display the same high degree of equivalence or are international to the same extent. Nevertheless, there are some, e.g. compromise (Eng.) / kompromis (Pl.) / compromiso (Sp.), negotiations (Eng.) / negocjacje (Pl.) / negoziati (It.) or discussion (Eng.) / dyskusja (Pl.) / discussione (It.), whose international character does not seem to raise doubts and which are expected not to pose problems to second language learners or translators on account of their deceptive formal similarity. The following remarks have emerged from a reflection on international terms for social situations, activities performed in these situations and their outcomes – to the exclusion of terminology of science and technology – inspired by, among others, Braun et al. (1990), Kolwa (2001) and Wierzbicka (1991). The aim of this paper is to suggest that in spite of a relatively high degree of interlingual equivalence between some such elements, there are still certain differences in the way the social situations are construed in respective cultures, and that these differences may occasion profile transfer between international terms of relatively high degree of equivalence. ‘Profile transfer’ will be used here to cover the area of cross-linguistic influence obtaining between systemic translation equivalents whose denotations (extensions), however closely overlapping, are not identical, that is in cases when not all the entities which can be correctly called by one of the terms will be referred to by the other by speakers of that language. Profile transfer will consist in the denotation of one of the terms being narrowed 78

9612Arabski.indb 78

13/01/2006 08:43:47

International Terms and Profile Transfer

79

down or extended as a result of both semantic proximity and discreteness of the other term. It does not account for morphological or syntactic inaccuracies that may result from differences in the grammatical properties of the units. Profile transfer between terms which are not only semantically but also formally related will be viewed as a particular instance of this process. The argument will focus on discussion in English and dyskusja in Polish, which has gained acceptance among Polish authors writing for home audiences as the title of a text unit corresponding to ‘Discussion’ in English academic papers. It will be hypothesised that this use of dyskusja – new to the Polish academic tradition, according to which such a section would be called ‘Omówienie wyników’, and, as I shall attempt to show, inconsistent with the current meaning of the Polish term – may be indicative of some changes in (or may have consequences for) the way dyskusja is construed.

International Terms and the Problem of Interlingual Correspondence The term ‘international element’ is used here in the narrower sense, to refer to lexemes, although in the broader sense it can refer to linguistic elements on other levels of organisation, such as, e.g. morphemes or idiomatic expressions. ‘International term’ is a synchronic category for lexical items which occur in many languages and whose form and meaning overlap to the extent that they can be identified as equivalents by speakers of only one language (Ma°kiewicz, 1984). Thus in calling an element ‘international’ no claim is made concerning its etymology or direction of borrowing. The study of international terms appears to have gained in importance in the last decades of the 20th century, with the European Union gradually expanding and with the idea of free movement of people, goods and services slowly taking shape. As Braun (1990: 15) observes, with the political borders playing minor role in the lives of people in Europe (Western Europe at that time), the limitations imposed by language barriers appear all the more severe and inconvenient. It seems that the search for a theoretical basis for the study of international terms undertaken by German scholars on the threshold of 1990s (Braun et al., 1990) was to a large extent inspired by the need to emphasise the common core, the European heritage of the EU nations and cultures, and to explicate the existing differences with a view to making international contacts more effective and less fraught with the risk of miscommunication.

9612Arabski.indb 79

13/01/2006 08:43:47

80

Cross-linguistic Influences in the Second Language Lexicon

The postulate of internationality There is no universal agreement on exactly how widespread a term should be to be considered international. The definition adopted here is that proposed by Ma°kiewicz (1984), who suggests that the necessary condition for an international term is the presence of corresponding lexemes in at least three languages representing three different branches of the same family and belonging to the same geographic and cultural area. Two additional criteria have been proposed to account for various degrees of internationality – first, the presence of corresponding lexical units in languages which are genetically unrelated but which belong to the same geographic and cultural area, and, second, the presence of corresponding lexical items in languages which are remote in origin, space and culture – but the necessary isogloss is formed by three non-cognate languages of the same family which are part of the same geographic and cultural scene (Ma°kiewicz, 1984). The postulate of formal correspondence Formal correspondence is established chiefly on the basis of the written form of a lexical unit, in particular the consonantal skeleton of its root (Ma°kiewicz, 1984). The degree of similarity between forms sufficient to call them international escapes definition – they should be similar enough for an average language user to identify as potentially equivalent. The postulate of semantic correspondence The problem of semantic correspondence is still more complex given that, first, one representation may have more than one meaning in the same system, and, second, that more often than not elements of different systems which are beyond doubt similar in form and meaning do differ semantically in some important respects. In the former case, a question arises as to whether the semantic correspondence between lexical units should pertain to all the meanings they have in their respective systems or, if not, to how many and which of the meanings. In the latter case, the difficulty lies in the permissible semantic change. For example, for a pair of verbs discriminate (Eng.) – 1. ‘distinguish, tell apart’, 2. ‘give unequal treatment’, and dyskryminowa° (Pl.), a claim of equivalence can be made only with regard to the second meaning of the English lexeme. Moreover, the correspondence is not complete, since in Polish the meaning is restricted to ‘victimise’, while in English the direction of unequal treatment is specified by means of syntactic devices – discriminate against or discriminate in favour of. To solve the definitional problem of semantic correspondence, Ma°kiewicz (1984) proposes that international terms are lexical units of different systems which meet the criteria of minimal isogloss and formal

9612Arabski.indb 80

13/01/2006 08:43:48

International Terms and Profile Transfer

81

similarity and which constitute a group of systemic translation equivalents. International terms proper are those which are equivalent with regard to all their meanings in respective systems, while those equivalent only in some of the meanings are partial international terms. Lexical units which meet the criteria of minimal isogloss and formal similarity but whose meanings differ to the extent that they are not systemic translation equivalents are pseudointernational terms. Lipczuk (1992) refers to this group of lexemes as tautonyms. International terms constitute an important part of the lexicon of science and technology. On the one hand, they facilitate communication between scientists from different parts of the world, who often rely on them while trying to grasp meanings and make themselves understood despite lack of working knowledge of each other’s language or of International Scientific English, hailed as the modern lingua franca of science (e.g. Wood, 2001). On the other hand, if their semantic equivalence is only partial, they may become potential sources of misunderstanding. Efforts to make international communication between scientists more effective have been and are being undertaken, e.g. by Valentina Skujina from the University of Latvia, who calls for a more disciplined and uniform use of international morphemes and in the long run for establishing a common international structural-semantic model for scientific terminology (Skujina, 2003). The vocabulary of science and technology lies outside the scope of this paper, though, which is concerned with words of everyday use, understood and used by average speakers of a language. The following sections focus on discussion (Eng.) and dyskusja (Pl.), nouns whose international status appears unquestionable and whose straightforward translatability, at first glance at least, seems evident. First, bilingual dictionaries will be consulted to determine the type of equivalence that holds between the two nouns. Second, definitions of discussion and dyskusja provided by monolingual dictionaries will be compared to detect possible differences in the underlying concepts. Finally, a sample of translators’ actual decisions will be examined to confirm, refute or modify the conclusions based on the dictionary analysis.

Discussion and Dyskusja Bilingual dictionaries In the Collins BGW English-Polish Polish-English Dictionary (1996), discussion and dyskusja are recorded as mutual translation equivalents (see Table 8.1). The Great Polish-English English-Polish Dictionary (StanisÙawski, 1999) renders dyskusja as ‘discussion; debate; controversy’ and discussion

9612Arabski.indb 81

13/01/2006 08:43:49

82

Cross-linguistic Influences in the Second Language Lexicon

Table 8.1 Bilingual dictionary definitions Collins BGW English-Polish Polish-English Dictionary (1996) discussion = ‘dyskusja’ dyskusja = ‘discussion’ The Great Polish-English English-Polish Dictionary (StanisÙawski, 1999) dyskusja = ‘discussion; debate; controversy’ discussion = ‘dyskusja; debata’ [‘discussion; debate’] Wielki sÙownik angielsko-polski PWN-Oxford (2002) discussion = 1. ‘dyskusja, rozmowa’ [‘discussion, conversation’] 2. ‘omówienie’ [‘a part of text – a lecture, book or article in which an issue is considered’] as ‘dyskusja; debata’ [‘discussion; debate’] respectively. The equivalence of discussion and dyskusja is also confirmed by the PWN-Oxford EnglishPolish dictionary, which translates discussion as 1. ‘dyskusja, rozmowa’ [‘discussion, conversation’] (Wielki sÙownik angielsko-polski PWN-Oxford, 2002), although here the second available translation equivalent, ‘conversation’, stands in contrast to ‘debate’ offered by StanisÙawski. More importantly, however, the PWN-Oxford dictionary, unlike StanisÙawski or the Collins BGW, records another distinct meaning of discussion, 2. ‘omówienie,’ referring specifically to a part of text (a lecture, book or article) in which an issue is considered. Thus, the nouns discussion and dyskusja are recorded as first translation equivalents in Collins BGW and StanisÙawski, which may suggest – the distributional and formal criteria satisfied – that the two are international terms proper. However, according to the PWN-Oxford dictionary, the most recent source consulted, equivalence holds only for the first sense of the English unit, which qualifies the relationship as partial. Furthermore, the choice of the second translation equivalent in StanisÙawski (‘debate’) and in the PWN-Oxford for sense 1. (‘conversation’) makes the semantic correspondence between discussion and dyskusja still less definitive. Monolingual dictionaries The Collins Cobuild English Language Dictionary (1987) defines discussion as 1. ‘the act of talking or writing about something in detail and from several points of view’, or 2. ‘a conversation, speech, or piece of writing in which a subject is considered in detail from several points of view’ (see Table 8.2). Similarly, the Longman Dictionary of Contemporary English (1987

9612Arabski.indb 82

13/01/2006 08:43:50

International Terms and Profile Transfer

83

Table 8.2 Monolingual dictionary definitions English Collins Cobuild English Language Dictionary (1987) discussion = 1. ‘the act of talking or writing about something in detail and from several points of view’ 2. ‘a conversation, speech, or piece of writing in which a subject is considered in detail from several points of view’ Longman Dictionary of Contemporary English (1987) discussion = ‘a case or the action of considering something by talking or writing about it from several points of view’ Oxford English Dictionary (1992) discussion = 1. ‘examination or investigation of a matter by arguments for and against’ 2. ‘argument or debate with a view to eliciting truth or establishing a point; a disquisition in which a subject is treated from different sides’ Microsoft Bookshelf (1996/7) discussion = 1. ‘consideration of a subject by a group; an earnest conversation’ 2. ‘a formal discourse on a topic; an exposition’ Polish (translation mine – K.W.) SÙownik j¿zyka polskiego (1994) dyskusja = ‘an exchange of opinions on a particular subject in speech or writing, an act of examining or considering an issue together; a conversation, dispute’ Praktyczny sÙownik wspóÙczesnej polszczyzny (1997) dyskusja = ‘an exchange of opinions on a subject, in speech or in writing, usually public; consideration of a subject by a group’

[1978]) explicates it as ‘a case or the action of considering something by talking or writing about it from several points of view’. In the same vein, the Oxford English Dictionary (1992) defines discussion as 1. ‘examination or investigation of a matter by arguments for and against’, or 2. ‘argument or

9612Arabski.indb 83

13/01/2006 08:43:50

84

Cross-linguistic Influences in the Second Language Lexicon

debate with a view to eliciting truth or establishing a point; a disquisition in which a subject is treated from different sides’.1 According to the Microsoft Bookshelf (1996/7), discussion means 1. ‘consideration of a subject by a group; an earnest conversation’, or 2. ‘a formal discourse on a topic; an exposition’. The definitions proposed by Collins, Longman and Oxford appear fairly homogeneous. They refer to an act of language use, in speech or in writing, which aims at a thorough examination of an issue from various perspectives, or to the product of the act. Interestingly, none of them foregrounds the group or solitary aspect of the activity, although they all more or less overtly admit both. Only Microsoft Bookshelf appears to differentiate between two meanings of discussion according to its conversational or expository, that is, essentially solitary form. According to the Polish language dictionary, dyskusja means ‘an exchange of opinions on a particular subject in speech or writing, an act of examining or considering an issue together; a conversation, dispute’ (SÙownik j¿zyka polskiego, 1994). A practical dictionary of contemporary Polish defines it in a very similar way, as ‘an exchange of opinions on a subject, in speech or in writing, usually public; consideration of a subject by a group’ (Praktyczny sÙownik wspóÙczesnej polszczyzny, 1997). Thus both discussion and dyskusja refer to the act of using language with a view to a detailed examination of an issue from various vantage points. In Polish, however, unlike in English, the conversational, collective aspect of exchanging views and considering something together appears explicit and valid. By contrast, the English term seems to emphasise the product of the activity if written (sense 2 in Collins Cobuild…, 1987) and, according to Microsoft Bookshelf (1996/7), refers also to expository writing on a topic, which may be approximated in Polish as ‘omówienie’ (sense 2. of discussion according to the PWN-Oxford, 2002) and which is absent from the Polish noun dyskusja. Analysis of text samples Leaving aside the problem of text unit which is referred to in English by discussion and for which the Polish language has other terms (that is the point where equivalence between discussion and dyskusja is not observed), it appears that although there is a large semantic overlap between the two, the English and Polish lexical units name concepts which differ slightly with regard to the number of active contributors, that is in the ‘collective’ component. This component, according to the dictionary definitions, seems obligatory in Polish but not in English. Since the underlying concepts do appear to differ, it may be expected that this variation will be somehow reflected in translators’ practice. In other words, the level of actual choices may reveal regularities which could

9612Arabski.indb 84

13/01/2006 08:43:51

International Terms and Profile Transfer

85

shed some light on other differences between the social situations referred to as discussion in English and dyskusja in Polish – if such differences exist. In order to find this out, four English texts and their Polish translations were consulted, namely Jane Austen’s Pride and Prejudice (1813), Joseph Conrad’s Nostromo (1904), Henry James’ The Portrait of a Lady (1881), and Emily Brontë’s Wuthering Heights (1847). The choice of these texts was motivated by the fact that, irrespective of individual literary tastes, they are recognised as examples of so-called ‘good prose’ and that, being in public domain, they are easily available in electronic form, which made the analysis easier. Moreover, it seemed advisable to choose works by different authors, who developed their own distinctive styles. Also important was to choose works translated by different translators, so as not to restrict the analysis to a particular translator’s style, strategy or predilections. The corpus analysed was certainly too small to promise a conclusive answer to the problem of (non)equivalence of concepts; the obtained results may thus serve only as indication that further research in this matter is or is not worth undertaking. It is understood that translation of any text, and of a literary text in particular, does not mean manipulating lexical equivalents. Therefore, the absence of dyskusja in Polish was not automatically regarded as indicative of its unsuitability in the context and, consequently, of a possible nonequivalence of the concepts. It was the presence of the word that was analysed in terms of the features of the social situation referred to. In the four English texts discussion appears 28 times, each time referring to a group situation, that is, including the ‘collective’ component inherent in the definition of dyskusja but not obligatory in English. Thus, it would seem that in all the identified contexts it is possible to refer to the situation by the Polish term. Dyskusja appears in the corresponding fragments of Polish translations 17 times. Interestingly, an analysis of these passages shows that all the contexts involve an explicit divergence of opinion between the speakers, although the extent to which the opinions differ varies. Generally speaking, dyskusja appears in two types of contexts: the context of disagreement (PP3; NO2; PL1, 2, 5, 7–12, 14–16) and the context of confusion (NO3, 4, 5), each time referring to an exchange of arguments between the speakers, who act as sides rather than a party.2 In the former type of context, the element of divergence of opinion is very strongly marked. Thus, in PP3 discussion refers to a difference of opinion between Mr. and Mrs. Bennet concerning the family’s summer visit in Brighton. In passage NO2 reference is made to Antonia Avellanos, an educated, well-read and resolute young lady, who not only could afford her own opinions but

9612Arabski.indb 85

13/01/2006 08:43:52

86

Cross-linguistic Influences in the Second Language Lexicon

also had the courage to defend them in front of male opponents. In PL1 Isabel Archer, Lord Warburton, Ralph and Mr. Touchett are exchanging conflicting opinions concerning old English houses, in PL2 Mrs. Touchett and Henrietta Stackpole express extreme opinions concerning American hotels, and in PL5 Madame Merle argues that a woman’s dress expresses her personality, a view to which Isabel Archer strongly objects. In the same vein, in PL7 Mrs. Touchett expresses a highly critical opinion about the Countess Gemini and disapproves of her visit, while Madame Merle defends her. Passages PL8, 9, 10, and 11 refer to a long conversation between Isabel Archer and Ralph Touchett in which they differ sharply in their opinions concerning Mr. Osmond, whom Isabel intends to marry. In PL12 the elements of disagreement and exchange of arguments are made plain by the immediate co-text. In PL14 Henrietta Stackpole and Ralph Touchett disagree over the character of Mr. Bantling. In PL15 Mr. Osmond objects to his wife Isabel’s going to England to see her dying cousin Ralph, while Isabel insists on going. In PL16 discussion refers to an earlier conversation between Mr. Bantling and Henrietta Stackpole, in which the former expressed the opinion that Isabel would not openly defy her husband’s wish by coming to see Ralph, while the latter was sure Isabel would come. In the second type of context, the context of confusion, the element of divergence of opinion is also present, but the emphasis shifts from an open conflict or polemic to a variety of opinions. Thus, passage NO3 refers to an outburst of political speculations concerning future events after the Monterist Revolution. Although the general mood is optimistic, the scenarios of future order and stability are different. This excerpt represents the lowest point on the ‘divergence of opinion’ scale. Passage NO4 depicts a state of confusion following a letter from father Corbelàn, who escaped to the bandit Hernandez. In his letter he reveals Martin Decoud’s plan to establish a new state and hopes that Hernandez will be able to defend the territory against Montero until General Barrios’s arrival. People gathered in the Amarilla Club, where the letter is brought, are confused with this bolt-from-the-blue declaration and debate whether to join Hernandez and his band or to remain in Sulaco waiting for Montero. Similarly, NO5, which is a continuation of NO4, contains an exchange of arguments and opinions concerning the future course of action – whether to join Hernandez or to stay in Sulaco.3 It appears that the word dyskusja was used by the translators only in those contexts which involved an element of divergence of opinion (see Table 8.3). In all the passages from which this element is absent (PP1, 2, 4, 5; NO1; PL3), the translators opted for other solutions (see Table 8.4).

9612Arabski.indb 86

13/01/2006 08:43:52

87

International Terms and Profile Transfer

Table 8.3 Text samples. Dyskusja appears in the Polish translation; discussion involves divergence of opinion English

Polish

PP3

She had not been many hours at home before she found that the Brighton scheme, of which Lydia had given them a hint at the inn, was under frequent discussion between her parents. (chap.39)

Po kilku godzinach stwierdziÙa, Če plan wyjazdu do Brighton, o którym Lidia wspominaÙa w gospodzie, jest przedmiotem cz¿stych dyskusji pomi¿dzy rodzicami. (p.223)

NO2

At the receptions – where the situation was saved by the presence of a very decrepit old lady (a relation of the Corbelàns), quite deaf and motionless in an armchair – Antonia could hold her own in a discussion with two or three men at a time. (part 2, chap.2)

W czasie przyj¿° – kiedy to sytuacj¿ ratowaÙa obecnoï° zgrzybiaÙej damy (krewnej Corbelànów), caÙkiem gÙuchej i nieruchomej w swym fotelu – Antonia potrafiÙa stawia° w dyskusji czoÙo dwom lub trzem m¿Čczyznom naraz. (pp.127–8)

NO3

Don Justè’s fan-shaped beard was still the centre of loud and animated discussions. (part 2, chap.5)

GÙoïne i oČywione dyskusje skupiaÙy si¿ nadal wokóÙ wachlarzowatej brody don Justègo. (p.159)

NO4

“Heaven itself wills it. Providence is on our side,” wrote Father Corbelàn; there was no time to reflect upon or to controvert his statement; and if the discussion started upon the reading of that letter in the Amarilla Club was violent, it was also shortlived. (part 3, chap.3)

“Samo niebo tego chce. Opatrznoï° jest po naszej stronie” – pisaÙ ksi­dz Corbelàn. Nie byÙo czasu, by zastanawia° si¿ nad tym oïwiadczeniem lub podawa° je w w­tpliwoï°. ToteČ jeïli dyskusja wywoÙana odczytaniem owego listu w klubie “Amarilla” byÙa gwaÙtowna, byÙa ona równieČ krótkotrwaÙa. (p.279)

NO5

All the latter part of the afternoon an animated discussion went on in the big rooms of the Amarilla Club. (part 3, chap.3)

Przez caÙe póĊniejsze popoÙudnie w wielkich salach klubu “Amarilla” toczyÙa si¿ oČywiona dyskusja. (p.279)

PL1

In this discussion Ralph Touchett took no interest whatever; he stood with his hands in his pockets, looking greatly as if he should like to renew his conversation with his new-found cousin. (chap.2)

Ralph Touchett nie zdradzaÙ najmniejszego zainteresowania t­ dyskusj­; staÙ z r¿kami w kieszeniach I najwidoczniej miaÙ ochot¿ nawi­za° znów rozmow¿ ze ïwieČo odkryt­ kuzynk­. (p.29)

continued…

9612Arabski.indb 87

13/01/2006 08:43:54

88

Cross-linguistic Influences in the Second Language Lexicon

PL2

This contribution to the discussion, however, Miss Stackpole rejected with scorn. Middling indeed! (chap.11)

Ale panna Stackpole z pogard­ odtr­ciÙa jego gÙos w tej dyskusji. (p.132)

PL5

“Should you prefer to go without them?” Madame Merle enquired in a tone which virtually terminated the discussion. (chap.19)

- WolaÙabyï obywa° si¿ bez tych osÙon? – spytaÙa madame Merle tonem, który poÙoČyÙ kres dalszej dyskusji. (p.277)

PL7

The Countess gave rise indeed to some discussion between the mistress of the house and the visitor from Rome, in which Madame Merle (who was not such a fool as to irritate people by always agreeing with them) availed herself felicitously enough of that large licence of dissent which her hostess permitted as freely as she practised it. (chap.26)

Wizyta hrabiny wywoÙaÙa dyskusj¿ mi¿dzy pani­ domu a jej przyjacióÙk­ z Rzymu; madame Merle (za m­dra, Čeby irytowa° ludzi nieustannym przytakiwaniem) doï° zr¿cznie wykorzystaÙa prawo do róČnicy pogl­dów, przywilej, który pani Touchett przyznawaÙa bliĊnim i który sama zawsze w praktyce stosowaÙa. (p.381)

PL8

He felt agitated, intensely eager; now that he had opened the discussion he wished to discharge his mind. (chap.34)

ByÙ podniecony, zapaliÙ si¿ teraz, skoro zacz­Ù t¿ dyskusj¿, chciaÙ zrzuci° wszystko z serca. (p.469)

PL9

Isabel had uttered her last words with a low solemnity of conviction which virtually terminated the discussion, and she closed it formally by turning away and walking back to the house. (chap.34)

Ostatnie zdania wypowiedziaÙa cicho, uroczyïcie, z siÙ­ przekonania, koÚcz­c dyskusj¿; zamkn¿Ùa j­ ostatecznie odwracaj­c si¿ i ruszaj­c w kierunku domu. (p.474)

PL10

He had uttered his thought, as we know, and after this had held his peace, Isabel not having invited him to resume a discussion which marked an era in their relations. (chap.39)

Jak wiadomo, wyraziÙ t¿ swoj­ opini¿, póĊniej jednak milczaÙ, poniewaČ Isabel nie zach¿ciÙa go do podejmowania na nowo dyskusji, która staÙa si¿ punktem zwrotnym w stosunkach mi¿dzy nimi. (p.530)

PL11

That discussion had made a difference – the difference he feared rather than the one he hoped. (chap.39)

Dyskusja poci­gn¿Ùa za sob­ pewne skutki, ale te, których si¿ obawiaÙ, zamiast tych, których pragn­Ù. (p.530)

continued…

9612Arabski.indb 88

13/01/2006 08:43:55

89

International Terms and Profile Transfer PL12

Certainly she had fallen into exaggerations – she who used to care so much for the pure truth; and whereas of old she had a great delight in good-humoured argument, in intellectual play (she never looked so charming as when in the genial heat of discussion she received a crushing blow full in the face and brushed it away as a feather), she appeared now to think there was nothing worth people’s either differing about or agreeing upon. (chap.39)

Dawniej lubiÙa dobrotliwe spory i pojedynki intelektualne (nigdy nie byÙa pi¿kniejsza niČ w chwilach, gdy w szczerym zapale dyskusji otrzymuj­c miaČdČ­cy cios, op¿dzaÙa si¿ od niego jak od muchy); teraz moČna by myïle°, Če nie ma dla niej spraw wartych tego, Čeby si¿ ludzie o nie sprzeczali lub co do nich zgadzali. (pp.536–7)

PL14

Discussion was stimulated of course by their inevitable difference of view - Ralph having amused himself with taking the ground that the genial ex-guardsman was a regular Machiavelli. (chap.47)

Oczywiïcie dyskusj¿ t¿ oČywiaÙa nieunikniona róČnica zdaÚ, bo Ralph zabawiaÙ si¿ udowadnianiem, Če pogodny ex-gwardzista jest drugim Machiavellim. (p.676)

PL15

I don’t see why we should prolong our discussion; you know my wish. (chap.51)

Niepotrzebnie przedÙuČasz dyskusj¿, znasz moje Čyczenia. (p.729)

PL16

“He knows all about it,” she added; “we had quite a discussion. He said you wouldn’t, I said you would.” (chap.53)

– Bantling wszystko wie – wyjaïniÙa przyjacióÙce. – Stoczyliïmy caÙ­ dyskusj¿. On twierdziÙ, Če tego nie zrobisz, a ja byÙam pewna, Če si¿ nie cofniesz. (p.766)

Thus, in passage PP1, where reference is made to a repeated analysis of the family’s situation after Elizabeth’s refusal to accept Mr. Collins’s offer of marriage and to the mother’s speculations concerning the family’s well-being, had the offer been accepted, the translator chose roztrz­sania (from the verb roztrz­sa°, semantically and morphologically related to the Latin discutere; BaÚkowski, 2000). In passage PP2, where discussion refers to a conversation between Elizabeth Bennet and Mr. Wickham concerning the reasons for Mr. Wickham’s absence at the ball as well as Mr. Darcy’s hostile character and his reportedly infamous conduct, the translator opted for a verbal construction, rozmawia° o [‘talk about’]. In PP4, where discussion refers to a conversation between Elizabeth Bennet and the Gardiners, who on their way to Longbourn analyse the situation after

9612Arabski.indb 89

13/01/2006 08:43:56

90

Cross-linguistic Influences in the Second Language Lexicon

Table 8.4 Text samples. Dyskusja does not appear in the Polish translation; discussion does not involve divergence of opinion English

Polish

PP1

The discussion of Mr. Collins’s offer was now nearly at an end, and Elizabeth had only to suffer from the uncomfortable feelings necessarily attending it, and occasionally from some peevish allusions of her mother. (chap.21)

Roztrz­sania nad oïwiadczynami pana Collinsa miaÙy si¿ ku koÚcowi i ElČbieta musiaÙa juČ tylko ïcierpie° nieprzyjemne wraČenie, jakie towarzyszy podobnym sytuacjom i od czasu do czasu znosi° zrz¿dzenie matki. (p.121)

PP2

She highly approved his forbearance, and they had leisure for a full discussion of it, and for all the commendation which they civilly bestowed on each other, as Wickham and another officer walked back with them to Longbourn, and during the walk he particularly attended to her. (chap.21)

PochwaliÙa gor­co jego powïci­gliwoï°. Mogli dÙugo rozmawia° o tej sprawie i mówi° sobie nawzajem wiele miÙych rzeczy, bowiem Wickham wraz z drugim oficerem odprowadziÙ je do domu, a przez caÙy czas poïwi¿caÙ jej osobie najwi¿cej uwagi. (p.122)

PP4

It may be easily believed, that however little of novelty could be added to their fears, hopes, and conjectures, on this interesting subject, by its repeated discussion, no other could detain them from it long, during the whole of the journey. (chap.47)

Ratwo uwierzy°, Če przez caÙy czas podróČni nie potrafili zaj­° si¿ dÙuČej innym tematem niČ ten, cho° niewiele juČ nowego moČna byÙo doda° do powtarzanych ci­gle obaw, nadziei i domysÙów. (p.278)

PP5

“What could become of Mr. Bingley and Jane!” was a wonder which introduced the discussion of their affairs. (chap.58)

– Gdzie si¿ podziaÙ Bingley i Jane? – tym pytaniem zapocz­tkowali rozmow¿ o tamtych dwojgu. (p.364)

NO1

For this natural reason these discussions were precious to Mrs. Gould in her engaged state. (part 1, chap.6)

Z tej naturalnej przyczyny rozmowy owe byÙy cenne dla pani Gould w jej narzeczeÚskich czasach. (p.61)

PL3

It had had, verily, two causes: part of it was to be accounted for by her long discussion with Mr. Goodwood, but it might be feared that the rest was simply the enjoyment she found in the exercise of her power. (chap.17)

W istocie zÙoČyÙy si¿ na to dwie przyczyny: cz¿ïciowo moČna je zÙoČy° na karb dÙugiej i trudnej rozmowy z Casparem, ale, niestety, znaczn­ rol¿ graÙo teČ rozkoszne poczucie wÙadzy. (p.226)

9612Arabski.indb 90

13/01/2006 10:18:11

9612Arabski.indb 91

Its occupants had recommended their angry discussion: Mrs. Linton, at least, was scolding with renewed vigour; Heathcliff had moved to the window, and hung his head, somewhat cowed by her violent rating apparently. (chap.11)

Hareton, during the discussion, stood with his hands in his pockets, too awkward to speak; though he looked as if he did not relish my intrusion. (chap.18)

Isabel was apparently on the point of assuring her cousin for the third time that he was odious; but the discussion was interrupted by the arrival of the lady who was the topic of it. (chap.18)

Isabel listened to the discussions taking place between the two with a sense that she might derive much benefit from them and that they were among the advantages she couldn’t have enjoyed for instance in Albany. (chap.23)

This had occurred to her just before her marriage, after her little discussion with her aunt and at a time when she was still capable of that large inward reference, the tone almost of the philosophic historian, to her scant young annals. (chap.40)

WH1

WH2

PL4

PL6

PL13

In Polish – argument or quarrel [‘they were arguing again’].

In Polish – conversation [‘during this conversation’].

In Polish – conversation [‘the lady about whom they were talking’].

In Polish – dispute [‘Isabel listened to these disputes’].

In Polish – conversation [‘after the conversation with her aunt’].

Tamci znów si¿ kÙócili; wÙaïciwie pani Linton krzyczaÙa w uniesieniu, a Heathcliff staÙ przy oknie ze zwieszon­ gÙow­, jakby nieco przeraČony jej gwaÙtownoïci­. (p.117)

Podczas tej rozmowy Hareton staÙ z r¿kami w kieszeniach, zbyt nieïmiaÙy, by si¿ odezwa°. Ale wiedziaÙam, Če byÙ nierad z mojego przybycia. (p.198) Isabel juČ otwieraÙa usta, Čeby po raz trzeci powiedzie° kuzynowi, Če jest wstr¿tny, lecz w tym momencie do salonu weszÙa osoba, o której mówili. (p.243) Isabel przysÙuchiwaÙa si¿ tym sporom, zadowolona, Če tak duČo moČe si¿ od nich nauczy° i stwierdzaÙa w duchu, Če to jeszcze jeden z wielu przywilejów, których nie zyskaÙaby nigdy siedz­c na przykÙad w Albany. (p.337) Tak myïlaÙa Isabel przed ïlubem, wkrótce po rozmowie z ciotk­, w okresie, gdy jeszcze byÙa zdolna do tak ogólnych refleksji i niemal filozoficznego spojrzenia na doï° jeszcze krótk­ I skromn­ kronik¿ swojego Čycia. (p.551)

Table 8.5 Text samples. Dyskusja does not appear in the Polish translation; discussion does involve divergence of opinion.

International Terms and Profile Transfer 91

13/01/2006 08:43:58

92

Cross-linguistic Influences in the Second Language Lexicon

Lydia’s elopement with Mr. Wickham, the translator chose powtarzane obawy… [‘repeatedly expressed fears’]. In passages PP5, NO1 and PL3, the translators opted for rozmowa [‘conversation’], which in PP5 refers to a conversation between Elizabeth Bennet and Mr. Darcy after declarations of love have been exchanged; in NO1 to conversations between Charles Gould and his fiancée about Mr. Gould, senior, and the silver mine in Costaguana, and in PL3 to a conversation between Isabel and Caspar Goodwood, whose offer of marriage Isabel rejected in favour of her plans to travel. In none of the above-mentioned passages is discussion used to refer to a situation which involves a clash of opinion. In the remaining passages (WH1, 2; PL4, 6, 13) the element of disagreement is present, but the translators opted for other solutions (see Table 8.5).

Profile Transfer: Concluding Remarks Although the number of studied samples is too small to allow for a definitive statement or conclusion, an analysis of the translated passages seems to indicate that there is a marked tendency for the situation referred to by the Polish term dyskusja to involve – apart from the collective component, manifest already at the dictionary level – an explicit divergence of opinion between participants in the interaction. This component seems irrelevant to the situation-type referred to by the English term discussion. I would like to propose, however tentatively, that discussion and dyskusja profile situations which are slightly different with regard to the role of and the relationship between the participants. According to Langacker (e.g. 1990), conceptualisation involves the foregrounding of certain elements within a specific cognitive domain or mental context, called base. The meaning of lexical items is constructed relative to the base and by selection of a particular substructure within the base – an element, relationship or process – which is given prominence and which constitutes the profile of the concept, its semantic identity and discreteness from other concepts profiled within the same base. An analysis of dictionary definitions and the results obtained from a small sample of contexts seem to suggest that dyskusja involves the foregrounding of elements within the domain of discourse which discussion does not foreground. These elements include specifically the role of the participants in the situation – dyskusja appears to presuppose active participation of at least two interactants, no matter whether in speech or in writing; and the relationship between them – which, if the situation is to be categorised as dyskusja, has to involve some divergence of opinion. The

9612Arabski.indb 92

13/01/2006 08:43:59

International Terms and Profile Transfer

93

differences between discussion and dyskusja could be recapitulated in the following way: •

• •

an academic text or its part can itself constitute and be referred to as discussion; dyskusja can be initiated by an academic text and consists in the interaction between various texts; a lecture can constitute and be referred to as discussion; dyskusja can take place after the lecture; a situation when all the participants agree or want to be perceived as agreeing with one another is not what Polish speakers normally call dyskusja.4

As noted in the introduction, dyskusja is becoming more and more common in Polish academic texts (especially by natural scientists) as the title of a text unit corresponding to ‘Discussion’ in English research articles, which in the Polish academic tradition is called ‘Omówienie wyników’ [‘discussion of results’]. In a sample of 21 monographs in biology and biochemistry published in Polish by the University of Silesia in the years 1985–2001, only three used the traditional heading and one opted for another title of the chapter following the presentation of results. In one case the author attempted to reconcile tradition with current trends in the rather cumbersome heading ‘Omówienie wyników badaÚ i dyskusja’ [‘discussion of results and discussion’]. In four other cases the heading – ‘Dyskusja wyników’ [‘discussion of results’], which is an attempt to translate the English prepositional phrase into the Polish inflectional paradigm – appears to violate grammatical restrictions: dyskusja does not admit postmodification by the genitive and requires a prepositional phrase (dyskusja nad/o [‘discussion over/about’] + Nlocative) or a participle (dyskusja dotycz­ca [‘discussion concerning’] + Ngenitive) to specify the subject matter. The process of substitution of a traditional section heading by dyskusja seems to be influenced by the conventions of English academic texts rather than result from a natural semantic extension of the term. It is worth noting that the proportion of references to English writings, especially in the field of natural sciences, is generally very high and that in some of the monographs consulted the percentage of English language bibliography entries is almost 100. Thus, the exposure of Polish authors to texts written in English is very high too, and many of them, recognising the importance of addressing international scientific audiences, attempt to have their works published in English. These two factors – exposure and aspiration to English academic conventions – may contribute to the fact that in some cases Polish writers appear

9612Arabski.indb 93

13/01/2006 08:43:59

94

Cross-linguistic Influences in the Second Language Lexicon

better versed in English norms of scientific writing than Polish traditions of the academy and more sensitive to the expectations of the international scientific community, which they then transfer, consciously or not, into the Polish text. However, in English the section-referring meaning of discussion appears consistent with the profile of the situation to which the term refers. This is not the case in Polish. This is why it seems possible to view the use of dyskusja in Polish academic texts as a result of borrowing transfer, with possible consequences for the profile of the Polish term. In other words, it can be hypothesised that a process initiated on the level of discourse structure may result in changes in the semantic structure of a lexeme. Notes 1. Obsolete, jocular and strictly technical senses have been omitted. 2. Symbols refer to analysed samples presented in Tables 8.3–8.5. PP – Jane Austen, Pride and Prejudice; WH – Emily Brontë, Wuthering Heights; NO – Joseph Conrad, Nostromo; PL – Henry James, The Portrait of a Lady. Samples are numbered successively according to their position in the text. Discussion and dyskusja are shown in italics; dotted line marks relevant fragments of Polish translations from which dyskusja is absent. 3. The element of clash of opinions and attitudes is made clear further in the text. ‘As dusk fell,’ we read, ‘Don Juste Lopez, inviting those caballeros who were of his way of thinking to follow him, withdrew into the corredor, where at a little table in the light of two candles he busied himself in composing an address, or rather a solemn declaration to be presented to Pedrito Montero by a deputation of such members of Assembly as had elected to remain in town’ (Conrad, 2000, part 3, chap. 3). 4. It seems that this point may also concern the Italian discussione. David Katan (1999: 231) mentions difficulties in translating the Italian term into English, pointing out that the situation profiled by discussione may in fact represent discussion, argument or fight for the British English.

References Braun, P. (1990) Internationalismen – Gleiche Wortschätze in europäischen Sprachen. In P. Braun, B. Schaeder and J. Volmert (eds) Internationalismen. Studien zur interlingualen Lexikologie und Lexikographie (pp. 13–33). Tübingen: Max Niemeyer Verlag. Braun, P., Schaeder, B. and Volmert, J. (eds) (1990) Internationalismen. Studien zur interlingualen Lexikologie und Lexikographie. Tübingen: Max Niemeyer Verlag. Katan, D. (1999) Translating Cultures. Manchester: St. Jerome. Kolwa, A. (2001) Internationalismen im Wortschatz der Politik. Frankfurt am Main: Peter Lang. Langacker, R.W. (1990) Concept, Image, and Symbol. The Cognitive Basis of Grammar. Berlin: Mouton de Gruyter. Lipczuk, R. (1992) Internacjonalizmy a ‘faÙszywi przyjaciele tÙumacza’. [International terms and ‘false friends’] In J. Ma°kiewicz and J. Siatkowski

9612Arabski.indb 94

13/01/2006 08:44:00

International Terms and Profile Transfer

95

(eds) J¿zyk a kultura 7: Kontakty j¿zyka polskiego z innymi j¿zykami na tle kontaktów kulturowych (pp. 135–43). WrocÙaw: Wiedza o Kulturze. Ma°kiewicz, J. (1984) Co to s­ tzw. internacjonalizmy? [What are the so called international terms?] J¿zyk Polski 3, 176–84. Skujina, V. (2000) Vitality of Latin and Greek word-elements in the modern interlingual term-building processes – Online document. Retrieved 7 February 2003 from http://www.liis.lv/latval/Darbinieki/Jauniecit/vasarefer.htm Wierzbicka, A. (1991) Cross-Cultural Pragmatics: The Semantics of Human Interaction. Berlin: Mouton de Gruyter. Wood, A. (2001) International scientific English: The language of research scientists around the world. In J. Flowerdew and M. Peacock (eds) Research Perspectives on English for Academic Purposes (pp. 71–83). Cambridge: Cambridge University Press. Dictionaries BaÚkowski, A. (2000) Etymologiczny sÙownik j¿zyka polskiego. [Etymological dictionary of the Polish language] Warszawa: Wydawnictwo Naukowe PWN. Collins BGW English-Polish Polish-English Dictionary (1996) Warszawa: Polska Oficyna Wydawnicza BGW. Collins Cobuild English Language Dictionary (1987) London: Harper Collins Publishers. Longman Dictionary of Contemporary English (1987)[1978] New Edition. Harlow, Essex: Longman. Microsoft Bookshelf (1996/7) CD-ROM. Microsoft Corporation. Oxford English Dictionary on CD-ROM (1992) Version 1.02. Oxford: Oxford University Press. Praktyczny sÙownik wspóÙczesnej polszczyzny [Practical dictionary of contemporary Polish] Vol. 10 (1997) PoznaÚ: Kurpisz. SÙownik j¿zyka polskiego [Polish language dictionary] (1994[1978]) Warszawa: Wydawnictwo Naukowe PWN. StanisÙawski, J. (1999) Wielki sÙownik angielsko-polski. The Great English-Polish Dictionary. Warszawa: Philip Wilson. StanisÙawski, J. (1999) Wielki sÙownik polsko-angielski. The Great Polish-English Dictionary. Warszawa: Philip Wilson. Wielki sÙownik angielsko-polski PWN-Oxford [Great English-Polish dictionary] (2002) Warszawa: PWN. Fiction Austen, J. (1998 [1813]) Pride and Prejudice. Urbana, IL: Project Gutenberg. Etext #1342. First Release: Jun 1998. Austen, J. (1996) Duma i uprzedzenie. Trans. A. PrzedpeÙska-Trzeciakowska. Warszawa: PrószyÚski i S-ka. Conrad, J. (2000 [1904]) Nostromo. Urbana, Illinois: Project Gutenberg. Etext #2021. First Release: Jan 2000. Conrad, J. (1987) Nostromo. Trans. J.J. SzczepaÚski. In DzieÙa wybrane III, 5-435. Warszawa: PaÚstwowy Instytut Wydawniczy. Brontë, E. (1996 [1847]) Wuthering Heights. Urbana, Illinois: Project Gutenberg. Etext #768. First Release: Dec 1996.

9612Arabski.indb 95

13/01/2006 08:44:01

96

Cross-linguistic Influences in the Second Language Lexicon

Brontë, E. (1990) Wichrowe Wzgórza. Trans. J. Sujkowska. WrocÙaw: ZakÙad Narodowy im. OssoliÚskich. James, H. (2001 [1881]) The Portrait of a Lady. Vols. 1-2. Urbana, Illinois: Project Gutenberg. Etext #2833-4. First Release: Sep 2001. James, H. (1977) Portret damy. Trans. M. Skibniewska. Warszawa: Czytelnik.

9612Arabski.indb 96

13/01/2006 08:44:02

Chapter 9

The Influence of English on Polish Drug-related Slang MAGDALENA BARTŁOMIEJCZYK Introduction Influence of English on Polish The first borrowings from English appeared in the Polish language as early as the 18th century (MaÚczak-Wohlfeld, 1995: 24); however, for many years their number has been relatively limited in comparison with loanwords from other languages. It was not before the second half of the 20th century that English became the most popular source of new borrowings, not only lexical ones, but also other types. The extensive influence English exerts on contemporary Polish is a well-known fact. MaÚczak-Wohlfeld (1995: 68–73) enumerates 45 semantic fields in which English borrowings appear, the three most modern ones, which, according to her, developed after 1992, comprise vocabulary connected with work, advertising and drugs. The English influence in various realms of language has already been described in many articles. To give just a few examples, the language of computing, including English loanwords in this field, was discussed by Matusiak (1997), anglicisms in the language of medicine by Górnicz (2000), and words of English origin in vocabulary related to professional career by Ociepa (2001). However, in such realms as those mentioned above, English loanwords are normally spread by well-educated professionals who know English and the specialised literature and magazines published in English about their subject of personal interest. As far as slang is concerned, Moch (2002) gives numerous examples of vocabulary specific for the Polish hip-hop subculture and includes, apart from vocabulary connected with music and social life, some drugrelated words and expressions, of which a large part is imported from English.

97

9612Arabski.indb 97

13/01/2006 08:44:03

98

Cross-linguistic Influences in the Second Language Lexicon

Specificity of Drug-related Slang The aim of the study whose results are described here was to see how English affects a realm of Polish which is different in its character from a professional slang, namely the vocabulary employed by Polish drug users to describe various concepts connected with drug culture, such as names of illegal substances, paraphernalia, people involved in use or sale of drugs and ways of administration. The difference mentioned above manifests itself in at least three aspects: (1) Unlike the case of a professional slang, effective communication as such is not the main reason behind developing and using this type of slang, whose crucial roles consist in self-identification of speakers as members of a consolidated group and preventing outsiders from getting to know the secrets shared by the group (Crystal, 1988: 53). (2) Drug slang manifests itself first and foremost in the spoken language. As drug use is illegal and, moreover, widely socially unacceptable, it is relatively rarely written about in an informal manner since this might be interpreted to imply approval. (3) Drug slang is generally not supposed to be developed by well-educated Poles knowing English. Therefore, we might expect the influence of English on this particular realm to be limited in comparison with any professional slang. Special symbols used in the article / is used to separate alternative spellings of the same word. Angle brackets < > comprise literal translations of Polish expressions into English. Meanings of slang words and expressions are given after an arrow: ª. The symbol ½ is used before the English word which was the source of a lexical borrowing in cases where Polish spelling might obscure such a relation.

Research Project: Method and Material The study was based on a comparative analysis of a corpus of English and Polish drug-related expressions aimed at discovering borrowings as well as other similarities. The English corpus included about 2500 words and expressions originating mainly from various slang dictionaries, as these proved to be quite comprehensive in this field. Johnson (1993) was a particularly useful

9612Arabski.indb 98

13/01/2006 08:44:03

Influence of English on Polish Drug-related Slang

99

source, but some items also came from other slang dictionaries, such as Ayto and Simpson (1993), Spears (1996) and Thorpe (1990). The origin of the Polish corpus, comprising about 1000 items, is more complex. Dictionaries were also used, Czeszewski (2001) and St¿pniak (1993) being of particular importance. Ostrowska’s glossary (1983) is not large, but it shows that some of the items originating from English were already present in the Polish drug slang as early as the beginning of the 1980s. Some of the most recent vocabulary was found in online slang glossaries. Two of them deserving a special mention are: www.republika.pl/ narkotyki/zargon/html and www.slang.elo.pl. Another important source of the Polish corpus is the spoken language, which means that a considerable part of the expressions in question was heard by the author in utterances made by her peers in casual conversations as well as in interviews she conducted with several drug addicts in the help centre for addicts in D­brówka near Katowice. Also items identified in lyrics of rock and hip-hop songs or in interviews with performers of such music had spoken language as their source. In many cases what had been heard was later verified by some written medium. The Internet provided a lot of relevant information as well. Apart from the on-line glossaries mentioned above, numerous expressions were found on websites devoted to drugs, such as www.hyperreal.info or www.marysite.prv.pl. Such sites claim to have an educational character, but most are very informal in their register and often provide ample information on topics such as modes of use, dosage, etc. Drugs are a controversial subject, so practically every article about them placed on any popular website (like www.onet.pl) triggers a lot of comments from internauts, some of which are pure slang bearing a very close similarity to spoken utterances. Lastly, some of the analysed expressions came from books written by former drug addicts and describing their experiences connected with drug use along with other issues. These included Michalewski (1992) and Rosiek (1997). As slang is very changeable and many expressions go out of use even before they are noted anywhere, the author realises that many of the examples she uses are likely not recent; some of them might even be ‘dead’ in slang use by now. This article, however, is meant to show some general trends, which are more durable than particular items illustrating them.

English Borrowings in Polish Drug-related Slang Lexical borrowings The phenomenon of assimilating English words into Polish is the most conspicuous aspect of the diverse influence of English on Polish in general,

9612Arabski.indb 99

13/01/2006 08:44:04

100

Cross-linguistic Influences in the Second Language Lexicon

and the same is valid for slang. Some loanwords undergo far-reaching phonological assimilation and enter a Polish declination pattern. However, as relatively recent additions to the Polish lexicon, such words are clearly recognisable as non-native items. We can distinguish between two types of loanwords in this group: (1) Loanwords having the same meaning in Polish as their English predecessors in English slang, for example: Charlie brown sugar gras, staf (½ stuff), skun (½ skunk), szit/szid (½ shit), baka (½ bacha) joint/dČoint, splif, blant/blunt acid crack spid (½ speed) bongo (½ bhong) flashback snifowa° (½ to sniff)

ª cocaine ª heroin for smoking ª marijuana ª a marijuana cigarette ª LSD ª cocaine for smoking ª amphetamine ª a water pipe for smoking marijuana ª a sudden reoccurrence of a drug experience ª to inhale a drug

The words given above do not constitute a homogeneous group. Firstly, they are at various stages of assimilation, which is represented by their spelling and declination or lack of it; brown sugar functions as a quote, whereas blant is both spelled in a way which shows Polish pronunciation and is declined like any native Polish noun having the same grammatical gender and ending (compare the expressions requiring genitive jara° blanta or skr¿ta ª smoke a marijuana cigarette). Secondly, the degree of their reach varies considerably, consequently, a word such as joint can perhaps hardly be described as belonging to drug slang, as it is commonly understood by young Poles even if they are not even occasional drug users. (2) Words originating from English, but not used in English drug-related slang with this particular meaning. These are mainly words which probably had been first assimilated by Polish in their basic meaning and afterwards underwent a semantic change involving an extension of meaning. The last example, however, is different, as it is clearly found in Polish only in slang. It is included here because it is an interesting case of an English non-slang word which in Polish both belongs to a different register and possesses a much narrower range

9612Arabski.indb 100

13/01/2006 08:44:05

Influence of English on Polish Drug-related Slang

101

of meaning, referring only to illegal substances and not also to medicines, as in English. hapening, party snack bar spray teletext dragi

ª an occasion of taking drugs together by several persons ª a place where drugs are taken ª a hypodermic syringe containing heroin ª a person intoxicated by amphetamine or ecstasy ª illegal drugs

Borrowings of this type in the analysed corpus are almost exclusively nouns, probably because they can, unlike verbs and adjectives, easily match Polish grammatical patterns. However, sometimes such a borrowed noun can function as a derivational base of other words representing various lexical categories, which proves that such a lexical item is already highly assimilated in the target language. For example, the word high can function in English both as a noun (denoting the state of intoxication caused by drugs) or as an adjective (meaning intoxicated by drugs). The word has been adopted as haj in Polish slang, but only in its nominal meaning. To express the meaning of the English adjective, Poles use the prepositional phrase na haju or the past participle form nahajowany. Another word derived from haj is the reflexive verb nahajowa° si¿ ª to get high. Similarly, the word blant gave birth to derivatives such as blantowiec, blantoholik ª a habitual marijuana smoker or blancina ª a weak marijuana cigarette, and the word baka – to the verb baka° ª to smoke marijuana. In this way, we are sometimes faced with a whole family of words belonging to various lexical categories derived from one loanword. Semantic borrowings: Analogous extensions of meaning One of the methods of creating vocabulary involves a semantic change as the result of which new meanings are ascribed to already existing words. The process or its product can be called an extension (Crystal, 1998: 145) although sometimes such words are also referred to as neosemanticisms. A word with quite an ‘innocent’ basic meaning may denote, for instance, a particular kind of drug in slang. Usually, the process of extension takes place when there is some sort of similarity between the object, process or phenomenon which the word denotes in its first meaning and the object, process or phenomenon which it is to denote in its new meaning. For instance, the meaning of snow has been extended in slang so that the word designates cocaine. It can be easily seen that snow and cocaine are similar to each other in

9612Arabski.indb 101

13/01/2006 08:44:06

102

Cross-linguistic Influences in the Second Language Lexicon

being white and most often having a powdery form. On the other hand, cocaine could be compared on the same basis to sugar, salt, flour, and a lot of other substances with the same effect. However, to the author’s best knowledge, none of the previously mentioned words has come to denote cocaine. It means that although the choice of a word whose meaning is to be extended is partially motivated, it is also characterised by a certain degree of arbitrariness. It is noticeable that many Polish slang words have both a similar basic meaning and slang meaning as English slang words: koks trawa ïnieg sól strzaÙ zielsko podróČ Ùadowa° waln­°

ª cocaine (ENG. coke) ª marijuana (ENG. grass) ª cocaine (ENG. snow) ª heroin (ENG. salt) ª a quick injection (ENG. shot) ª marijuana (ENG. weed) ª an experience of hallucinations (ENG. trip) ª to inject drugs (ENG. to load up) ª to inject drugs (ENG. to hit)

Although just a few examples are provided here, such word pairs are very numerous. This suggests that many of them are not coincidental and that the Polish slang words in question most probably are semantic borrowings rather than true neosemanticisms, as the resulting form does not widen the basic meaning of Polish words independently of any foreign influence. Borrowings belonging to this category are sometimes classified as a type of calque (Bussmann, 1996: 61), but we have decided to reserve this term for expressions of more complex structure. Calques What is meant by ‘calques’ here are literal translations of English words and expressions into Polish (Crystal, 1998: 51). Such borrowings, as well as borrowings belonging to the group of analogous extensions described above, are only possible to notice as a result of a thorough comparative analysis, because Polish expressions do not contain any foreign elements. Calques combine several distinct lexical or morphological constituents. This reduces the probability of their emergence as native Polish expressions rather than loan translations. Examples include the following expressions: biaÙa dama czerwone ptaszki

9612Arabski.indb 102

ª an illegal drug (ENG. white lady) ª barbiturates (ENG. red bird ª secobarbital)

13/01/2006 08:44:07

Influence of English on Polish Drug-related Slang

103

ª a place where drug addicts meet (ENG. needle park) ČóÙte kubraczki ª barbiturates (ENG. yellow jackets) kwaïna gÙowa ª a user of LSD (ENG. acid head) szcz¿ïliwy pyÙ ª an illegal drug in powder form (ENG. happy dust) goni° smoka ª to use heroin (ENG. to chase the dragon) przypudrowa° sobie nos ª to inhale a drug (ENG. to powder one’s nose) park igieÙ

Some analogous trends in creating slang vocabulary Many items in Polish drug-related slang cannot be classified as belonging to any of the categories of borrowings described above, and yet they are created according to similar rules or use metaphors coming from the same field. This phenomenon can also be interpreted as a result of the influence exerted by English: our hypothesis is that first several expressions of a given type are borrowed into Polish as calques, and afterwards more of the same type are created. Only the three most conspicuous analogous trends will be presented here, omitting subjects such as religious metaphors or metaphors connected with war and destruction. Shortenings One of the methods of creating slang words involves shortening of already existing words which may be neutral in the sense of their register. The method is highly productive both in English and in Polish and is used mainly to build nouns denoting various types of illegal substances. Shortenings include abbreviations as well as clippings. In the case of abbreviations, words are shortened so that only the initial letter remains. The method is very widespread in English, with almost every letter of the alphabet denoting some kind of illegal drug, for example: A ª LSD (shortened from acid), C ª cocaine, C&M ª cocaine mixed with morphine, E ª ecstasy, H ª heroin and j ª marijuana (shortened from joint). Abbreviations found in Polish slang are: C ª cocaine, H ª heroin, HMC ª a mixture containing heroin, morphine and cocaine, M1 ª morphine, MJ ª marijuana. Examples of the same derivational method functioning in Polish are less numerous, but using the letter C as an abbreviation for cocaine suggests the relationship to English, since the full name of the drug is spelled in Polish with a ‘k’. The productivity of this method is obviously limited for practical reasons. One letter can designate one type of drug only, because the terms

9612Arabski.indb 103

13/01/2006 08:44:07

104

Cross-linguistic Influences in the Second Language Lexicon

can be used in the same context and it would inevitably lead to misunderstandings if, for instance, H were to mean both heroin and hashish or M - morphine, marijuana and methedrine. The process of clipping neutral words designating drugs or other notions connected with them is a method more productive then abbreviating, and used widely both in Polish and English. English examples are usually onesyllable words and include, among others: amp ª an ampoule of a drug, barb ª barbiturate, hash ª hashish, morph ª morphine, dex ª dexedrine, hypo ª a hypodermic syringe, shrooms ª psychedelic mushrooms. The method functions similarly in Polish, although there are more clippings than in English longer than one syllable. One of the reasons behind this is that Polish clippings as a rule seemingly must have the same grammatical gender as the words they are derived from. This means, for example, that the feminine suffix ‘-a’ is added to those clippings which originate from feminine nouns. Examples found in our corpus are: cent ª a cubic centimetre of a liquid drug, hasz ª hashish, dolar ª Dolargan, hera ª heroin, kofa ª caffeine, morfa ª morphine, meska ª mescaline, amfa or feta ª amphetamine, haluny ª hallucinations caused by drugs, narki or tyki (shortened non-slang word narkotyki) ª illegal drugs. In both languages, clipping involves removing the end of the shortened word in most cases. However, sometimes the beginning may also be removed as the examples of shrooms in English as well as feta and tyki in Polish show us. Colours When examining the Polish or English corpus one can hardly fail to notice a great abundance of names containing colour adjectives present in both languages. In many instances, the colour adjective used in an expression is closely related to the colour of the illegal substance denoted, which is often the case in ‘brand names’ of drugs. Nonetheless, sometimes a phrase becomes so popular that it begins to denote a particular type of drug in any form indiscriminately of its actual colour. Most Polish expressions of this type are not calques, but bear considerable stylistic similarity to English slang items. Table 9.1 presents a few examples of expressions connected with four colours. Personifications Drugs are often referred to as if they were human beings. One type of personification involves the use of proper names when referring to drugs. English examples of such expressions include: Adam ª MMDA, Eve ª MDEA, Oliver ª amphetamine, Mary Jane, Mary Ann and Johnny

9612Arabski.indb 104

13/01/2006 08:44:08

Influence of English on Polish Drug-related Slang

105

Table 9.1 Examples of slang expressions involving colours Colour

English

Polish

black

black diamonds ª amphetamine black bomber ª Benzedrine Black Heart ª a “brand name” of heroin

czarne kakao ª cocaine czarna msza ª an occasion of taking drugs together czarny koÚ ª a drug dealer

blue

blue cheer ª LSD blue cloud ª Amytal Blue Thunder ª a “brand name” of heroin

bÙ¿kit królewski ª LSD

red

red dragon ª LSD red devils ª amphetamines

czerwone ptaszki ª barbiturates czerwony krzyČyk ª morphine

white

white cross ª methamphetamine White Cloud ª a “brand name” of cocaine white knight ª a cocaine user

biaÙy krzyČyk ª cocaine biaÙa wst­Čka ª morphine biaÙa ïmier° ª an illegal drug

ª marijuana. Polish examples contain both Polish and English names, often being lexical borrowings: Lucy ª LSD, Charlie ª cocaine (both used in English as well), marycha, maryïka or Maria ª marijuana, Majka ª morphine, Dorota ª Dolargan, Ewka ª cocaine, helena ª heroin. An interesting rule seems to be that the Polish items of this type not borrowed from English seem to be exclusively female names. The reason for this might be the fact that most non-slang names of drugs in Polish are feminine nouns. The other type of personification, also encountered frequently in the corpus, does not involve the use of names per se, but consists in using other, often intimate descriptions referring to people. English examples are: white lady ª heroin or cocaine, blue lady ª very potent heroin, the chief ª LSD, boy, brother ª heroin, girl ª cocaine, silver girl ª a hypodermic syringe, rainy day woman ª a cigarette containing marijuana, while Polish gives us the following examples: siostra ª morphine, ojciec ª

9612Arabski.indb 105

13/01/2006 08:44:09

106

Cross-linguistic Influences in the Second Language Lexicon

marijuana, wesoÙy chÙopiec , zausznik ª an illegal drug, biaÙa czarodziejka ª cocaine, marzyciel ª a hallucinogenic drug.

Conclusions As can be seen from the numerous examples, the influence of English on Polish drug users’ slang is both widespread and deep and by no means limited to adopting English expressions into Polish, although this particular aspect is most easily noticed. Lexical borrowings are only the tip of the iceberg since a thorough analysis reveals other, more subtle traces of English influence. These are only possible to discover after comparing numerous expressions from Polish and English slang. This has several implications. First of all, what is true for drug slang may also be true for other realms of the Polish language. Since this type of slang covers only a very limited field of human activity, it is relatively easy to discover calques and semantic borrowings in drug-related slang as the corpus to examine is small. The traces of English that we usually notice in standard Polish or in specialised language belonging to larger semantic fields are lexical borrowings. Perhaps other realms of Polish may also abound in semantic borrowings and calques more than we realise, only it is difficult to notice them, as no thorough comparative analysis of such a huge amount of data is presently possible. Moreover, how many trends in creating neologisms, such as building metaphors for a given realm using imagery from another one, are actually imported? Secondly, there is a question of how all these borrowings and imported methods for vocabulary creation have permeated Polish slang. Some of the borrowings can be explained by the spread of the widely understood American and British youth culture. For example, Lucy obviously comes from the Beatles’ song ‘Lucy in the Sky with Diamonds’, which was supposed to describe the effects of LSD although the authors denied it. Brown sugar is also the title of a song by the Rolling Stones and used there as a nickname for a dark-skinned girl. Similarly, the slang term siostra may be connected with the very explicit song ‘Sister Morphine’ by the same group. The calque of the expression to powder one’s nose probably emerged as a result of ‘Pulp Fiction’, as the film has enjoyed great popularity in Poland. Such a cultural influence, however, does not account for everything. It is generally believed that borrowings from one language into another are introduced by bilingual speakers of both the languages in question (MaÚczak-Wohlfeld, 1995: 15). Who are the bilingual speakers who bring

9612Arabski.indb 106

13/01/2006 08:44:10

Influence of English on Polish Drug-related Slang

107

English borrowings into Polish drug slang? Probably not the heroin addicts of the kind that we can see at any large railway station. Rather, it is likely young, educated Poles who are speakers of English, indicating that occasional, ‘recreational’ drug use is a widespread phenomenon in this social group. Or perhaps equally, it is the people involved in production, import and sale of drugs who also propagate English borrowings, using English as a good ‘marketing’ tool due to its high prestige. Some of the above mentioned problems related to drug slang cannot be resolved by purely linguistic research, as the issues involved border on sociology or even criminology. However, the author believes that drug slang is not something linguists should shun and ignore. It is a part of our language, just like drug use is a part of our social reality, and whatever people’s opinions on the phenomenon might be, it will not disappear simply because we avoid talking about it. Knowledge of drug-related slang may contribute to helping physicians and other health care professionals who work with drug abusers and addicts or police and customs officers who are fighting drug crime. On the other hand, the creativity, ingenuity and even humour found in this realm of language make it a very interesting subject of study for a linguist. References Ayto, J. and Simpson, J. (1993) The Oxford Dictionary of Modern Slang. Oxford and New York: Oxford University Press. Bussmann, H. (1996) Routledge Dictionary of Language and Linguistics. London and New York: Routledge. Crystal, D. (1988) The Cambridge Encyclopedia of Language. Cambridge and New York: Cambridge University Press. Crystal, D. (1998) A Dictionary of Linguistics and Phonetics. Oxford and Malden: Blackwell Publishers Ltd. Czeszewski, M. (2001) SÙownik slangu mÙodzieČowego. PiÙa: Ekolog. Górnicz, M. (2000) Elementy angloj¿zyczne w tekstach medycznych. Poradnik J¿zykowy 2/2000. Johnson, P.N. (ed.) (1993) Dictionary of Street Alcohol and Drug Terms. Columbia: School of Medicine, University of South Carolina. MaÚczak-Wohlfeld, E. (1995) Tendencje rozwojowe wspóÙczesnych zapoČyczeÚ angielskich w j¿zyku polskim. Kraków: Universitas. Matusiak, J. (1997) Polskie sÙownictwo komputerowe. Poradnik J¿zykowy 1/1997. Michalewski, W. (1992) Mistycy i narkomani. Ethos. Moch, W. (2002) SÙownictwo swoiste polskiej subkultury hiphopowej. J¿zyk Polski, 3/2002. Ociepa, R. (2001) O wspóÙczesnych anglicyzmach w nazewnictwie zawodów i stanowisk sÙuČbowych. Poradnik J¿zykowy 2001/6. Ostrowska, Z. (1983) SÙownik niektórych poj¿° z kr¿gu narkomanów. In R.R. D­browski (ed.) Narkomania znakiem czasu. Warszawa: Wydawnictwo ‘Znaki Czasu’.

9612Arabski.indb 107

13/01/2006 08:44:11

108

Cross-linguistic Influences in the Second Language Lexicon

Rosiek, B. (1997) Pami¿tnik narkomanki. Katowice: Krajowa Agencja Wydawnicza. Spears, R.A. (1996) NTC’s Dictionary of American Slang and Colloquial Expressions. Lincolnwood: NTC Publishing Group. St¿pniak, K. (1993) SÙownik tajemnych gwar przest¿pczych. London: Plus. Thorpe, T. (1990) The Dictionary of Contemporary Slang. New York: Pantheon Books.

9612Arabski.indb 108

13/01/2006 08:44:12

Part 3

Lexical Transfer in Language Processing

9612Arabski.indb 109

13/01/2006 08:44:13

9612Arabski.indb 110

13/01/2006 08:44:13

Chapter 10

Why Money Can’t Buy You Anything in German: A Functional-Typological Approach to the Mapping of Semantic Roles to Syntactic Functions in SLA MARCUS CALLIES Introduction Native speakers of English show a very subtle knowledge of the lexical properties of verbs. For instance, they are aware of the possible occurrences of a verb with a range of combinations of arguments and adjuncts in various syntactic contexts such as transitivity alternations (1–2), or the deletion of arguments (3–4): (1) (a) They protested against the law. (b) They protested the law. (2) (a) He washes the new shirts. (b) The new shirts wash well. (3) (a) He broke the window with a stone. (b) The window broke. (4) (a) John eats tomato salad. (b) John eats. These examples show that native speakers’ knowledge of the variability of the argument structure of a verb goes beyond what is commonly ascribed to the respective verb in a subcategorisation frame as used in Generative Grammar. Levin (1993) presents a large-scale investigation of the behaviour of English verbs, analysing them with respect to the specific syntactic behaviour they show, so-called alternations. These are language-specific changes in the realisation of the argument structure of a verb as exemplified in (1)–(4) above. Since English shows a rather wide range of alternations, it is by no means clear to what extent learners of 111

9612Arabski.indb 111

13/01/2006 08:44:14

112

Cross-linguistic Influences in the Second Language Lexicon

English as a foreign language are aware of these possibilities, let alone are competent in using them for specific purposes. Generally speaking, the acquisition of argument structure and its interrelationship with the syntax-semantics interface in a second language (L2), as well as the influence of the native language (L1) in that process is a relatively underexplored field in second language acquisition (SLA) research (Kortmann, 1998; R. Hawkins, 2001). In the traditional approach to the semantic description of verbal arguments known as Case Grammar (Fillmore, 1968), each argument is assigned a special participant role in relation to the verb, i.e. the way in which the referent of an NP contributes to the state, action or situation described by the sentence. These participant roles have also been called deep cases or case roles (Fillmore, 1968), semantic and thematic roles. They are not inherent properties of NPs but are relational notions and depend on their occurrence in the context of a given sentence. It is the verb that controls the range of semantic roles it takes, imposing selectional restrictions on the semantic characteristics of its arguments. Although the notion of semantic role is relational and there is no oneto-one relationship between grammatical function and semantic role, there are typical matchings. It has frequently been observed that crosslinguistically, languages show a preference for the grammatical subject of a sentence typically taking the semantic role of AGENT or related animate semantic roles such as EXPERIENCER and RECIPIENT, whereas the object typically carries the role of PATIENT or THEME (Comrie, 1989: 107). This also holds for English and German. However, a contrastive analysis of the two languages reveals that in English a much broader range of different semantic roles can occur in subject position, whereas German is much more restricted in that respect (cf. Rohdenburg, 1974; Hawkins, 1981, 1986): (5) (a) California grows the best oranges. (b) * Kalifornien wächst die besten Orangen. (c) In Kalifornien wachsen die besten Orangen. (6) (a) The 3rd of October celebrates the day of German reunification. (b) * Der 3.Oktober feiert den Tag der Deutschen Einheit. (c) Am 3. Oktober wird der Tag der Deutschen Einheit gefeiert. (7) (a) 69 million dollars bought him the election. (b) * 69 Millionen Dollar kauften ihm die Wahl. (c) Mit 69 Millionen Dollar (er)kaufte er sich die Wahl. (8) (a) A bus crash in Austria has injured 24 people. (b) * Ein Busunfall in Österreich hat 24 Personen verletzt. (c) Bei einem Busunfall in Österreich wurden 24 Personen verletzt.

9612Arabski.indb 112

13/01/2006 08:44:14

Why Money Can’t Buy you Anything in German

113

On the basis of a traditional list of semantic roles (e.g. Saeed, 2003: 149ff.), the underlined subject NPs in the English sentences (5a)– (8a) above may be assigned the roles of LOCATION, TEMPORAL, INSTRUMENT and CAUSE, respectively. They can be considered noncanonical or marked because they are inanimate subjects which are used nonmetaphorically and non-metonymically, lacking features typical of animate subjects such as volition, intention and sentience. Equivalent German translations of the English sentences, retaining both the subject NP and the verb, are highly unidiomatic if not ungrammatical, as indicated by the asterisks in (5b)–(8b). From that it follows that with semantically equivalent, i.e. nearsynonymous transitive verbs, English allows a larger inventory of nonagentive semantic roles to occupy the subject position than German: the German verbs show more severe selectional restrictions than their English counterparts. In view of the detailed studies on the semantic diversity of subjects in the two languages (Hawkins, 1981, 1986; Plank, 1983; Hansen, 1987), it is surprising that there are very few substantial empirical studies that investigate potential problems that may arise for German learners of English as a foreign language. This paper considers the mapping of semantic roles to syntactic functions in English and German and intends to shed light on the following questions: How are marked subjects in English accepted by German learners? What strategies do the learners employ when translating those sentences into idiomatic German?

The Thematic Role Hierarchy In English and other languages, non-agentive roles may occur in subject position either by omission of some roles and the shifting of other roles as a reaction to the changed constellation, or by deliberate choice of the speaker, often accompanied by a change in verbal voice (Saeed, 2003: 154): (9) (a) David opened the door with a key. (b) The key opened the door. (c) The door opened. In (9a) the NP David is the agent. This information is omitted in (9b), and the NP carrying the INSTRUMENT-role (the key) moves into subject position. In (9c), the INSTRUMENT is deleted and the former object NP the door, taking the role of THEME, becomes the grammatical subject of the sentence. The mapping of semantic roles to syntactic functions is generally thought to be constrained by a functionally motivated thematic role hierarchy which

9612Arabski.indb 113

13/01/2006 08:44:15

114

Cross-linguistic Influences in the Second Language Lexicon

is considered to be universal and implicational. Various versions of such a hierarchy have been proposed in the literature (e.g. Dik, 1997; Fillmore, 1968; Givón, 2001; for an overview see Levin & Rappaport-Hovav, 1996: 43f.). The thematic hierarchy is essentially ‘a ranking order of the various semantic case-roles according to the likelihood of their […] occupying the pragmatic case-roles of subject or direct object in simple clauses’ (Givón, 1984: 139). Given a possible hierarchy (cf. Givón, 2001: 200): AGENT > RECIPIENT > PATIENT > LOCATION > INSTRUMENT > OTHERS there are two ways of reading it: (1) subjects are selected from left to right, the leftmost role being the most likely one to be realised as subject, and (2) if a language allows e.g. LOCATION to occupy subject position, it will also allow all the roles to the left of this role (i.e. all the higher ranked roles) to appear as subject in that language. It has to be noted that the classification of verbal arguments based on a varying list of semantic roles faces severe theoretical and practical problems. Saeed (2003) gives a concise summary of the criticism on the traditional approach. First of all, there is no agreement among linguists on a limited inventory of semantic roles. Being overspecific, one would have to identify an individual role for each verb, resulting in verb-specific semantic roles such as RUNNER, DANCER or LISTENER. This would of course be counterproductive to our intention of observing generalisations among semantic roles and to track down principles of the mapping of roles onto grammatical relations. Another problem is that even for the trained linguistic observer, there are considerable difficulties to clearly identify the semantic role of an NP in a specific context: (10) Ewa comes from Poland. (11) John weighs 180 pounds. What kind of semantic roles shall we assign to the underlined NPs in the sentences above? In many cases, the proposed inventory of roles does not seem to be sufficient to incorporate a specific semantic relation between a verb and its argument. A reaction to these shortcomings is the proposition of so-called generalised semantic roles, e.g. macro-roles like ACTOR and UNDERGOER (Foley & VanValin, 1984) or proto-roles (Dowty, 1991). Dowty adopts a prototypeapproach in that he does not consider thematic roles to be primitives, but assumes both an AGENT and a PATIENT proto-role consisting of more specific properties such as volition, sentience and movement. This approach is more flexible in the identification of individual roles and accounts for

9612Arabski.indb 114

13/01/2006 08:44:16

Why Money Can’t Buy you Anything in German

115

phenomena such as fuzziness and gradual membership. Similarly to Dowty’s proto-roles, Comrie recognises that the relations among individual sentence participants are ‘not so much a set of discrete semantic relations, but rather a continuum’ (1989: 59). He proposes the classification of sentence participants according to their position on a continuum, depending on the degree of control that one NP-referent has over another NP-referent in relation to the verb. For the present purposes, instead of attempting to clearly identify individual semantic roles, I will broadly distinguish between animate (AGENT, PATIENT, EXPERIENCER, RECIPIENT) and inanimate roles (THEME, INSTRUMENT, LOCATION, TEMPORAL, CAUSE, FORCE, etc.).

Markedness and the Predictability of Language Transfer Throughout this paper, the term language transfer will be used in the sense of cross-linguistic influence, defined by Odlin as ‘the influence resulting from similarities and differences between the target language and any other language that has been previously (and perhaps imperfectly) acquired’ (1989: 27), thereby incorporating positive transfer as well as interference, avoidance and overproduction. With regard to making predictions as to the learnability of linguistic structures and their (non-) transferability from a learner’s native language, Kellerman (1979) initiated a re-evaluation of the notion of transfer. In his view, the learner is an active decision-maker on what linguistic structures may be transferable into the second language. Transfer is seen as a cognitive process subject to three constraints: (1) the learner’s perception of the distance between the first and the second language (what Kellerman calls the learner’s ‘psychotypology’); (2) the learner’s perception of the degree of markedness of a potentially transferable item in the L1 (perceived transferability); and (3) the nature of the learner’s knowledge of the target language. It is important to note that Kellerman’s approach to transfer underlies a psycholinguistic rather than linguistic understanding of markedness in terms of psychological and perceptual complexity, not structural complexity (Rutherford, 1982: 92). Psychotypological aspects of the cognitive organisation of a learner’s interlanguage, such as perceived transferability, are essentially based on subjective learner perception, a phenomenon hard to objectify and with little predictive power for the researcher. Applying the concept of typological markedness seems a much more practicable solution, especially since it enables us to objectify language distance along typological properties. Eckman (1977) argued that on the basis of a contrastive analysis of two languages (L1 and L2) and the inclusion of the concepts of

9612Arabski.indb 115

13/01/2006 08:44:16

116

Cross-linguistic Influences in the Second Language Lexicon

typological markedness and cross-linguistic influence (or language transfer), it should be possible to predict areas of difficulty for an L2 learner. In a nutshell, Eckman’s Markedness Differential Hypothesis (MDH) claims that L1 structures that are different from L2 structures and typologically more marked will not be transferred, whereas L1 structures that are different from L2 structures and typologically less marked are more likely to be transferred. Thus, predictions can be made to both the order and difficulty of linguistic features in the acquisition process: Less marked structures will be acquired first or without difficulty, more marked structures are expected to be acquired later or with greater difficulty. In contrast to the concept of perceived similarity, the functional-typological approach to SLA (Ramat, 2003) has significantly more predictive power since potential difficulties in the L2 learning process are identified not merely on the basis of similarities and differences derived from a contrastive analysis of two languages, but through a combination of the concepts of typological markedness and cross-linguistic influence.

Previous Research Following Fillmore’s theory of Case Grammar and its significance for foreign language learning and teaching, there is a considerable amount of literature from the 1970s providing contrastive descriptions of the set of semantic roles that both English and German allow in subject position (Dirven, 1979; König & Nickel, 1970; Radden & Dirven, 1977, 1981; Zimmermann, 1972). Rohdenburg (1974) offers a vast collection of authentic material on semantically unusual, i.e. marked subjects in English, a phenomenon he calls ‘secondary subjectivisation’. Rohdenburg examined – among other things – German translations of 16 English sentences containing marked subjects. Participants in his survey were advanced German learners of English (high school and university students) and university staff members. Rohdenburg identified two commonly used strategies for translating English marked-subject sentences into German: (1) the subject is retained and the verb is changed in order to adapt to the needs of the sentence as in (12b) below, if possible by using a particle verb to convey the intended semantic meaning by morphological means (12b), and (2) the verb is retained and the subject is changed, the former subject usually being realised as an adverbial/adjunct as in (12c): (12) (a) Other serious fighting killed 5 and injured 12. (b) Weitere schwere Gefechte forderten 5 Tote und 12 Verletzte. (c) Bei weiteren schweren Gefechten wurden 5 Personen getötet und 12 verletzt.

9612Arabski.indb 116

13/01/2006 08:44:17

Why Money Can’t Buy you Anything in German

117

As reported by Kellerman (1979, 1983), Jordens (1977, 1978) found that with near-synonymous and structurally related sentences (raising, mediopassive constructions and marked subjects, i.e. secondary subjectivisation), advanced Dutch learners of German showed different levels of acceptability, i.e. sentences with canonical subjects were significantly better accepted than those with marked subjects. Further evidence for learnability comes from the basic variety. Klein and Perdue (1992) have identified semantic constraints that influence the formfunction mapping in the early stages of L2 acquisition. These relate to the thematic roles of the participants and to the degree of control that one NPreferent with a certain semantic role may have over another NP-referent (cf. also the continuum of control in Comrie, 1989: 59). According to what Klein and Perdue call the ‘controller principle’, ‘the NP with the highest degree of control comes first’ (1992: 49). Broadly speaking, this means that NPs having a higher degree of control over the other NP-referents in a sentence, consequently being more agentive, appear in subject position and are acquired earlier.

Research Hypotheses Givón (1991) proposes a definition of markedness that includes the following correlates: structural complexity, frequency distribution, and cognitive complexity. According to these criteria, marked elements are structurally more complex, less frequent and therefore cognitively more salient. They require more attention, mental effort and cause more processing time for the recipient. Givón’s understanding of markedness is extremely useful in the present context since it integrates the notion of cognitive complexity, reminiscent of Naturalness Theory (Dressler et al., 1987) which postulates a correlation between markedness and the cognitive-physiological complexity of linguistic units: marked structures require more cognitive work in order to be processed. As far as the mapping of semantic roles to syntactic functions is concerned, a contrastive analysis shows that English is not only highly marked when compared with German, but also in a cross-linguistic perspective: it allows an unusually broad range of different non-agentive semantic roles to surface as subject (Comrie, 1989: 75, 79; Blake, 1987: 315, 319f.). Consequently, English clearly exhibits a higher degree of typological markedness. Similarly to raising constructions, there is a larger distance between form and function in English marked-subject sentences of the type given in (5a)–(8a) above, since the logical subject of these sentences is

9612Arabski.indb 117

13/01/2006 08:44:18

118

Cross-linguistic Influences in the Second Language Lexicon

not the grammatical subject. They are functionally and semantically more complex, less transparent and less explicit, causing more cognitive cost and requiring more processing time for the recipient in terms of the analysability and decoding of the form-function relation. In view of the constraints on the learnability of linguistic structures and their (non-) transferability from a learner’s native language outlined earlier the following research hypotheses can be formulated: •

•

German learners of English will show comparatively low acceptability rates for English marked subject constructions, which are expected to be underrepresented if not avoided in their interlanguage due to interference from L1, which lacks or disprefers those structures, and higher typological markedness (Kortmann, 1998: 141, 156). Learners will accept near-synonymous and structurally related sentences in which a marked subject-NP is realised as an adjunct significantly better.

Experimental Design and Methodology Forty-four mostly undergraduate students of English language and literature participated in the study1. The data were elicited by using acceptability judgements, error identification, and an English/German translation task. Studies on research methodology in SLA suggest that the use of metalinguistic judgements is effective but controversial. The data obtained cannot be used uncritically and should be validated by other data collection methods. There is ongoing dispute about the question whether judgements are really indicative of a learner’s grammatical competence or whether they actually provide performance data. It now seems reasonable to assume that they ‘do not provide direct access to the learner’s competence and they are indeed performance data, but they do give us an insight into the learner’s competence by telling what is possible and what is not acceptable in interlanguage’ (Gass, 1994: 306f.). Despite the criticism, metalinguistic judgements are a valuable data collection method if used cautiously. They are open to the influence of extralinguistic variables such as task-related factors which may influence a learner’s judgement, notably parsing strategies, the context and ordering of test items, or the degree of the informant’s metalinguistic training. However, these unwanted side-effects can be controlled by thoughtfully selecting the test items and participants and by carefully designing and administering the tasks.

9612Arabski.indb 118

13/01/2006 08:44:18

Why Money Can’t Buy you Anything in German

119

In a non-timed data collection session, students were asked to carry out three tasks. In the first task, participants evaluated sentences with a nonagentive subject as to their acceptability in a given context on a seven-point scale. Additionally, they were asked to indicate reasons for unacceptability of a sentence, if applicable. In the second task, the students judged nearsynonymous sentences in which the marked NPs presented earlier were realised as adjuncts (see the Appendix for a list of all 32 test sentences that were used in the study). Finally, participants had to translate 10 sentences selected from the first judgement task into idiomatic German. Potential task-related factors were minimised by varying the presentation order of the test items to avoid sequencing effects, and by including distractors in order to control the attitude and behaviour of the informants and to conceal the actual research goal. The test items used were authentic and only slightly adapted sentences taken from a self-compiled corpus based on a three-month analysis of several online sources, mainly The Times online edition (London), and CNN online (New York). The investigation was restricted to transitive action verbs which normally take animate semantic roles such as AGENT, EXPERIENCER or RECIPIENT as subject, but which also allow other semantic roles in subject position. Special interest lay on the subjectivisation of non-subcategorised arguments, so-called oblique subject alternations as exemplified in (5a)–(8a) above, i.e. cases involving ‘verbs that have “agent” subjects, but that alternatively may take as subjects noun phrases that can be expressed in some type of prepositional phrase when the verb takes its canonical “agent” subject. […] When the verbs take the oblique subject, the “agent” is no longer expressed’ (Levin, 1993: 79). The study excluded the so-called middle alternation or medio-passive construction (Levin, 1993: 25), exemplified in (13), and the locative alternation (Levin, 1993: 50) as in (14): (13) (a) The meat cuts easily. (b) His latest book sells well. (14) (a) The square is seething with people. (b) The garden is swarming with bees.

Results and Discussion What will be presented in the following sections are first quantitative and qualitative results. It has to be noted that the results need to be interpreted with caution, since there are no data from a sufficiently large control group available yet.

9612Arabski.indb 119

13/01/2006 08:44:19

120

Cross-linguistic Influences in the Second Language Lexicon

Acceptability rates Generally speaking, there were rather high standard deviation scores for the majority of test sentences, possibly indicative of a heterogeneous learner group. Nevertheless, in 50% of the test items (8 out of 16), the German learners accepted the sentences in which the respective NP was realised as an adjunct significantly better than those in which it appeared as a marked subject. Highly significant differences between the mean acceptability rates (p 90%

to break the heart (Russian) have a thorn in the heart (Japanese)

Source: FOOD Target: PROBLEMS

something is hard to swallow etwas ist schwer zu verdauen

Percentage

Examples

~ 90%

something is hard to digest (Czech) something is hard to chew (Japanese)

Source: OBSTACLES ON ONE’S WAY Target: PROBLEMS

to be on a bumpy road einen steinigen Weg vor sich haben

Percentage

Examples

~ 80%

to walk a thorny/steep path (Turkish) the way to paradise is paved with cactuses (Arabic)

9612Arabski.indb 201

13/01/2006 08:45:28

202

Cross-linguistic Influences in the Second Language Lexicon

Source: HOT OBJECTS Target: DANGER

to burn one’s fingers sich die Finger an etwas verbrennen

Percentage

Examples

~ 70%

to burn one’s fingers on something (Finnish) to be afraid of hot butter (Arabic)

Source: UPRIGHT POSITION Target: STRENGTH

to help someone back on his feet jemandem auf die Beine helfen

Percentage

Examples

~ 70%

to help somebody to stand on his feet (Greek) to pull someone up on his hands (Turkish)

Source: DEEP WATER Target: DANGER

to keep one’s head above water sich über Wasser halten

Percentage

Examples

< 40%

to stay above water (Spanish) to keep the nose above water (Swedish)

Considering these observations it seems worthwhile to follow up this pilot study along several lines: • •

by testing more informants form the same languages in order to exclude the possibility of chance results by triangulating these interview data with more traditional ones such as dictionary data and recognition tests

9612Arabski.indb 202

13/01/2006 08:45:29

Metaphorical Transferability • •

203

by collecting data for more metaphorical expressions from the same domains in the languages investigated and, of course, by extending the range of domains as well as languages.

Beyond the tendencies observed in terms of subject matter, it is perhaps an at least as relevant result that we have at our disposal a tested method of data collection for everyday idioms, on which further research can build. Note 1. This contribution focuses rather on methodological issues; the few results mentioned on pp. 200ff are based on a preliminary data analysis. For a fuller presentation and discussion of results cf. Zimmermann (2005).

References Barcelona, A. (2001) On the systematic contrastive analysis of conceptual metaphors: Case studies and proposed methodology. In M. Pütz, S. Niemeier and R. Dirven (eds) Applied Cognitive Linguistics, Vol. 2 (pp. 117–46). Berlin: Mouton de Gruyter. Barcelona, A. (ed.) (2003) Metaphor and Metonomy at the Crossroads – A Cognitive Perspective. Berlin: Mouton de Gruyter. Boers, F. (2000) Metaphor awareness and vocabuary retention. Applied Linguistics 21 (4), 553–71. Cameron, L. and Low, G. (1999) Metaphor. Language Teaching 32 (2), 77–96. Fernando, C. and Flavell, R. (1981) On Idiom: Critical Views and Perspectives. Exeter: University of Exeter Press. Kellermann, J. (1978) Giving learners a break: Native language intuitions as a source of predictions about transferability. Working Papers on Bilingualism 15, 59–92. Kövecses, Z. (1995) The ‘Container’ metaphor of anger in English, Chinese, Japanese and Hungarian. In Z. Radman (ed.) From a Metaphorical Point of View: A Multidisciplinary Approach to the Cognitive Content of Metaphor (pp. 117–45). Berlin: Mouton de Gruyter. Lakoff, G. and Johnson, M. (1980) Metaphors We Live By. Chicago: University of Chicago Press. Lakoff, G. and Johnson, M. (1999) Philosophy in the Flesh. The Embodied Mind and its Challenge to Western Thought. New York: Basic Books. Mühlhäusler, P. (1995) Metaphors others live by. Language and Communication 15 (3), 281–7. Niemeier, S. (2003) Straight from the heart – metonymical and metaphorical explorations. In A. Barcelona (ed.) Metaphor and Metonomy at the Crossroads – A Cognitive Perspective (pp. 195–213). Berlin: Mouton de Gruyter. Seliger, H.W. and Shohamy, E. (1989) Second Language Research Methods. Oxford: Oxford University Press. Van Brabant, K. (1986) The cross-cultural study of metaphor: Towards a theoretical and practical analysis of figurative speech. Communication and Cognition 19 (3-4), 395–434. Zimmermann, R. (2002) Advanced foreign language vocabulary – a closer look at word-formation and idioms. In J. Arabski (ed.) Time for Words. Studies in Foreign Language Vocabulary Acquisition (pp. 83–95). Frankfurt am Main: Lang.

9612Arabski.indb 203

13/01/2006 08:45:30

204

Cross-linguistic Influences in the Second Language Lexicon

Zimmermann, R. (2005) Cross-cultural idioms? An empirical study. In A.J. Schuth, C. Horner and J.J. Weber (eds) Life in Language: Studies in Honour of Wolfgang Kühlwein (pp. 83–95). Trier: Wissenschaftlicher Verlag Trier.

Appendix 1: Test Run Questionnaire (English Version) PHILIPPS-UNIVERSITÄT MARBURG FACHBEREICH FREMDSPRACHLICHE PHILOLOGIEN INSTITUT FÜR ANGLISTIK UND AMERIKANISTIK Data Collection 2001/2002 You are taking part in a research project of the Institute for English and American Studies. Your participation in this data collection will be kept anonymous and the data gained will be used only for research and teaching purposes. Please do not speak with other students about the content of this data collection because they may also be test candidates. Thank you in advance for your willingness to be a part of this project. Please read the following instructions carefully. This data collection is not intended to be a (foreign language) test. What we are interested in is your competence in your respective native language, especially in the field of imageful language. In the course of the interview you will be given short dialogue sequences. Your task is to answer a few questions as to the expression set in bold letters. In case you have any questions or comments or problems of understanding please do not hesitate to interrupt the interviewer and ask questions. Example: A: ‘I’m afraid to go to the doctor right now. I think I’ll wait a bit longer.’ B: ‘I think you shouldn’t hide your head in the sand.’ HIDE ONE’S HEAD IN THE SAND to avoid/postpone a difficult decision, not wanting to admit something DEN KOPF IN DEN SAND STECKEN einer Sache aus dem Weg gehen, etwas nicht wahrhaben wollen

9612Arabski.indb 204

13/01/2006 08:45:31

Metaphorical Transferability

205

Questions: 1.

Is there a similar expression in your native language using a similar idea or image? Please write it down (in your writing system).

2.

Can you please translate the expression word-by-word, writing it down?

3.

Can you write your words for HEAD, SAND and HIDE as you pronounce them in your language, using Roman letters?

4.

Can you think of other imageful expressions in your language using the concepts HEAD, SAND and HIDE? Please give English translations.

5.

Can you think of other imageful expressions to talk about AVOIDING OR POSTPONING A DIFFICULT DECISION in your language? Please give English translations

Appendix 2 3. A: ‘Did you have a good time with your host family?’ B: ‘Yes, very much so. I was given a warm welcome.’ A WARM WELCOME/RECEPTION – a friendly welcome/reception Is there a similar expression in your native language using a similar idea or image? Please write it down (in your writing system). Can you please translate the expression word-by-word, writing it down? Can you write your words for WARM and WELCOME as you pronounce them in your language, using Roman letters? Can you think of other imageful expressions in your language using the concepts WARM and WELCOME? Please give English translations.

9612Arabski.indb 205

13/01/2006 08:45:31

206

Cross-linguistic Influences in the Second Language Lexicon

Can you think of other imageful expressions to talk about A FRIENDLY RECEPTION in your language? Please give English translations.

4. A: ‘Has your cat had her kitten yet?’ B: ‘Yes. It really warms my heart when I see them playing.’ TO WARM SOMEONE’S HEART – to affect someone’s feelings positively Is there a similar expression in your native language using a similar idea or image? Please write it down (in your writing system). Can you please translate the expression word-by-word, writing it down? Can you write your words for WARM and HEART as you pronounce them in your language, using Roman letters? Can you think of other imageful expressions in your language using the concepts WARM and HEART? Please give English translations. Can you think of other imageful expressions to talk about AFFECTING SOMEONE’S FEELINGS POSITIVELY in your language? Please give English translations

Appendix 3: Supportive Items (English) 1. EMOTIONS ARE PARTS OF THE BODY a) b) c) d) e)

my heart bleeds for someone – to feel very sorry for someone one’s heart sinks into one’s boots – to lose confidence, to get frightened to take a load off one’s mind – to feel relieved, to feel emotionally better in reserve: it made my hair stand on end – to feel frightened or shocked to rub salt into the wound – make a painful experience even more painful for someone

9612Arabski.indb 206

13/01/2006 08:45:32

Metaphorical Transferability

207

2. PROBLEMS/IDEAS ARE FOOD a) b) c)

to find something a tough nut to crack – to find it hard to solve a difficult problem to have to swallow hard – to resolve a difficult problem or situation to grasp the nettle – to tackle a difficulty boldly

3. DANGER IS DEEP WATER / STRENGTH IS WATER a) b) c) d)

to be up to one’s neck in it – to have serious financial problems, to be in big trouble to be in deep water – to be in a dangerous or serious situation, to be in trouble to pull the rug from under someone’s feet – to get someone into trouble in reserve: grist to the mill – to bring advantage to someone, to support someone’s point of view

4. PROBLEMS ARE HOT OR DANGEROUS OBJECTS a) b)

to pull someone’s chestnuts out of the fire – to get someone out of trouble to tackle/take on a hot issue – trying to solve a difficult or unpleasant problem or situation

5. PROBLEMS ARE OBSTACLES ON ONE’S WAY a) b)

to smooth someone’s path – to help someone solve a problem to put a spoke in someone’s wheel – to cause problems for someone

6. STRENGTH IS FIRM, UPRIGHT POSITION a) b)

to stand on one’s own two feet – to be independent, to manage one’s own affairs without help to lose one’s head – to panic in a crisis or difficult situation

9612Arabski.indb 207

13/01/2006 08:45:33

208

Cross-linguistic Influences in the Second Language Lexicon

Appendix 4:

9612Arabski.indb 208

13/01/2006 08:45:33

Metaphorical Transferability

9612Arabski.indb 209

209

13/01/2006 08:46:34

Chapter 16

On the Use of Translation in Studies of Language Contact JOLANTA LATKOWSKA Translation Skill as an Aspect of Bilingualism The popular perception of bilingualism is that apart from the knowledge and regular use of two languages (Weinreich, 1963), it also involves the ability to translate, i.e. express in one language (target language) the meanings conveyed in another (source language), preserving semantic and stylistic equivalences (Bell, 1991). This view is in keeping with reports of bilingual children as young as two years of age, who successfully translate for their monolingual parents, relatives and/or playmates (Harris & Sherwood, 1978; Grosjean, 1982). Cases like these have, undoubtedly, given rise to the belief that all bilinguals have translation competence, which may develop to the same degree as their proficiency in the languages concerned (Lorscher, 1991). On the other hand, there is evidence to suggest that bilinguals experience difficulty in translating and that overall fluency may not be a decisive factor in successful translation. As Grosjean (1982, 1997) rightly observes, this is because bilinguals rarely develop the same level of competence in their languages, whose use is domain specific and, as such, is seldom required to the same extent. What is more, since bilingual aphasics often retain their ability to translate into the language they cannot speak (Grosjean, 1982), which hints at the possibility that the skill has different neurolinguistic underpinning, there is every good reason to believe that translation is qualitatively different from typical linguistic production, i.e. speaking and writing. In fact, in defining translation competence theorists focus not only on its product but also on the processes involved, which, among other things, implies that translation is a skill that can be trained and investigated in terms of relevant strategies and/or competencies. This, however, is the domain of translation studies and falls outside the scope of this paper. Still, it cannot escape notice that a lot of everyday translation is undertaken by non-professional translators, who despite the fact that they have no formal training, are capable of producing translations that show a degree 210

9612Arabski.indb 210

13/01/2006 08:49:13

Translation in Studies of Language Contact

211

of semantic and stylistic equivalence. By the same token, research into child bilingualism has provided further evidence in favour of bilinguals’ natural capacity to produce translations within their range of comprehension and vocabulary (Malakoff & Hakuta, 1991: 144). This finding is in line with Toury’s (1986a, 1986b) contention that all humans have an innate predisposition to translate, which is co-extensive with bilingualism. (In his view, translation competence is a function of practice in actual translation [Lorscher, 1991: 42].) A similar stance was adopted by Lorscher (1991: 34) who in his study of oral translation by non-professionals proves convincingly that every individual who has two or more languages at his disposal also possesses a rudimentary ability to mediate between these languages. It is this natural translation ability (Harris, 1977) that is central to the present investigation.

Translation in Language Contact Research: Review of Selected Studies The fact that translation is a natural attribute of bilingualism creates a window of opportunity for its use in research into aspects of bilingualism and foreign language learning. Indeed, according to Underhill (1987), it is one of the most likely communicative activities that anyone with a reasonable command of an L2 may be required to engage in. This makes its use in bilingualism research both viable and indispensable. The research procedures applied to date vary from word translation tasks that focus mainly on the accuracy and speed of translation (cf. Kroll & Tokowicz, 2001) and thus provide insights into bilingual neurolinguistic processing and conceptual organisation (cf. Singleton, 1999; Forster & Jiang, 2001), through sentence translation, both oral and written, (Malakoff & Hakuta, 1991; Zabor, 2001; Huffines, 1991) to translation tasks involving texts above the level of the sentence. These, too, are conducted in both oral and written modes (Singleton, 1999; Malakoff & Hakuta, 1991; Lorscher, 1991). The last two types of tasks may be used to provide data on the nature of bilingual competence and interlanguage, as well as on the mechanics of cross-linguistic interaction. The rationale for using translation in this manner is best explained by Larsen-Freeman and Long (1991) who state that sentence translation calls for the decoding of the stimulus sentence and the encoding of the translation, so that subjects’ performance approximates natural speech production (Larsen-Freeman & Long, 1991: 29). To provide support for this perspective, she quotes evidence from the Swain et al. study (1974: 76), which found that the errors made in translation are essentially the same as those produced in spontaneous communication and imitation (cf.

9612Arabski.indb 211

13/01/2006 08:49:16

212

Cross-linguistic Influences in the Second Language Lexicon

Weller, 1989). Larsen-Freeman and Long warn, however, that researchers should beware of the potential methodological pitfalls that are inherent in translation, namely the consistent occurrence of errors which may be traced to strategies of translation, i.e. word-for-word rendition, rather than to the underlying linguistic competence. What needs to be noted, however, is that not everyone produces such errors. In stark contrast to these findings is Burmeister and Ufert’s (1980) analysis of the data collected from spontaneous speech and translation tasks. What they discovered was that the elicited data contained a higher level of L1 interference and avoidance, as well as structures that were rare or virtually non-existent under natural conditions. Burmeister and Ufert attributed these discrepancies to the subjects’ low L2 proficiency, which made them unable to successfully meet the demands of the task in question. Significant differences between lexical errors in translation and free composition were also found by Källkvist (1998), who established that both tasks exhibited similar types of errors. What was different was their distribution, i.e. translation contained more non-existent words and semantic flaws while free composition produced collocational errors and lacked semantic specificity. Unfortunately, the study does not look into the causes of these errors. Interestingly, a caution has also been expressed by linguists working in the realm of L1 attrition. For example, Huffines (1991) who used a sentence translation task to investigate word order rules and dative pronouns in Pennsylvania German discovered ample proof of convergence towards English in the word order task, which was not confirmed by evidence from free conversation. She interprets this as the function of a strategy to translate a sentence linearly, as presented in English (1991: 134), and observes that translation data alone may distort the overall picture of the process under investigation. By contrast, the data from the morphology task she used supported the translation results. Likewise, Yagmur (2002) observes that recent studies in the area of L1 attrition report massive linguistic loss not because of actual attrition but on account of inappropriate data collection instruments, one of them being translation from L2 into L1. More evidence against the relevance of translation has been found by Malakoff and Hakuta (1991) whose findings testify to the presence in translation of source-word order errors, and source-word intrusion errors, where a word from the source language works its way into the translation (1991: 155). Their second finding regarding translation is that in the written mode the incidence of errors is higher, which might imply that the modality is more demanding in terms of processing. Accordingly, written work contains more word order errors resulting from word-for-word rendition. It must be stressed, however, that although the incidence of literal translation is

9612Arabski.indb 212

13/01/2006 08:49:17

Translation in Studies of Language Contact

213

quite high in their studies, it does not lead to incorrect sentence structures in the vast majority of cases. Likewise, the overall rate of word-intrusion errors is low in both oral and written rendition. Consequently, on weighing up their findings, Malakoff and Hakuta (1991) conclude that when the quality of translation is affected, it is the sentence structure that suffers and not the meaning (Malakoff & Hakuta, 1991). There can be no doubt that these findings significantly increase the credibility of translation as a tool for investigating morphology and lexico-semantics rather than wordorder phenomena. Further light on the viability of the research method under discussion is shed by the results of the MLRP conducted at Trinity College Dublin (Singleton, 1991) to obtain insights into the relationship of L2 development and previous learning experience, as well as aspects of bilingual mental representations. The project used a written translation task together with a number of vocabulary eliciting methods and, overall, found no major discrepancy in results that could undermine the reliability of findings produced by translation. On the contrary, these were confirmed by the data collected through the other methods. Few literal translation errors were reported in studies of translation conducted among advanced L2 English learners by Zabor (2001) either. Although part of the evidence discussed so far seems to imply that the adverse effects of translation are practically negligible, they do nevertheless occur and as such pose a threat to the validity of research results. After all, one must not forget that written translation offers the temptation of translating and then understanding (Seleskovitch, 1976). Consequently, the search for the target language segments is performed in a form-oriented way which leads to form-induced errors (Lorscher, 1991). To counter these allegations, it is necessary to turn to the work of Fischer and Pipp (1984, in Malakoff & Hakuta, 1991), who point out that such strategies are not the only factors affecting the effectiveness of translation. In their opinion, translation proficiency is a function of an interplay between bilingual proficiency, metalinguistic maturity, which is required for monitoring the outcomes of the process, as well as the tactic(s) adopted. Given that the strategies enhance performance within limits delineated by proficiency and metalinguistic awareness, it can safely be concluded that translation performance is primarily indicative of the underlying competencies, linguistic and metalinguistic, and only secondarily of the strategy/ies used. Also of significance to this discussion are semantic incentives, which according to Hulstijn and Tangelder (1991, cited in Singleton, 1999) are the driving force behind advanced vocabulary processing. While there is no doubt that L2 learners rely on form in translation and other L2-related

9612Arabski.indb 213

13/01/2006 08:49:18

214

Cross-linguistic Influences in the Second Language Lexicon

tasks (Singleton, 1999), this seems to be connected with their low L2 proficiency rather than being an artefact of the activity performed. Studies of L2 vocabulary learning show that native speakers and learners with nearnative proficiency are more likely to focus on semantic relations rather than on purely formal and phonological features. Thus, if lexical interference obtains, it may be the result of semantic factors which are at play in natural speech production as well. Incidentally, Lorscher (1991) remarks that form-oriented translation is often used to disguise problems with comprehension, and thus may be indicative of incomplete knowledge of the source language, which is a case in point. A somewhat different picture emerges from the comparison of borrowings found in translations from English by advanced L2 German and French learners who took part in the already-mentioned MLRP (Singleton, 1999). Namely, the French learners borrowed from English twice as often as the German-learning subjects. Apart from reports of extensive cross-linguistic consultation obtained through introspection, these findings provide data which confirm that psycho-typological perceptions, i.e. the realisation that French and English share a number of cognates, may influence the incidence of word-intrusion, i.e. form-oriented, errors in translation. In this case, however, a similar rate of borrowing was found in other tasks, which proves that translation may provide empirically reliable insights into cross-linguistic processes.

The Study The primary objective of the study was to determine the extent of crosslinguistic influence in the area of L1 vocabulary as evidenced by an L2–L1 written translation task. Moreover, in the light of contradictory reports on translation, it was hoped that the study would provide information on whether or not translation produces large numbers of errors which could be attributed to translation strategies rather than to cross-linguistic phenomena such as borrowing transfer (Odlin, 1989) or backlash interference (Singleton, 1999), as well as the underlying bilingual competence.

The subjects Although the Weinreich (1963) definition of bilingualism is broad enough to include foreign language acquisition contexts (Romaine, 1995), as well as learners with less than communicative proficiency in the L2, this study is concerned primarily with Polish-English bilinguals who have developed an advanced level of proficiency in their L2, as attested to by relevant testing procedures (university entrance examinations, end-of-year examinations in

9612Arabski.indb 214

13/01/2006 08:51:21

Translation in Studies of Language Contact

215

General English, CAE and CPE), and who regularly use both their languages for communicative purposes in either natural or formal settings such as, for example, university programmes with instruction in the L2. In line with these assumptions, the research was conducted at the English Department of the University of Silesia, Poland. Its subjects were selected at random. The sample was made up of 30 students in their first, third and fifth years of study and a group of lecturing staff employed in the department. The group consisted of six lecturers. The instruments The test used in the study was a written translation task which elicited translation into Polish of 24 English sentences containing idiomatic expressions and fixed phrases as well as collocational pairs of words and single vocabulary items. All of them had semantic equivalents in the Polish language which were lexically different. The rationale for using a sentence translation task was based on Catford’s (1965) contention that equivalence of meaning between two languages cannot be established to the same degree below the level of the sentence. Despite evidence that it is the clause rather than the sentence which is the unit of translation (Bell, 1991), Catford argues that it is the sentence that provides enough contextualisation to make the meaning of its components explicit enough to guarantee their successful rendition into the target language. Sufficient contextualisation may be central to the translation of idioms whose meaning is not the sum of their components and which need to be treated as single lexical units (Hatch & Brown, 1995; McCarthy, 1990). Likewise, many collocations and set expressions are so tight that it is impossible to reformulate them without changing their meaning. This, in Hatch and Brown’s opinion, proves that they form lexical units in themselves and should be thought of and processed as single items. From the translator’s point of view, however, such expressions constitute potential pitfalls as they are liable to be translated literally. It is this potential for literal translation that was the main reason for their inclusion in the test. Implementation The subjects were required to translate the sentences included in the test as precisely as possible. They were also informed that the emphasis was on correctness and a sense of equivalence. Since the task was timed (55 mins), the subjects were forced to compare surface features and semantic attributes of both languages in circumstances when time was scarce. According to Sharwood-Smith (1983), this kind

9612Arabski.indb 215

13/01/2006 08:51:22

216

Cross-linguistic Influences in the Second Language Lexicon

of strain is conducive to cross-linguistic interaction, which is basically a measure taken to cope with the increased processing burden faced by the bilingual. In the translation context, it may induce the subjects to resort to word-for-word translation as an easy way out. Results Overall, the data obtained in this study can be divided into three broad categories, which include: • •

•

correct translations, where the exact equivalent of the item concerned was provided; literal translations (calques); this category also encompasses incomplete calques (creations, approximations), i.e. expressions which constitute literal translations in part only, and are made up of coinages, which are unique by L1/L2 standards and thus fail to conform to monolingual norms; avoidance strategies, which are attempts by the subjects to provide a translation which expresses the meaning of the relevant item in a roundabout fashion. Such indirectness may be achieved by using a single word in place of a phrase or collocational pair, circumlocution, a new coinage which is not a calque and which, in the testee’s opinion, substitutes for the item elicited by the test. What needs to be stressed is that avoidance does not result in forms that violate the norms of the L1.

As indicated in Table 16.1, most subjects experienced difficulty while translating into their mother tongue. This is evidenced by their scores for accurate translations which constitute just 69% of the collected corpus. However, the results for literal translation are relatively low and do not exceed 11%. This is because the subjects used avoidance strategies that included a predominance of circumlocution (17%). Idiomatic expressions that resembled the elicited items in meaning and yet failed to qualify as their direct equivalents were extremely rare (0.4%). Table 16.2 presents examples of the phrases which were most often calqued into Polish. Table 16.1 The results of the translation test in percentages Results

(%)

Correct translations

69

Calques

11

Avoidance

19

9612Arabski.indb 216

13/01/2006 08:51:23

217

Translation in Studies of Language Contact

Table 16.2 The most frequently calqued L2 items Calqued L2 items

(%)

As white as chalk

81

The eye of the needle

42

How do you find + obj.

25

Catch fire

22

Goat dies

19

Read between the lines

19

Replication To obtain more data on the incidence of literal translation, the test was replicated in March 2003; the original project was conducted in March 1997 as part of a larger investigation of cross-linguistic interaction in an L1 environment (cf. Latkowska, 2001). The sample was composed of 12 advanced students of English who were taught through the medium of English, and who were in their fourth year of the English Department’s MA programme. This time, too, the subjects worked to a time limit which was shorter than in the previous study (40 mins), the underlying assumption being that when pressed for time the subjects were likely to resort to measures facilitating both processing and production. Drawing on the results of previous research, it was hypothesised that one of them could be wordfor-word translation. By and large, the results of the replicated test confirm the findings of the previous study. As shown in Table 16.3, the subjects did experience difficulty while translating into their mother tongue, which manifested itself as the inability to retrieve specific L1 equivalents. It did not lead to massive borrowing from the source language, however, as evidenced by the scores for literal translation (9%) and avoidance (29%). Table 16.3 Results of the replicated test Results

(%)

Correct translation

62

Calques

9

Avoidance

29

9612Arabski.indb 217

13/01/2006 08:51:24

218

Cross-linguistic Influences in the Second Language Lexicon

Table 16.4 Avoidance as observed in the replicated translation test Avoided L2 expressions

(%)

Put an end

83

Catch fire

83

Goat dies

67

Pay through the nose

67

The eye of the needle

50

Hang wallpaper

50

Among the most frequently calqued expressions were as white as chalk (*biaÙy jak kreda – 84%) and grey hair (*szare wÙosy – 33%). The results for avoidance are displayed in Table 16.4.

Discussion The most conspicuous finding of both studies is that in translations into the L1 by proficient bilinguals the incidence of literal/word-for-word rendition is relatively low and does not seem to result from a strategy to translate sentences linearly as presented in the source language. On the contrary, given that the subjects in the second study were forced to work under considerable time pressure, which incidentally did not result in an increase in calques, the inevitable conclusion must be that they were driven by semantic concerns. To be more specific, rather than translate sentences literally they reduced them to sense and processed in terms of meaning, which was then rendered into the L1. The much higher incidence of avoidance in the second study seems to prove this point. Also, the fact that build castles in the air was twice replaced by semantically similar but not exactly equivalent idioms, i.e. buja° w obÙokach, myïle° o niebieskich migdaÙach suggests that the processes operating in advanced translation are essentially meaning-oriented. Drawing on the evidence from previous research, which shows that the results of translation correlate with those of acceptability judgements and that the incidence of transfer in spontaneous translation without any ensuing impact on acceptability judgements is an isolated occurrence (cf. Latkowska, 2001), one is led to believe that this indeed is the case. The outcomes of this study might have been affected by the fact that apart from two words that resembled their English equivalents

9612Arabski.indb 218

13/01/2006 08:51:25

Translation in Studies of Language Contact

219

phonetically and to a certain extent semantically (treat-traktowa°, linelinijka [tekstu]), there were virtually no other cognates that could function as potential stimulus words and thus give rise to formally induced transfer. Surprisingly, however, formal similarity was not conducive to extensive lexical borrowing as evidenced by the results for calqued expressions. The scores for lines and treat constitute 19% and 6% respectively of the collected corpus. Yet another insightful perspective is offered by transfer studies, which show that idioms are generally non-transferable (Kellerman, 1979; Jordens, 1977). This position seems to be upheld by the findings of this study as avoidance, i.e. indirect rendition of the meaning of the elicited items was used in preference to their direct equivalents. This was also true of the test’s only idiom whose literal translation was correct by Polish standards and whose transparency was unquestionable, i.e. poÙoČy° kres/koniec (to put an end to something). In this case, too, the subjects opted for a single word translation which conveyed the meaning of the sentence accurately without resorting to idiomatic expressions (to finish/end). One cannot rule out the possibility, however, that in the present study avoidance was used as a compensation strategy in a situation when the subjects acted under the pressure of time, and when the elicited items were not readily available. Furthermore, it must be stressed that the non-transferability view has been questioned on the grounds that bilinguals often make use of calques which are in principle transferred idioms characterised by a degree of opacity (Odlin, 1989). Odlin (1989: 147) links their occurrence to the social context of language use, which suggests that transfer may also function as a social rather than purely linguistic phenomenon. What is more, different settings are likely to induce a different intensity of transfer. Thus, focused contexts feature efforts to minimise the influence from other languages while in unfocused settings the constraints on negative transfer tend to be weaker. He also adds that formal situations require highly focused language. In the light of this statement, one can venture a conjecture that perhaps the low incidence of literal translation reported in this study was the effect of the formal setting in which it was conducted. Since the focus was on correctness, the subjects opted for safe solutions that reflected L1 norms without running the risk of using borrowings from the source language. Some calquing did, however, occur and on closer inspection, it also appears to be semantically motivated, the determining factors being either greater transparency (catch fire/*zÙapa° ogieÚ) and typicality (as white as chalk/*biaÙy jak kreda, grey hair/*szare wÙosy) or shared semantic concepts (the eye of the needle/*oko igÙy). The latter possibility is of special

9612Arabski.indb 219

13/01/2006 08:51:25

220

Cross-linguistic Influences in the Second Language Lexicon

relevance to this discussion since the translations offered for the eye of the needle included words such as dziurka, otwór and otworek, all of which constitute a form of opening which is a concept the English eye shares with its Polish equivalent oko. Interestingly enough, it is a by-product of the test that further upholds the semantic character of the translation process. Namely, the rationale behind the decision to include the item I’ve been watching this goat for an hour. It’s extremely lazy; it hasn’t even moved. Perhaps it’s died in the test was that it would help to observe if the umiera°/ zdycha° (die) distinction, which is absent from English, was preserved in English-Polish translation (in Polish zdycha° is used mainly in connection with animals). While the results obtained for this sentence suggest that the distinction is vulnerable to cross-linguistic influences, an informal interview conducted with the subjects after the completion of the replicated study revealed that most of them perceived it’s died as it’s dead. All things considered, this is a classic example of expectation - driven understanding, which once again proves that in translation proficient bilinguals attend to meaning rather than form. A question that merits consideration is whether or not translation offers a window on bilingual competence, and if so, what exactly it reveals. Although it is true that elicited data often contain items which never or rarely occur in spontaneous speech, e.g. approximations [*kupi° zamek w powietrzu (buy a castle in the air) for build castles in the air], it is the very nature of translation to elicit the language that the subjects have at their command. Unfortunately, in practice they often have limited control over the content and linguistic resources required to perform the task and are, in effect, forced to resort to language they would not normally use, i.e. items they are not confident of or familiar with. In such a scenario the use of avoidance seems only natural as it helps one to keep clear of potential lexical pitfalls without affecting the effectiveness of communication. Incidentally, the use of avoidance as a soft option has also been the focus of translation studies. Indeed, among the strategies employed by translators when they encounter a translational problem, Chesterman (1999: 140) mentions that of ‘changing something’, where the suggested shift refers to the word class, clause structure or entire unit, i.e. word, phrase, clause, etc. (Catford, 1965), as well as to other syntactic and semantic modulations. Since strategies of this type are conscious measures to solve a variety of translation problems, they may be able to provide data on the performative aspects of the bilingual’s competence, including natural translation ability. As explained earlier, translation has been used in research into language contact phenomena such as L1 attrition, where avoidance of a form may be symptomatic of the bilingual’s underlying uncertainty and confusion.

9612Arabski.indb 220

13/01/2006 08:51:26

Translation in Studies of Language Contact

221

Although this is certainly indicative of insecurity in performance, which according to Andersen (1982) is characteristic of the process in question, the absence of a form cannot be used as a basis for categorical statements about language loss due to the fact that avoidance has multiple facets and uses. To put it another way, translation offers no guarantee that the forms avoided have really been lost, which substantially reduces the reliability of findings obtained from it. On the other hand, in view of the fact that literal rendition into the L1 is relatively rare, and that when it does occur, it tends to result from a limited command of the target language (see Part 2), there are grounds to assume that its occurrence might be symptomatic of imminent attrition and/or an L2-induced restructuring of the bilingual’s L1 competence. Although this conjecture needs to be tested empirically, an argument in its favour is the noticeable occurrence of *jak znajdujesz (how do you find + obj) in the translations obtained in the original study. Despite its relative opacity, the calque found its way into the subjects’ L1. Given that an identical expression used to exist in the Polish language, the immediate conclusion must be that this archaism was reinstated in the bilinguals’ lexicon under the influence of L2 English. Moreover, since research findings show that translators into L1 are able to make better judgements of output quality (Campbell, 1998) than translators into L2, it can be concluded that the subjects in both studies were probably fairly confident of the correctness of the calques they applied. All things considered, this is yet another argument in support of Cook’s (1996: 64) multi-competence theory which advances the view that people who know two languages do not have identical linguistic competence to monolinguals. No discussion of translation data can ever be complete without a mention of the bilingual’s language modes and their impact on the task in question. This undoubtedly calls for the activation of the languages involved, i.e. the bilingual mode (Grosjean, 1997, 2001). Unlike most bilingual contexts which require a weaker activation of the guest language, in translation both the source language and the target language are active to the same extent as both are needed for perception and production (Grosjean, 1997: 174). Consequently, one could expect a certain amount of interference and borrowing to occur, as has indeed been reported in the literature on the subject (Gile, 1995). However, both research and popular experience show that translators are capable of suppressing the source language, at least at the lexical level. The low percentage of borrowing observed in this study is a case in point. By contrast, the subjects’ noticeable preference for avoidance seems to suggest that in a bilingual mode cross-linguistic interaction occurs primarily in the semantic domain and may lie at the root of production irregularities and/or interference (cf. Latkowska, 2001).

9612Arabski.indb 221

13/01/2006 08:51:27

222

Cross-linguistic Influences in the Second Language Lexicon

Conclusion The central finding of this study has been that word-for-word rendition is relatively rare in translations into L1 by proficient bilinguals. Its low incidence may be attributed to the nature of the operations involved in bilingual language processing, which, as demonstrated above, give priority to semantic factors. Also of relevance here are the context of language use and general proficiency in both languages, as well as the widely reported resistance to the transfer of idioms. A somewhat surprising finding concerns the role of avoidance in translation. Apart from being a cognitive measure adopted as an opt-out when the bilingual is faced with two contrasting forms expressing the same idea, avoidance is undoubtedly a conscious strategy used to solve emerging translation problems. Consequently, it provides insights into the mechanics of translation, as well as aspects of cross-linguistic interaction such as, for instance, underproduction and psycholinguistic incentives. Unfortunately, researchers have no way of telling whether the forms avoided in translation have been lost/restructured or whether they have been omitted due to processing factors. This substantially reduces the utility of translation as an elicitation measure. The occurrence of literal rendition, by contrast, as it seems to be relatively rare, may be indicative of the L2-induced restructuring of the bilingual’s competence and thus may shed light on the nature of the conceptual representations underlying bilingual performance. In summary, the accumulated evidence suggests that translation may be of value to research into proficient and functional bilingualism provided it is used selectively to examine relevant phenomena and that its findings are interpreted with care and in relation to data obtained from other sources. References Andersen, R.W. (1982) Determining the linguistic attributes of language attrition. In R.D. Lambert and B.F. Freed (eds) The Loss of Language Skills. Rowley, MA: Newbury House Publishers. Bell, R.T. (1991) Translation and Translating. Theory and Practice. London: Longman. Burmeister, A. and Ufert, D. (1980) Strategy switching. In S. Felix (ed.) Second Language Development. TĀbingen: Gunther Narr Verlag. Campbell, S. (1998) Translation into the Second Language. London: Longman. Catford, J.C. (1965) A Linguistic Theory of Translation. Oxford: Oxford University Press. Chesterman, A. (1999) Communication strategies, learning strategies and translation strategies. In K. Malmkjær (ed.) Translation and Language Teaching. Manchester: St Jerome Publishing. Cook, V. (1996) Competence and multi-competence. In G. Brown, K. Malmkjær and J. Williams (eds) Performance & Competence in Second Language Acquisition (pp. 57–69). Cambridge: Cambridge University Press.

9612Arabski.indb 222

13/01/2006 08:51:28

Translation in Studies of Language Contact

223

Fischer, K. and Pipp, S. (1984) Processes of cognitive development: Optimal level and skill acquisition. In M. Malakoff and K. Hakuta (1991) Translation Skill and Metalinguistic Awareness in Bilinguals (pp. 141–66). Cambridge: Cambridge University Press. Forster, K.I. and Jiang, N. (2001) The nature of the bilingual lexicon: Experiments with the masked priming paradigm. In J.L. Nicol (ed.) One Mind, Two Languages (pp. 72–83). Oxford: Blackwell. Gile, D. (1995) Basic Concepts and Models for Interpreter and Translator Training. Amsterdam: John Benjamins. Grosjean, F. (1982) Life with Two Languages. Cambridge, MA: Harvard University Press. Grosjean, F. (1997) The bilingual individual. Interpreting: International Journal of Research and Practice in Interpreting 2, 163–91. Grosjean, F. (2001) The bilingual’s language modes. In J.L. Nicol (ed.) One Mind, Two Languages. Oxford: Blackwell. Harris, B. (1977) The importance of natural translation. Working Papers in Bilingualism 12, 96–114. Harris, B. and Sherwood, B. (1978) Translating as an innate skill. In D. Gerver and H. W. Sinaiko (eds) Language Interpretation and Communication. New York: Plenum Press. Hatch, E. and Brown, C. (1995) Vocabulary, Semantics and Language Education. Cambridge: Cambridge University Press. Huffines, M.L. (1991) Pennsylvania German: Convergence and change as strategies of discourse. In H.W. Seliger and R.M. Vago (eds) First Language Attrition (pp. 125–38). Cambridge: Cambridge University Press. Hulstijn, J. and Tangelder, C. (1991) Intralingual interference: Lexical errors as a result of formal and semantic similarities among English word-pairs. Paper presented at the First Annual Meeting of the European Second Language Association (EUROSLA), Salzburg. Cited in Singleton (1999). Jordens, P. (1977) Rules, grammatical intuitions, and strategies in foreign language learning. Interlanguage Studies Bulletin 2, 5-76. Källkvist, M. (1998) How different are the results of translation tasks? A study of lexical errors. In K. Malmkjær (ed.) Translation and Language Teaching. Manchester: St Jerome Publishing. Kellerman, E. (1979) Transfer and non-transfer: Where are we now ? Studies in Second Language Acquisition 2: 37-57. Kroll, J.F. and Tokowicz, N. (2001) The development of conceptual representation for words in a second language. In J.L. Nicol (ed.) One Mind, Two Languages. Oxford: Blackwell. Larsen-Freeman, D. and Long, M. (1991) An Introduction to Second Language Acquisition Research. London: Longman. Latkowska, J. (2001) The influence of the L2 on L1 competence: A study of crosslinguistic interaction in an L1 environment. In J. Arabski (ed.) Insights into Foreign Language Acquisition and Teaching. Katowice: University of Silesia Press. Lorscher, W. (1991) Translation Performance, Translation Process, and Translation Strategies. TĀbingen: Gunther Narr Verlag. Malakoff, M. and Hakuta, K. (1991) Translation skill and metalinguistic awareness in bilinguals. In E. Bialystok (ed.) Language Processing in Bilingual Children. Cambridge: Cambridge University Press. McCarthy, M. (1990) Vocabulary. Oxford: Oxford University Press.

9612Arabski.indb 223

13/01/2006 08:51:29

224

Cross-linguistic Influences in the Second Language Lexicon

Odlin, T. (1989) Language Transfer: Cross-linguistic Influence in Language Learning. Cambridge: Cambridge University Press. Romaine, S. (1995) Bilingualism. Oxford: Blackwell. Seleskovitch, D. (1976) Interpretation, a psychological approach to translation. In R.W. Brislin (ed.) Translation: Applications and Research. New York: Gardner Press. Sharwood-Smith, M. (1983) On first language loss in the second language acquirer: Problems of transfer. In S. Gass (ed.) Language Transfer in Language Learning. Rowley, MA: Newbury House Publishers. Singleton, D. (1999) Exploring the Second Language Mental Lexicon. Cambridge: Cambridge University Press. Swain, M., Dumas, G. and Naiman, N. (1974) Alternatives to spontaneous speech: Elicited translation and imitation as indicators of second language competence. Working Papers on Bilingualism 3, 68–79. Toury, G. (1986a) Monitoring discourse transfer: A testcase for a developmental model of translation. In J. House and S. Blum-Kulka (eds) Interlingual and Intercultural Communication. Discourse and Cognition in Translation and Second Language Acquisition Studies. TĀbingen: Gunther Narr Verlag. Toury, G. (1986b) Natural translation and the making of a native translator. TEXTconTEXT 1, 11-29. Underhill, N. (1987) Testing Spoken Language. Cambridge: Cambridge University Press. Weinreich, U. (1963) Languages in Contact. The Hague: Mouton. Weller, G. (1989) Some polemic aspects of translation in foreign language pedagogy revisited. Translator and Interpreter Training and Foreign Language Pedagogy. State University of New York at Binghamton: American Translators Association Scholarly Monograph Series, Vol. III. Yagmur, K. (2002) Issues in finding the appropriate methodology in language attrition research. Paper presented at the International Conference on First Language Attrition, Vrije Universiteit Amsterdam. Zabor, L. (2001) Grammaticality judgement and elicited translation tasks in SLA research. In J. Arabski (ed.) Insights into Foreign Language Acquisition and Teaching. Katowice: University of Silesia Press.

Appendix Types of sentences in the translation test (1) I can’t thread this bloody needle. Its eye is so small that I keep missing it. (2) I’d never seen anyone looking as ill as Jeff. He was as white as chalk when he came home from work that night. (3) How do you find Prof. Ratkin? She gives the impression of being unusually intelligent, but if you ask me, she’s a bit too pedantic. (4) Paper catches fire easily. (5) I’ve been watching this goat for an hour. It’s extremely lazy, it hasn’t even moved. Perhaps it’s died. (6) She didn’t give me a definite answer but reading between the lines, I think she’ll take the job.

9612Arabski.indb 224

13/01/2006 08:51:29

Translation in Studies of Language Contact

225

(7) I must put an end to his advances. Either he marries her or he’ll have to stop coming here. (8) I’ve got a ticket for today’s football match but I didn’t think I’d pay through the nose for it. It cost $30 but I had to pay twice as much. (9) We’ve almost finished redecorating our flat. We only have to hang wallpaper in the kitchen and bedroom. (10) Your plans are unrealistic. Rather than building castles in the air you’d better get a decent job and earn a living for your family.

9612Arabski.indb 225

13/01/2006 08:51:30

Chapter 17

On Building Castles on the Sand, or Exploring the Issue of Transfer in the Interpretation and Production of L2 Fixed Expressions ´LICKA ANNA CIES Introduction This chapter addresses the degree to which L2 learners’ native language influences their understanding and production of fixed expressions in a second language. The term fixed expressions has been deliberately chosen here as a very general term to cover multi-word idiomatic phrases. Traditional definitions of idioms would treat them as ‘multi-word items which are not the sum of their parts; they have holistic meanings which cannot be retrieved from the individual meanings of the component words’ (Moon, 1997: 46). Similar, non-compositional definitions of idioms were suggested, among others, by Fraser, 1979; Irujo, 1986; Makkai, 1972; Nicolas, 1995; Schenk, 1995; and Weinreich, 1969. Defining idioms as non-compositional phrases, however, fails to acknowledge a growing strand of research that has demonstrated the essentially compositional nature of these expressions, with individual meanings of idiom constituent words actively contributing to their figurative interpretation (see, for example, Cacciari, 1993; Cacciari & Glucksberg, 1991; Cacciari & Tabossi, 1988, 1993; Gibbs, 1993; Gibbs & Nayak, 1989; Gibbs et al., 1989a; Gibbs et al., 1989b; Simpson, 1991; Titone & Connine, 1994, 1999). Therefore, rather than adopting the narrow definition of idioms as conventionalised expressions, whose meaning cannot be determined from the meaning of their parts, I will assume that idioms are multi-word expressions falling on a continuum of compositionality – from highly fixed and opaque phrases (like trip the light fantastic, for example) to highly compositional and transparent expressions (like play with fire). Despite a long tradition of research into idiomaticity in the first language, the issue of the comprehension and production of idioms by second language 226

9612Arabski.indb 226

13/01/2006 08:51:31

On Building Castles on the Sand

227

learners has received surprisingly little attention. Scarcity of studies into the processing of idioms by second/foreign language learners is particularly intriguing, given the widely held view that idioms constitute one of the most problematic areas of foreign language learning (see, for example, Alexander, 1987; Cacciari, 1993; Carter, 1998; Fernando, 1996; Gairns & Redman, 1986; Irujo, 1993; Kovecses & Szabo, 1996; Lattey, 1986; McCarthy, 1990; Moon, 1997; Zughoul, 1991). In an attempt to fill this research gap, as well as to address the issue of cross-linguistic influence in the processing of L2 idiomatic expressions, the present paper describes the study addressing the role of the learners’ native language in their comprehension and production of L2 idiomatic phrases.

The Study Research questions The study focused on the degree to which L2 learners’ native language influences their comprehension and production of L2 idioms. In addition, it attempted to probe interpretation strategies employed by second language learners when faced with the task of interpreting and explaining L2 idioms varying in terms of their translatability into the learners’ L1. Depending on the degree to which English idioms employed in the study could be literally translated into Polish, they were divided into three categories: Lexical-Level Idioms, that is, expressions with exact translation equivalents in Polish; Semi-Lexical Level Idioms; or partially matching idioms; and PostLexical Level Idioms, that is, non-matching ones. The classification of idioms into three types has been borrowed from Liontas (2002). The following research questions were formulated: (1) How does the learners’ L1 affect the comprehension and production of second language idioms? (2) Will transfer effects differ for the three types of L2 idiomatic expressions? (3) What reading and interpretation strategies do advanced L2 learners employ to interpret familiar and less familiar L2 idioms? Based on the previous research into the role of L1 in processing L2 idioms, it was assumed that the learners’ native language will considerably affect their performance on the production and comprehension tasks. It was further assumed that transfer effects will differ depending on the degree of L2 idioms’ translatability into L1, with L2 idioms having direct translation equivalents in the learners’ L1 being easiest to process and partially matching idioms most prone to L1 influence. Finally, it was assumed that a literal analysis of idiom constituent words will constitute

9612Arabski.indb 227

13/01/2006 08:51:31

228

Cross-linguistic Influences in the Second Language Lexicon

the major interpretation strategy employed by learners when reading and explaining the meaning of L2 idiomatic expressions. Following from these assumptions, four hypotheses were advanced: (1) Lexical-Level (LL) Idioms (L2 idioms with direct translation equivalents in L1) should be the easiest to comprehend and produce correctly. (2) Semi-Lexical Level (SLL) Idioms (L2 expressions only partially matching their L1 equivalents) should be particularly prone to negative transfer. Such negative transfer should manifest itself in the production of these idioms, whereas their comprehension could be comparable to that obtained for LL Idioms. (3) Post-Lexical Level (PLL) Idioms should pose the most difficulties for L2 learners both on comprehension and production tasks. Transfer effects obtained for these idioms should be much smaller than those manifested for the remaining two idiom types. (4) Analysis of interpretation strategies reported by L2 learners should reveal the importance of literal meanings of idiom constituents in constructing the overall figurative interpretation of an L2 idiom, irrespective of the idiom type. Participants A total of 36 advanced learners of English from the School of English, Adam Mickiewicz University, PoznaÚ, Poland, volunteered to take part in the experiment. They were all Year 2 students who had successfully passed their Practical English Examination delivered at the end of Year 1 and roughly corresponding to the Cambridge Advanced English level examination administered by the University of Cambridge Local Examinations Syndicate. Among the participants, 10 were male and 26 female, and their average age was 21.9 years. The sample was divided into three groups. Group one consisted of 12 students who were administered both the comprehension and the follow-up production tests. Group two comprised 13 students who were only tested on their production skills, while Group three consisted of 11 students who were only asked to complete the comprehension test. Materials and methodology As stated in the introductory section, the L2 idioms chosen for the study could be broadly categorised as Lexical-Level, Semi-Lexical Level, and Post-Lexical Level, depending on whether they had fully matching, partially matching, or non-matching translation equivalents in Polish. A

9612Arabski.indb 228

13/01/2006 08:51:32

On Building Castles on the Sand

229

full list of the 21 L2 idioms employed in the study, along with their Polish equivalents, is provided in Appendix 1. An example of an LL Idiom is move heaven and earth, which can be translated word-for-word into its Polish counterpart poruszy° niebo i ziemi¿. In turn, SLL Idioms are only partially matching, in that one or two lexical items in their Polish counterparts are different, as in the English idiom promise somebody the moon, rendered in Polish as obieca° komuï zÙote góry (literally promise somebody golden mountains). Finally, PLL Idioms have L1 equivalents widely varying in terms of their syntactic structure and semantic composition. To provide an example, the English PLL idiom have an axe to grind corresponds to the Polish phrase upiec wÙasn­ pieczeÚ przy cudzym ogniu (lit. roast your own roast over somebody else’s fire). The idioms for the study were drawn from Longman (1998), Jaworska (2002) and B­ba and Liberek (2002). The comprehension task required participants to look at each of the L2 idioms in turn and answer three questions about it. Question 1 asked for the explanation of the idiom’s meaning, whereas Question 2 asked participants to write down, in their own words, their thought processes, reading strategies, or mental images they had during the processing of the idiom. Finally, Question 3 asked them to guess the best Polish equivalent of the English idiom. The 21 idioms were printed, two to a page, along with the three questions about them, in a 12-page booklet. The order of the idioms was randomised, so that each student received a booklet with a different ordering of idiomatic expressions. Twenty-three students taking part in this task were encouraged to try and interpret every idiom, even if they were not familiar with it. Even though participants were allowed to use Polish for explaining the meaning and reporting on their interpretation strategies, the overwhelming majority (99%) provided answers in English. Students were tested in two groups, one of which consisted of 12, and the other of 11 people. The comprehension task took approximately 60 minutes to complete. In turn, the production task consisted of two parts: a discourse completion and a translation section. The discourse completion task included 21 short contexts, one for each idiom. One word in each idiom was missing, and participants had to supply the missing word. A sample entry for the discourse completion task is provided in Appendix 2. The contexts for the idioms were adapted from Wolfram-Romanowska et al. (1999) and Wright (1999). Turning to the translation task, it included Polish sentences which contained idioms, as well as the English translation of the sentences with the translation of the idioms omitted. Participants were asked to supply the English idiom which they would use in the context of the sentence. Examples of all three types of idioms were provided in the instructions, so

9612Arabski.indb 229

13/01/2006 08:51:33

230

Cross-linguistic Influences in the Second Language Lexicon

participants were aware that a literal translation was not always possible. A sample entry for the translation task is provided in Appendix 3. Participants were first given the discourse completion and then the translation tasks. They were tested in two groups. One of them comprised 12 students who had completed the comprehension task two weeks earlier and so now they encountered the same idioms again in a different task. Another group comprised 13 students who had not participated in the comprehension task, so their responses to the production task were unaffected by the earlier exposure to the idiom stimuli. The discourse completion and translation tasks took approximately 25 minutes to complete. Results and discussion Scoring and analysis: The production task

The items on the discourse completion and translation tests were scored as correct, incorrect, or incorrect with transfer. The term transfer is used here to specifically indicate cases of the incorrect use of a translation of one or more content words from a corresponding Polish idiom. To provide an example, the phrase promised her the golden mountains, provided instead of the required promised her the moon, consists of the content words golden and mountains translated from the Polish counterpart (obiecywa° zÙote góry). Answers scored as incorrect included cases where participants modified the idioms by substituting their content words with other lexical items or used paraphrases rather than fixed expressions. To illustrate with an example, I had my fingers and toes, I had my fingers and thumbs, or I was all fingers and toes, provided instead of the correct form I was all fingers and thumbs, were all classified as incorrect. So was the response dream about unattainable, provided in place of build castles in the air. Further examples of the responses classified as incorrect and incorrect with transfer for all three idiom types are provided in Table 17.1. To ensure reliability of scoring, a Polish-English bilingual was asked to score the discourse and translation test. Interrater reliability between the experimenter and the rater was 0.98.

Results: The Production Task The rated responses were converted into percentages, which were calculated separately for the group that had been previously administered the comprehension task (henceforth the follow-up group) and for the group that had encountered the idioms for the first time in the experimental situation (henceforth the production-only group). The percentages for the follow-up group are presented in Table 17.2.

9612Arabski.indb 230

13/01/2006 08:51:34

231

On Building Castles on the Sand

Table 17.1 Examples of the responses classified as incorrect and incorrect with transfer on the translation and discourse completion task for each type of idiom. Content words transferred from Polish are marked in bold Idiom

Incorrect

Incorrect with transfer

loosen somebody’s tongue (LL)

loosed people’s tongues/opened their tongues

untied/untangled/ resolved/unfolded people’s tongues

add fuel to the fire (SLL)

put some sticks to the fire/added add olive/oil to the wood to the fire fire

promise somebody the moon (SLL)

promised paradise/promised her wonders/promised he would do anything for her/ promised her the heaven and stars

promised her the golden mountains/ heaps of gold/the mountains of gold

build castles in the air (SLL)

dream about unattainable

build castles on the sand/build sandcastles/build castles out of the sand/building castles on the ice

go like hot cakes (SLL)

go like hot sausages/sold very well

go like warm buns/ rolls/ bread-rolls

keep a stiff upper lip (PLL)

stay calm/remain as cool as a cucumber/be very calm

retain cold blood/stay cold blooded/keep cold blood

beat around the bush (PLL)

go around the bush/beat around put cotton around it the details/hide anything

be all fingers and thumbs (PLL)

I had my fingers and toes/I had I was all left handed/ my fingers and thumbs/I was all I had two left arms/I thumbs/I was all fingers and toes had two left hands

have an axe to grind (PLL)

make profit using somebody cook his roast on else’s account/kill two birds with somebody’s oven one stone/ use somebody for his own business

9612Arabski.indb 231

13/01/2006 08:51:34

232

Cross-linguistic Influences in the Second Language Lexicon

Table 17.2 Percentage of the correct, incorrect, and transfer-based responses provided by the follow-up group in the production task (DC stands for the discourse completion and T for the translation task) Idiom type

Correct responses

Incorrect responses

Incorrect responses with transfer

DC

T

DC

T

DC

T

LL

79%

60%

11%

28%

10%

12%

SLL

65%

49%

19%

42%

16%

9%

PLL

78%

38%

22%

52%

0%

10%

Beginning with the percentages obtained for the discourse completion task, what strikes one as immediately apparent is the relatively high proportion of correct responses for all three idiom types and a very low percentage of incorrect responses, or incorrect responses with transfer. In the feedback session organised at the end of the test, the students admitted that they had simply memorised the idioms they had seen on the comprehension test two weeks earlier. Even though the comprehension booklets were collected by the experimenter from the students, they were diligent enough to note down all the idioms whose meanings they were unsure and to commit them to memory. Hence, the discourse completion task administered two weeks afterwards could really be viewed as a recall or memory test. Despite the essentially recall-based nature of the discourse completion task for the follow-up group, the tendency for the incorrect responses to increase from LL to SLL and PLL idioms is apparent. However, a closer look at the table reveals that, while LL idioms were, in accordance with Hypothesis 1, the easiest to produce, yielding 79% of correct responses, the corresponding percentage of the correct responses obtained for PLL idioms is highly comparable (78%), contrary to the predictions of Hypothesis 3, under which PLL idioms should prove most problematic for L2 learners. As was mentioned earlier, however, participants comprising the followup group admitted memorising the idioms they had encountered on the earlier administered comprehension task, so their performance might have resulted from their familiarity with PLL idioms. It seems clear from the data that SLL idioms appeared most difficult for the learners, eliciting the smallest number of correct responses (65%) and the overall biggest number of incorrect ones (19% of incorrect responses and 16% of incorrect

9612Arabski.indb 232

13/01/2006 08:51:35

233

On Building Castles on the Sand

responses with transfer). The transfer percentages support Hypotheses 1, 2, and 3 advanced earlier, in that the percentage of incorrect responses with transfer is the biggest with SLL idioms (16%) and the smallest, or virtually non-existent (0%) with PLL idioms. The research hypotheses are further confirmed by the data obtained on the translation test. While LL idioms elicited 60% of correct responses and only 28% of incorrect ones, the percentage of correct responses for SLL idioms was 49% and of incorrect ones as high as 42%. PLL idioms posed the most difficulty, resulting in only 38% of correct responses and as much as 52% of incorrect responses. These data support the prediction that LL idioms are the easiest to produce whereas PLL idioms pose particular difficulties for second language learners. On the other hand, the percentage of incorrect responses with transfer obtained for the follow-up group is slightly unexpected. Contrary to Hypothesis 3, which predicts small transfer effects for PLL idioms, the percentage of incorrect responses with transfer obtained for PLL idioms (10%) is not significantly different from the data obtained for either LL or SLL idioms. This may indicate that L2 learners do not hesitate to translate the opaque Polish idiom literally, even if the resulting English rendering may look odd to them. Turning now to the responses provided by the production-only group on the discourse completion and translation tests, the relevant percentages are summarised in Table 17.3. Since the learners of English comprising the group had not encountered the idioms in the comprehension test, as did the follow-up group, their overall performance on both the discourse completion and translation tasks is worse than that of the other group, as is visible from the analysis of correct and incorrect responses. Beginning with the discourse completion test, the

Table 17.3 Percentage of the correct, incorrect, and transfer-based responses provided by the production-only group in the production task (DC stands for the discourse completion and T for the translation task) Idiom type

Correct responses

Incorrect responses

Incorrect responses with transfer

DC

T

DC

T

DC

T

LL

49%

51%

43%

36%

8%

14%

SLL

12%

12%

53%

35%

36%

53%

PLL

42%

26%

58%

57%

0%

17%

9612Arabski.indb 233

13/01/2006 08:51:36

234

Cross-linguistic Influences in the Second Language Lexicon

overall tendency is identical to that obtained for the follow-up group, in that the percentage of correct responses is the biggest (49%) for LL idioms, relatively smaller for PLL idioms (42%), and the smallest for SLL idioms (12%). Likewise, the percentage of incorrect responses and incorrect responses with transfer appears the smallest for LL idioms (43% and 8% respectively), slightly bigger overall for PLL idioms (58% and 0%), and the biggest for SLL idioms (52% and 36%). Thus, contrary to Hypothesis 3, under which PLL idioms should be the most difficult for L2 learners, it appears that SLL idioms surpass PLL ones in terms of difficulty. Support for another tenet of Hypothesis 3, under which PLL idioms should manifest small transfer effects, has been obtained, however, in that no instances of incorrect responses with transfer were found for those idioms. In accordance with Hypotheses 1 and 2, LL idioms were found to be the easiest for L2 learners and SLL idioms the most prone to transfer, with the percentage of transfer-based responses as high as 36%. Virtually identical tendencies were demonstrated for the production-only group’s performance on the translation task. Thus, SLL idioms appeared the most difficult to produce, eliciting only 12% of correct answers, 35% of incorrect ones and 53% of transfer-based answers. In accordance with Hypothesis 1, LL idioms were the easiest to produce, eliciting overall the biggest percentage of correct (51%) and the smallest percentage of incorrect answers with or without transfer (49% altogether). As found for the follow-up group, also here an unexpectedly high percentage of transfer responses was obtained for PLL idioms (17%), contrary to what was assumed in Hypothesis 3. Even though, overall, the biggest transfer effect occurred, as predicted, in SLL idioms, the presence of transfer effect in learners’ production of PLL testifies to the essentially literal, L1-based mechanism of L2 idiom production. Scoring and analysis: The comprehension task

The comprehension task was analysed for the correctness of idiom interpretation, correctness of the Polish equivalent provided for each idiom, as well as the nature of interpretation strategies reported by participants in the protocols. For Questions 1 (explain the meaning) and 3 (provide the Polish equivalent), a score of 2 was given for those answers where the idiom was correctly interpreted and where the correct Polish equivalent was provided. A score of 1 was given for those answers where the idiomatic expressions were only partially interpreted. Those expressions which were misinterpreted, wrongly translated, or contained no answer were scored as 0. The scoring procedure has been borrowed from Liontas’ (2002) study into L2 idiom interpretation strategies. The scores were calculated separately for

9612Arabski.indb 234

13/01/2006 08:51:37

On Building Castles on the Sand

235

LL, SLL, and PLL idioms for the interpretation and translation questions. Since 23 people filled out the comprehension protocols and each of them responded to three questions about seven idioms of each type, the maximum total score to obtain for each idiom type was 322. For clarity of presentation and ease of comparison, the scores were subsequently converted into percentages showing correct responses. Responses to Question 2 (describe interpretation strategies, thought processes and mental image) were analysed by the experimenter and assigned to one of the 10 categories that emerged from a close analysis of participants’ metacognitive protocols. A list of strategies, along with examples of metacognitive comments exemplifying a given strategy, is provided in Appendix 4. To start with, the strategy labelled Prior knowledge/Familiarity, refers to those instances where participants explicitly stated that they had learned the idiom before and provided no other comments concerning their thought processes or mental images accompanying the reading of the idiom. Guessing based on literal analysis subsumes those responses where participants attempted to guess the figurative meaning of an L2 idiom by analysing literal meanings of its constituent words and combining the obtained meanings to infer an idiomatic interpretation. In turn, Guessing based on one key word refers to those cases where L2 learners specifically emphasised the importance of one word in the idiom which helped them to infer the idiom’s meaning, whereas Guessing based on two keys words to cases where two words apparently helped the learners deduce the meaning of the idiom. Another strategy, labelled Analogy between literal and figurative senses, includes responses in which participants consciously referred to the analogy between the literal meaning of the idiom and its symbolic reference, as if aware of the dual (literal and figurative) senses of idiom constituent words. Responses which included description of a person, action, or event denoted by the literal meaning of the idiom have been classified as exemplifying the next category, referred to as Reference to the person/action/ event denoted by the literal meaning of the idiom. Some respondents failed to provide any information concerning their interpretation strategies and focused instead on the mental image that the idiom evoked in their minds. Since all of such images were tied to the literal meanings of idioms, the answers have been classified as exemplifying the category Literal based image. Some learners inferred the meaning of the idiom by seeking a Polish equivalent first. Such responses belong to the category labelled Reference to or association with a Polish idiom. A number of responses could not be classified as unambiguously exemplifying a particular category, but instead constituted a combination of strategies. Such responses contained, for example, an explicit reference to the Polish translation equivalent, as well

9612Arabski.indb 235

13/01/2006 08:51:38

236

Cross-linguistic Influences in the Second Language Lexicon

as a metacognitive comment on the usefulness of one key word that was helpful in deducing the idiom’s meaning. Such responses were categorised as belonging to the category labelled Combination of strategies. An independent judge was asked to analyse the protocols and assign the responses elicited to Question 2 to the 10 strategies identified by the experimenter. The judge and the experimenter were mostly unanimous in classifying the responses, with the interrater reliability as high as 0.90. The numeric values of total tallies obtained for each strategy were converted into the total percent that each strategy occupies in the total scheme of strategies for LL, SLL, and PLL idioms. Result: The comprehension task

Beginning with the scores obtained for the interpretation and translation questions, the overall results are highly similar for each idiom type, as shown in Table 17.4. In accordance with the predictions of Hypothesis 1, LL idioms turned out to be the easiest to comprehend, eliciting 96% of correct interpretation answers and 95% of correct Polish translations. In turn, SLL idioms appeared more problematic to interpret and translate, eliciting, respectively, 94% and 89% of correct answers. Finally, the percentage of correct interpretations obtained for PLL idioms was as low as 39%, and of correct Polish translations 37%, confirming Hypothesis 3, according to which PLL idioms pose most difficulties for L2 learners on both comprehension and production levels. Turning now to the analysis of the strategies described by participants in response to Question 2, summary of the percentages of various types of strategies obtained by idioms within each group is provided in Table 17.5. Table 17.4 Total scores and percentages obtained for each idiom type on Question 1 (explaining the meaning of the idiom) and Question 3 (providing the best Polish counterpart) in the comprehension task Idiom type

Question 1 – Interpret the meaning Total score (322 max)

Percent of correct responses

Question 3 – Provide a Polish translation Total score (322 max)

Percent of correct responses

LL

309

96%

307

95%

SLL

302

94%

285

89%

PLL

127

39%

119

37%

9612Arabski.indb 236

13/01/2006 08:51:39

237

On Building Castles on the Sand

Table 17.5 Summary of interpretation strategies reported by subjects in protocols Strategy

Percent of total responses within a given group of idioms LL idioms SLL idioms PLL idioms

Prior knowledge/ Familiarity

4%

7%

11%

Guessing based on literal analysis 9%

14%

29%

Guessing based on one key word

2%

6%

16%

Guessing based on two key words 1%

3%

2%

Analogy between literal and figurative senses

29%

28%

6%

Reference to the person/ action/ event denoted by the literal meaning of the idiom

9%

6%

4%

Literal based image

29%

20%

4%

Reference to or association with 16% idiom in Polish

10%

3%

Combination of strategies

–

–

6%

No information given

1%

6%

19%

A number of interesting patterns emerge from the data. For one thing, very few subjects admitted having previous knowledge of the idioms from the three categories. For LL idioms, only 4% of subjects reported familiarity with the idioms, for SLL idioms 7%, and for PLL idioms 11%. The figures obtained for this strategy should be treated with caution, as in some cases participants might have felt obliged to elaborate on their reading strategy and might have done so, forgetting to report on their earlier familiarity with a given idiom. Another interesting pattern is the predominance of strategies based on the literal analysis of idiom constituents. Paradoxically, guessing based on literal analysis is much more frequent for PLL idioms (29%) than for either SLL idioms (14%) or LL idioms (9%). This finding fits in nicely with the assumption of Hypothesis 4, under which literal meanings of idiom elements should significantly contribute to L2 learners’ interpretation of

9612Arabski.indb 237

13/01/2006 08:51:40

238

Cross-linguistic Influences in the Second Language Lexicon

L2 idioms, irrespective of the idiom type. Quite surprisingly, L2 learners participating in the task did not hesitate to analyse literally even the most opaque L2 idioms, ascribing to those idioms a meaning deduced from the literal analysis of their component words. Another essentially literally based strategy frequently employed for the interpretation of idioms turned out to be Analogy between literal and figurative senses. This strategy comprises 29% of the responses provided for LL idioms and 28% of the answers elicited for SLL idioms. Given the transparent nature of those two types of idioms, such figures come as no surprise. For PLL idioms, the analogy-based responses constitute only a small proportion of the total scheme of responses (6%), which is most likely due to their opaque nature. Literal-based image is most frequent for LL idioms (29%), less so for SLL idioms (20%), and even less substantial for PLL idioms, which elicited only 4% of responses based on literal image. The fact that participants did think either of an analogy between literal and figurative senses or of a literalbased image when processing opaque idioms is still surprising, irrespective of the small percentage of such responses. Further importance of literal meanings of idiom constituents is demonstrated by the scores obtained for the strategy labelled Reference to the person/action/event denoted by the literal meaning of the idiom. Here, the percentages drop from LL idioms (9%) to SLL idioms (6%) and PLL idioms (4%), but the fact that even opaque PLL idioms have elicited literal-based responses again provides a strong support for Research Hypothesis 4. Finally, as might have been expected, Reference to a Polish idiom accounts for a bigger percentage of responses to LL idioms (16%) than to SLL (10%) and PLL (3%) ones. Some of the metacognitive comments offered by L2 learners with reference to Polish-based inference throw an interesting light on the presence of transfer in the learners’ minds. In some responses, participants claimed to have deduced the meaning of the idiom from its Polish equivalent, even if no exact translation for the idiom can be found in Polish. Thus, for example, one student explaining the idiom beat around the bush claimed that he thought of the Polish equivalent first (owija° w baweÙn¿; lit. wrap in cotton) and only then came up with its correct meaning.

Conclusions All in all, analysis of the results obtained on the production and comprehension tasks broadly supports Research Hypotheses postulated at the beginning of this paper. Thus, in accordance with Research Hypothesis 1, Lexical Level idioms turned out to be the easiest to comprehend and produce, eliciting the overall biggest number of correct responses and the

9612Arabski.indb 238

13/01/2006 08:51:41

On Building Castles on the Sand

239

smallest number of incorrect responses, on both the comprehension and production tasks. Likewise, consistent with Research Hypothesis 2, SemiLexical Level idioms were found to be particularly prone to transfer, eliciting the highest percentage of transfer-based responses on the discourse completion and translation tasks. Hypothesis 2 further predicted that, while transfer effects should manifest themselves mainly in production of SLL idioms, their comprehension should be comparable to that obtained for LL idioms. This tenet of Research Hypothesis 2 has also been confirmed, in that participants’ performance on Questions 1 and 3 of the comprehension task was highly similar for LL and SLL idioms, with the former exceeding the latter only slightly in terms of the percentage of correct responses. Following Research Hypothesis 3, Post-Lexical Level idioms appeared the most difficult to interpret on the comprehension task. However, contrary to the claim that those idioms would also pose most difficulties in production, PLL idioms turned out to score much better for correctness on the discourse completion and translation tasks than SLL idioms. Finally, as predicted by Research Hypothesis 4, literal-based strategies appeared dominant in L2 learners’ accounts of their idiom interpretation process, which testifies to the essentially literal mechanism of L2 idiom comprehension. Overall, it appears that L2 learners’ native language plays a prominent role in the way they process second language figurative expressions. Interestingly, despite the typological distance between Polish and English and the participants’ awareness of this distance (all of them being trained in linguistics), Polish learners of English were found to actively employ their knowledge of idioms in their native language in the comprehension and production of L2 idioms. Such transfer was manifested even for opaque L2 expressions, which should be relatively uninfluenced by L1. The study reported in this paper thus further strengthens the status of cross-linguistic influence in the comprehension and production of second language figurative phrases. Differential transfer effects obtained for idioms varying in terms of their translatability into the learners’ L1 throw an interesting light into the nature of factors affecting cross-linguistic influence. Finally, predominance of literal-based strategies employed by second language learners in interpreting the meaning of L2 fixed expressions testifies to the relevance of literal analysis of idioms in constructing their figurative interpretations by L2 learners. It is hoped that the present study constitutes a contribution towards a more complete understanding of the processes involved in second language idiom comprehension and production.

9612Arabski.indb 239

13/01/2006 08:51:42

240

Cross-linguistic Influences in the Second Language Lexicon

References Alexander, R.J. (1987) Problems in understanding and teaching idiomaticity in English. Anglistik und Englischunterricht, 32, 105–22. B­ba, S. and Liberek, J. (2002) SÙownik Frazeologiczny WspóÙczesnej Polszczyzny. Warszawa PWN. Cacciari, C. (1993) The place of idioms in literal and metaphorical world. In C. Cacciari and P. Tabossi (eds) Idioms: Processing, Structure, and Interpretation (pp. 27–56). Hillsdale, NJ: Lawrence Erlbaum Associates. Cacciari, C. and Glucksberg, S. (1991) Understanding idiomatic expressions: The contribution of word meanings. In G.B. Simpson (ed.) Understanding Word and Sentence (pp. 217–40). Amsterdam: North-Holland. Cacciari, C. and Tabossi, P. (1988) The comprehension of idioms. Journal of Memory and Language 27, 668–83. Cacciari, C. and Tabossi, P. (eds) (1993) Idioms: Processing, Structure, and Interpretation. Hillsdale, NJ: Lawrence Erlbaum Associates. Carter, R. (1998) Vocabulary: Applied Linguistics Perspectives. London and New York: Routledge. Fernando, C. (1996) Idioms and Idiomaticity. Oxford: Oxford University Press. Fraser, B. (1979) The interpretation of novel metaphors. In A. Ortony (ed.) Metaphor and Thought (pp. 172–85). Cambridge: Cambridge University Press. Gairns, R. and Redman, S. (1986) Working with Words: A Guide to Teaching and Learning Vocabulary. Cambridge: Cambridge University Press. Gibbs, R.W. (1993) Why idioms are not dead metaphors. In C. Cacciari and P. Tabossi (eds) Idioms: Processing, Structure, and Interpretation (pp. 57–77). Hillsdale, NJ: Lawrence Erlbaum Associates. Gibbs, R.W. and Nayak, N.P. (1989) Psycholinguistic studies on the syntactic behavior of idioms. Cognitive Psychology 21, 100–38. Gibbs, R.W., Nayak, N.P. and Cutting, C. (1989a) How to kick the bucket and not decompose: Analyzability and idiom processing. Journal of Memory and Language 28, 576–93. Gibbs, R.W., Nayak, N., Bolton, J. and Keppel, M. (1989b) Speakers’ assumptions about the lexical flexibility of idioms. Memory and Cognition 17, 58–68. Irujo, S. (1986) Don’t put your leg in your mouth: Transfer in the acquisition of idoms in a second language. TESOL Quarterly 20/2, 287–304. Irujo, S. (1993) Steering clear: Avoidance in the production of idoms. International Review of Applied Liguistics 21, 205–19. Jaworska, T. (2002) SÙownik Frazeologiczny Angielsko-Polski i Polsko-Angielski. Warszawa: Wydawnictwa Naukowo-Techniczne. Kovesces, Z. and Szabo, P. (1996) Idioms: A view from cognitive semantics. Applied Linguistics 17 (3), 326–55. Lattey, E. (1986) Pragmatic classification of idioms as an aid for the language learner. IRAL 24 (3), 217–33. Liontas, J. (2002) Context and idiom understanding in second languages. EUROSLA Yearbook 2 (pp. 155–85). John Benjamins Publishing Company. Longman Idioms Dictionary (1998) Essex: Pearson Education Limited. Makkai, A. (1972) Idiom Structure in English. The Hague, Netherlands: Mouton. McCarthy, M. (1990) Vocabulary. Oxford: Oxford University Press.

9612Arabski.indb 240

13/01/2006 08:51:42

On Building Castles on the Sand

241

Moon, R. (1997) Vocabulary connections: Multi-word items in English. In N. Schmitt and M. McCarthy (eds) Vocabulary: Description, Acquisition and Pedagogy (pp. 40-63). Cambridge: Cambridge University Press. Nicolas, T. (1995) Semantics of idiom modification. In M. Everaert, E. van der Linden, A. Schenk and R. Schreuder (eds) Idioms: Structural and Psychological Perspectives (pp. 233–52). Hillsdale, NJ: Lawrence Erlbaum Associates. Puhvel, J. (ed.) (1969) Substance and Structure of Language. Los Angeles: University of California Press. Schenk, A. (1995) The syntactic behavior of idioms. In M. Everaert, E. van der Linden, A. Schenk and R. Schreuder (eds) Idioms: Structural and Psychological Perspectives (pp. 253–72). Hillsdale, NJ: Lawrence Erlbaum Associates. Simpson, G.B. (ed.) (1991) Understanding Word and Sentence. Amsterdam, Netherlands: Elsevier. Titone, D.A. and Connine, C.M. (1994) Comprehension of idiomatic expressions: Effects of predictability and literality. Journal of Experimental Psychology: Learning, Memory, and Cognition 20 (5), 1126–38. Titone, D.A. and Connine, C.M. (1999) On the compositional and noncompositional nature of idiomatic expressions. Journal of Pragmatics 31 (12), 1655–74. Weinreich, U. (1969) Problems in the analysis of idioms. In J. Puhvel (ed.) Substance and Structure of Language (pp. 23–81). Los Angeles: University of California Press. Wolfram-Romanowska, D., Kaszubski, P. and Parker, M. (1999) Idiomy PolskoAngielskie. Warszawa: PWN. Wright, J. (1999) Idioms Organizer: Organized by Metaphor, Topic, and Key Word. Hove: Language Teaching Publications. Zughoul, M.R. (1991) Lexical choice: Towards writing problematic word lists. IRAL 29 (1), 45–60.

9612Arabski.indb 241

13/01/2006 08:51:43

242

Cross-linguistic Influences in the Second Language Lexicon

Appendix 1 A list of English idioms employed in the experiment, along with their Polish counterparts. Literal translations of Polish Semi-Lexical Level and Post-Lexical Level idioms are provided in brackets. English idioms

Polish equivalents LEXICAL LEVEL IDIOMS

loosen somebody’s tongue

rozwi­za° komuï j¿zyk

take the bull by the horns

wzi­° byka za rogi

put one’s head in the sand

chowa° gÙow¿ w piasek

to swallow the bait

poÙkn­° przyn¿t¿

be the alpha and omega

by° alf­ i omeg­

move heaven and earth

poruszy° niebo i ziemi¿

swim against the current

pÙyn­° pod pr­d

SEMI-LEXICAL LEVEL IDIOMS cry one’s heart out

wypÙakiwa° sobie oczy (cry one’s eyes out)

add fuel to the fire

dolewa° oliwy do ognia (add olive to the fire)

go like hot cakes

iï° jak ciepÙe buÙeczki (go like warm buns)

promise somebody the moon

obiecywa° zÙote gory (promise somebody golden mountains)

be tied to one’s mother’s apron strings

trzyma° si¿ matczynej spódnicy (to hold on to one’s mother’s skirt)

build castles in the air

budowa° zamki na piasku/na lodzie (build castles on the sand/on ice)

sleep like a log

spa° jak suseÙ (sleep like a gopher) POST-LEXICAL LEVEL IDIOMS

keep a stiff upper lip

zachowa° zimn­ krew (to maintain cold blood)

be all fingers and thumbs

mie° dwie lewe r¿ce (have two left hands)

face the music

stawi° (czemuï) czoÙa (to put your forehead against something)

be the apple of sb’s eye

by° czyimï oczkiem w gÙowie (to be somebody’s eye in the head)

9612Arabski.indb 242

13/01/2006 08:51:44

On Building Castles on the Sand

243

have an axe to grind

upiec wÙasn­ pieczeÚ przy cudzym ogniu (to roast your own roast over sb else’s fire)

keep a high profile

rzuca° si¿ w oczy (throw oneself in eyes)

beat around the bush

owija° w baweÙn¿ (wrap in cotton)

Appendix 2 A sample entry for the discourse completion task:

Susan only just met Rickey in a pub and they started drinking wine and talking. Rickey was a nice guy and Susan felt she was growing closer and closer to him. Three-quarters of a bottle of wine had ……………………… her tongue and she found herself telling Rickey her life story

Appendix 3 A sample entry for the translation task:

Po doï° formalnej ceremonii otwarcia, zarz­d firmy zaprosiÙ nowych pracowników na przyj¿cie. Darmowy alkohol szybko rozwi­zaÙ ludziom j¿zyki i amosfera staÙa si¿ bardziej rozluĊniona. After a rather formal opening ceremony, the company management invited the new employees to a party. The free alcohol quickly …………………… and the atmosphere became more relaxed.

Appendix 4 Examples of metacognitive comments exemplifying a given strategy ⇒ ⇒

Prior knowledge/ Familiarity I’ve known the idiom beat about the bush too long to even have images about it! Face the music – no association, simply learned it by heart!

9612Arabski.indb 243

13/01/2006 08:51:45

244

⇒ ⇒

⇒ ⇒ ⇒

⇒

⇒

⇒

⇒

⇒ ⇒

Cross-linguistic Influences in the Second Language Lexicon

Guessing based on literal analysis Grinding an axe is extremely difficult, it seems, because an axe is hard, so to have an axe to grind probably means mie° twardy orzech do zgryzienia. ‘Profile’ for me is an association with nose and ‘high’ indicates ‘zarozumiaÙoï°’; so keep a high profile means zadziera° nosa, by° zarozumiaÙym. Guessing based on one key word ‘To grind’ means ‘rozdrabnia°’, so have an axe to grind can mean by° drobiazgowym. Only the word ‘face’ helped me to interpret this idiom; I don’t know why the word ‘music’ is there? An apple – you know where this sorry incident with Adam and Eve led us; it’s an icon of sexuality, so be the apple of sb’s eye means to be erotically tasty for sb or to be attractive for them. Guessing based on two key words ‘Tongue’ – association with speech; ‘to loosen’ – mie° luĊny j¿zykrozpowiada° o wszystkim co si¿ sÙyszaÙo; so the idiom means to gossip; to say the things you shouldn’t say; in Polish, nie trzyma° j¿zyka za z¿bami. A ‘loose tongue’ would probably mean a tongue which is not restricted in any way; able to utter anything that a person thinks; therefore the process of ‘loosening sb’s tongue’ is an action leading to his carelessness in speech. Analogy between literal and figurative senses ‘Air’ means something ephemeral and ‘castles’ mean buildings, projects, plans, or dreams. Therefore, if one builds castles in the air they have high dreams, hopes, projects, and plans that are not likely to succeed. The apple is a very good and precious thing as well as the eye so if someone is the apple of somebody’s eye it means that he or she is extremely important. Reference to the person/ action/event denoted by the literal meaning of the idiom I have an image of someone putting his/her head in the sand. I think of a person that fears nothing and is not afraid to take risk. For example, during corrida a matador fighting with the bull.

9612Arabski.indb 244

13/01/2006 08:51:45

On Building Castles on the Sand

⇒

⇒ ⇒

⇒ ⇒ ⇒

⇒

⇒ ⇒

245

Some people fall asleep and no sound around them may wake them up. You could take the person on a boat and he or she wouldn’t wake up. Literal based image I see a mythical person trying to move earth and heaven; to shake it by the power of his/her will. I see a person who is lost in the bushes and goes around them, not knowing what to do. (beat around the bush) Reference to or association with a Polish idiom I thought about a Polish idiom, which sounds similar: rozwi­za° komuï j¿zyk. I saw a moon and stars and associated this with a Polish proverb: Dam ci gwiazdk¿ z nieba – something impossible to achieve. I try to translate the idiom (loosen sb’s tongue) into Polish and try to guess or find out whether I know an equivalent in my language. Combination of strategies Lips shake when a person is touched by great emotions; e.g., wants to stop tears from falling; so a stiff upper lip will be associated with a calm person and the idiom might mean to be serious, calm and balanced. (literal image + guessing) When I saw the word ‘eye’, I had an association with the Polish saying by° czyimï oczkiem w gÙowie. (key word + association with Polish idiom) I saw a big axe and small grinds of coffee to grind. It seems impossible for me to be able to cut the grinds with a big axe, so the idiom is used to describe a situation in which you have inadequate means to achieve sth, like in Polish próbowa° zÙapa° komara w r¿kawicach bokserskich. (literal image + guessing based on literal analysis)

9612Arabski.indb 245

13/01/2006 08:51:46

Chapter 18

‘Don’t Lose Your Head’ or How Polish Learners of English Cope with L2 Idiomatic Expressions LILIANA PIASECKA The study investigates how advanced Polish learners of English cope with English idiomatic expressions. The hypothesis was made that if the learners are faced with L1 idioms that have identical counterparts in L2, then they will use exact translation from L1 and thus will show the effects of transfer. A total of 59 university students and 31 teacher training college students of English were asked to provide L2 equivalents to 23 L1 idioms in two tasks: contextualised fill-in translation task and decontextualised discrete items translation task. Results show that learners used a variety of strategies in the tasks, including exact translation which, however, was neither the major nor the only strategy they used. This suggests that transfer from L1 is one of many factors that affect language production. What’s in a name? That which we call a rose By any other name would smell as sweet. (William Shakespeare, Romeo and Juliet. Act 2, scene 2)

Introduction Whether language transfer or cross-linguistic influence, the knowledge of the world, including the knowledge of language(s), that the L2 learner brings to the task of acquiring another language cannot be deleted or ignored. It is recognised as an important factor (James, 1994; Laufer, 2000; Nickel et al., 1999) which does not have to result in error – a mortal sin for behavioural psychologists – but which may facilitate the learning process. It can lead to avoidance of forms that the languages in contact do not share; it may also affect the rate of development either by slowing it down or by accelerating it. It can influence the route of acquisition (as observed in Chinese learners of English) and may manifest itself in the overproduction of certain forms, e.g. speakers of Romance languages tend to overuse Latinate cognates (Benson, 2002). 246

9612Arabski.indb 246

13/01/2006 08:51:47

247

‘Don’t Lose Your Head’

Moreover, transfer defined as ‘the influence resulting from similarities and differences between the target language and any other language that has been previously (and perhaps imperfectly) acquired’ (Odlin, 1989: 27) can be observed in all linguistic subsystems, or at all levels of language use. Since the use of prior linguistic knowledge also depends on the type of context (formal/informal), the age of the learners, typological proximity of the languages in contact as well as on the learners’ levels of proficiency, and their linguistic awareness (Odlin, 1989), the degree to which the learner will resort to its use will vary.

Transfer Revisited In an eloquent discussion of Contrastive and Error Analysis in the early 1990s, James (1994) presents a number of refinements to the concept of transfer. The refinements and reformulations come from researchers involved in Contrastive and Error Analysis. Thus Kellerman and Sharwood-Smith (1986) (in James 1994) suggest to replace the term ‘transfer’ with the term ‘crosslinguistic influence’ to refer to ‘more subtle kinds of L1 influence’ (1994: 181). Avoidance is an example of such influence and it may result from the learner’s prior experience with transfer, i.e. in the past the learner tried to transfer some L1 form to L2, which caused an error, and the learner received negative feedback for the attempt (Gass, 1988 in James, 1994). Investigating the influence of L1 on the learner’s confusion with similar lexical forms (‘synforms’), Laufer (1992) distinguishes between indirect or systemic and direct transfer. Learners in whose L1 (e.g. German) words are built by means of affixation, transfer this principle to L2 (e.g. English), thus showing indirect or systemic transfer. Direct transfer occurs when similar lexical forms appear in the learners’ L1 (e.g. French simuler/stimuler), so they know the difference and they will transfer it to L2 (e.g. English simulate/stimulate). Table 18.1 Transfer types (on the basis of James, 1994) Researcher

Transfer types

Laufer (1992)

Indirect/systemic

Direct

Cook (1992)

Diachronic (systemic)

Synchronic (direct)

James (1988)

Primary

Secondary

Hammerly (1991)

Intrusive interference

Inhibitive interference

Nemser (1991)

Direct

Secondary (Indirect or ‘mediated’)

9612Arabski.indb 247

13/01/2006 08:51:47

248

Cross-linguistic Influences in the Second Language Lexicon

Cook (1992) proposes two types of transfer: diachronic and synchronic. The former is observed when the learner learns the language, the latter – when the learner uses, i.e. encodes or decodes messages. James (1994) claims that Cook’s synchronic transfer corresponds to Laufer’s direct, as it is ‘item-bound’, whereas diachronic corresponds to indirect or systemic transfer. James (1994) himself adds one more distinction based on his study of transfer and translation (James, 1988) and so he discusses primary transfer, which is ‘the spontaneous, untaught strategy of each individual learner’ (James, 1994: 183), and secondary transfer, i.e. the individual’s ‘legacy from the community in a language contact situation’ (1994: 183). Intrusive interference is observed when L1 forms occur in L2, but Hammerly (1991) argues that learners may decide not to learn and use L2 forms that are absent in their L1 and this he terms ‘inhibitive interference’. Hence the learner’s use of wrong forms is due to intrusion, whereas omission of certain forms is caused by inhibition. Nemser (1991) assumes that transfer should be discussed in terms of three language systems: L1, L2 and IL. Direct transfer operates when L1 or L2 forms appear in the learner’s IL. However, the learner may use L1 or the current IL as an ‘adaptor of importations’ (p. 351) from L2, and this is when secondary, indirect or ‘mediated’ transfer occurs, for example: L2 (English) cider road

L1 (Polish) jabÙecznik droga

IL I baked some cider (meaning apple pie) Road Mum (the beginning of a letter)

Although confusing at times (e.g. direct transfer), transfer terminology shows the researchers’ attempt to capture various aspects of this phenomenon. It also suggests that the learner’s knowledge of the world, prior experience with language learning and language use, contexts of language use, as well as the learner’s personality can account for transfer. Whatever name is used, transfer is at work.

Traditional and Cognitive View of Idioms As transfer occurs at all language levels, it also affects lexicon which comprises idioms, defined as ‘conventionalised expression(s) whose meaning cannot be determined from (their) parts’ (Irujo, 1986: 288), or phrases ‘whose meaning is difficult or sometimes impossible to guess by looking at the meanings of the individual words’ they contain (Oxford Advanced Learners’ Dictionary, 2000). They are also referred to as ‘lexical phrases’ (Nattinger & DeCarrico, 1992), or ‘formulae’ (Moon, 1998)1.

9612Arabski.indb 248

13/01/2006 08:51:48

‘Don’t Lose Your Head’

249

According to Kövecses and Szabó (1996), the approach to idioms presented above is traditional in the sense that idioms are a special category of words and as such ‘they are taken to be items of lexicon that are independent of any conceptual system’ (p. 328) and of each other. The authors argue that most idioms are conceptual in nature because they ‘arise from our more general knowledge of the world (embodied in our conceptual system)’ (p. 330). This knowledge can help the language user/learner to understand the meanings of idioms, thus making them motivated and not arbitrary.

Comparing Idioms A three-dimensional framework for comparing language forms has been proposed by Laufer (2000) and includes concept, form and distribution. If a selected category is present in two languages, they are conceptually similar, if it is present in only one language – they are conceptually different. Formal similarity is observed when the intended meaning is expressed by the same linguistic means in two languages, and distributional similarity occurs when a given category serves the same purpose in two languages. Within this framework, idioms in Polish and English are conceptually similar, though they may differ formally and distributionally. However, idioms that show total formal similarity (e.g. Pol. mie° zwi­zane r¿ce, Eng. have one’s hands tied) carry a promise for a researcher since they may be rightfully expected to induce positive transfer. From the perspective of cognitive semantics, idioms can represent the relations between two domains of knowledge (metaphor-based idioms) or can refer to one Figure 18.1 The conceptual motivation for many idioms (Kövecses and Szabó, 1996: 331) Idiomatic meaning: the overall special meaning of an idiom Cognitive mechanisms: metaphor, metonymy, conventional knowledge Conceptual domain(s) one or more domains of knowledge Linguistic forms and their meanings: the words that comprise an idiom, their syntactic properties, together with their meanings

9612Arabski.indb 249

13/01/2006 08:51:49

250

Cross-linguistic Influences in the Second Language Lexicon

domain (through the cognitive mechanisms of metonymy and conventional knowledge). Following this way of thinking, the idiom have one’s hands tied is motivated by the metonymy THE HAND STANDS FOR THE ACTIVITY (Kövecses & Szabó, 1996: 340). Therefore, idioms selected for the study are formally similar and conceptually motivated.

The Study The aim of the study described below was to find out how advanced learners of English would cope with a translation task including L1 idioms that are formally identical to L2 idioms. The assumption was made that such a task may induce transfer of L1 forms into L2. The following research questions were formulated: How will advanced learners of English render Polish idiomatic expressions in English? Will they translate Polish idioms into English, thus showing the effects of transfer? Will they avoid translation? Will they avoid idiomatic expressions in L2? Subjects A total of 90 subjects participated in the study; 59 were first-year students at the English Department of Opole University and they will be referred to as Group 1; 31 were second-year students of English at the Teacher Training College in Opole and they will be called Group 2. The division into two separate groups has been introduced because the groups vary in terms of the educational institution (college-university) as well as in terms of the length of instruction at a given institution (one year – two years). Table 18.2 summarises the most important characteristics of the groups. Both groups have been learning English for 9.2 years but more subjects from Group 2 know another foreign language which may be due to the fact that they have to take another foreign language. The groups differ in the number of foreign languages they know and so more university students than college students declare to know two foreign languages. The information in columns 5 and 6 shows the subjects’ proficiency in English as measured by the mean grade they got for Practical English courses (there is a difference between groups) and all the courses they had to take in winter term (almost no difference). University students’ self-assesment of proficiency in English on a scale from 1 to 7 is higher than that of college students.

9612Arabski.indb 250

13/01/2006 08:51:50

9612Arabski.indb 251

3 YES: 46 (78%) NO: 13 (22%)

YES: 27 (87%) NO: 4 (13%)

2 M = 9.2 Range: 11 Low: 4 High: 15

M = 9.2 Range: 12 Low: 3 High: 15

1

Group 1 (N=59) 1st year University students

Group 2 (N=31) 2nd year TTC students

Other foreign languages

Years of English

Group

1 FL: 20 (74.07%) 2 FL: 4 (12.90%) 3 FL: 2 (6.45%)

1 FL: 31 (67.39%) 2 FL:13 (22.03%) 3 FL: 2 (3.38%)

4

Number of FLs

Table 18.2 Group characteristics (M = mean value)

5

M=3.72

M=4.18

Average grade for Practical English courses 6

M=4.08

M=4.09

Average grade for all the courses 7

M=4.32

M=4.59

Self assessment of proficiency

‘Don’t Lose Your Head’ 251

13/01/2006 08:51:51

252

Cross-linguistic Influences in the Second Language Lexicon

Materials A short questionnaire and two translation tasks (Appendix 1) were prepared. The questionnaire included seven questions to identify learner characteristics (summarised in Table 18.2) and 13 Polish idioms connected with parts of the body were included in Task 1 – a partial translation of a conversation between two friends. I decided to put the idioms into a wider context to find out what coping strategies the subjects will use. Moreover, the selected idioms refer to the domain of knowledge that the subjects are well familiar with both in their L1 and L2. Names of various parts of human body are taught at early stages of language learning. Task 2 consisted in translating 10 Polish idioms into English. This time no context was provided. Analyses Subjects’ responses from two tasks were recorded and analysed separately. The responses were divided into five classes, i.e.: • • • •

•

exact L1–L2 translations resulting in the target idiom; parts of idioms were translated, which did not result in target idioms; another English idiom was used; non-idiom was used (i.e. the subjects provided expressions that reflected the meaning of L1 idioms but were non-idiomatic, e.g. where the target idiom was he always falls on his feet; the subjects wrote he is able to cope with difficulties, or he can deal with every situation) no answer was given.

After the responses had been classified and coded as belonging to any of the classes specified above, frequences of classes of responses were counted for the two tasks. To show how the two groups compare in terms of coping with the tasks, i.e. which way of coping they select, the frequences were turned into per cent.

Results The results of the classifications and calculations are presented in the following tables. The N for each task was calculated in the following way: number of subjects × number of idioms in the task. The data show that in Task 1 (idioms in context), the subjects used a number of ways to provide the answers. The most frequent response (apart from ‘no answer’) was exact translation from L1, then a non-idiomatic response followed by partial translation and another L2 idiom.

9612Arabski.indb 252

13/01/2006 08:51:51

253

‘Don’t Lose Your Head’

Table 18.3 Performance of all the subjects (N=90) on two tasks Exact L1–L2 Partial L1–L2 Another translation translation L2 idiom 1

2

TASK 1 305 144 (N = 1170) 26.07%

12.30%

TASK 2 (N = 900)

23.33%

186

210 20.87%

Non-idiom

No answer

4

5

3

127 280 314 10.85% 23.90% 26.81% 31

45

428

3.44%

5%

47.55%

In Task 2 (a list of idioms) the response pattern is different. The most noticeable feature is the lack of answer (almost half of all responses). The second most frequent response is partial translation from L1 to L2, then exact translation, non-idiom and another L2 idiom. As the groups differ in terms of the length of instruction at the educational institution (first-year students versus second-year students) as well as the type of educational institution (University and Teacher Training College), the results for the groups have also been presented in separate tables to see if the two variables have any effect on the response patterns. Thus, Table 18.4 shows Group 1 results, Table 18.5 includes the results from Group 2. Again, there are differences in the subjects’ performance on two tasks. In Task 1, non-idiom responses are most frequent, then no responses appear. Exact L1–L2 translation comes third and is followed by another idiom and partial translation. In Task 2, no answer responses dominate. Then most frequent is partial translation, exact translation, non-idiom and another idiom. Responses to Task 1 by college students reveal that when they decide to take the risk, they use exact translation, then non-idioms, partial translation and another idiom. In Task 2, over half of the responses were not given. When given, these were exact translation and partial translation, then non-idioms and other idioms. Table 18.4 Performance of Group 1 (N=59) on two tasks Exact L1–L2 Partial L1–L2 Another translation translation L2 idiom

Non-idiom

No answer

TASK 1 (N = 767)

171

212

211

TASK 2 (N = 590)

110

9612Arabski.indb 253

86 22.29%

87 11.21%

155 18.64%

11.34% 22

26.27%

27.64% 32

3.73%

27.50% 271

5.42%

45.93%

13/01/2006 08:51:52

254

Cross-linguistic Influences in the Second Language Lexicon

Table 18.5 Performance of Group 2 (N=31) on two tasks Exact L1–L2 Partial L1–L2 Another translation translation L2 idiom

Non-idiom

No answer

TASK 1 (N = 403)

134

68

102

TASK 2 (N = 310)

76

59 33.25%

40 14.64%

55 24.52%

9.92% 9

17.74%

16.87% 13

2.90%

25.31% 157

4.19%

50.64%

Response classes identified for the groups are presented in Table 18.6. They have been ordered from most to least frequent. As already observed, the tasks themselves generated different types of responses. In Task 1, both groups first tried to provide an equivalent of Polish expessions, either using a paraphrase or circumlocution (non-idiom) (Group 1), or exact L1–L2 translation (Group 2). The next most frequent behaviour in Task 1 was ‘no answer’, shared by both groups. Then groups differ in terms strategies used. In Task 2, the most frequent response for both groups was ‘no answer’, followed by partial and exact translation in Group 1, and by exact and partial translation in Group 2. Then both groups used non-idiomatic expressions and other L2 idioms.

Table 18.6 Classes of groups’ responses on the tasks, ranked from most frequent (1) to least frequent (5) Task 1 Group 1

Task 2 Group 2

Group 1

Group 2

1

Non-idiom

Exact L1–L2 translation

No answer

No answer

2

No answer

No answer

Partial translation

Exact L1–L2 translation

3

Exact L1–L2 translation

Non-idiom

Exact L1–L2 translation

Partial translation

4

Another L2 idiom

Partial translation

Non-idiom

Non-idiom

5

Partial translation

Another L2 idiom

Another L2 idiom

Another L2 idiom

9612Arabski.indb 254

13/01/2006 08:51:53

‘Don’t Lose Your Head’

255

Discussion On the basis of the data presented above research questions may be answered. Advanced Polish learners of English vary in their responses to the task of translating idioms from Polish into English. Thus, as far as Task 1 is concerned, Group 1 subjects use non-idiomatic expressions (paraphrase, circumlocution), although Group 2 subjects use exact L1–L2 translation, which resulted in providing accurate target forms. In Group 1, exact translation was third in terms of frequency. This may suggest that Group 1 subjects, who have been studying English Language and Literature at the University for less than a year, either are not willing to translate directly from one language to another, or their L2 vocabulary may not be sufficient to let them translate with confidence. Therefore, they used expressions whose meaning they were familiar with and confident about. On the other hand, subjects from Group 2, who have been studying English Language and Literature at the College for almost two years, use exact L1–L2 translation in Task 1 most frequently, although this is not their only strategy. They may be more familiar with certain English idioms and their L2 lexical knowledge is probably more extensive than university students’, so they are more confident when translating. Moreover, when the tasks were completed, I asked them why they did not translate directly and they told me that they had been told not to translate. In Task 2, the subjects most frequently decided not to take the risk of making a mistake and gave no anwer. However, it is in this task where they used partial and exact translation (Group 1) and exact or partial translation (Group 2). When the subjects used translation, it resulted either in correct target forms thus showing positive transfer, or in incorrect (although communicatively acceptable) forms, thus possibly pointing to negative transfer from L1 or other languages they know, although other factors might have been involved. For example, when translating igra° ze ïmiercia, the subjects wrote to play with death, to play with fire, to tease with death, to dance on the egde. To play with death appeared quite frequently in the data which may suggest that the subjects were familiar with the English expression to play with fire and decided to transfer it to another context. Actually, the class of responses classified as ‘partial translation’ provides some evidence that transfer, both from L1 and L2, does work Thus, one subject translated Ona jest zawsze pod r¿k­ (She’s always at hand) as *She’s always to my dispositon (Pol. Jest zawsze do mej dyspozycji). Such cases, however, were not frequent.

9612Arabski.indb 255

13/01/2006 08:51:54

256

Cross-linguistic Influences in the Second Language Lexicon

Another factor accounting for incorrect forms may be the subjects’ confusion about words with similar meanings. When translating prawo dČungli (the law of the jungle), instead of law, they used such forms as right(s), rule(s), principle(s) and produced jungle rule, jungle’s right, the rights of the jungle. As regards idioms avoidance, the data suggest that they were not avoided in Task 1, although they were less frequent in Task 2 (see Table 18.3, columns 1 and 3). The lack of answer does not have to imply avoidance but may signal inadequate linguistic resources, or the subjects’ lack of motivation to provide any answer. Finally, the best known idioms, i.e. the ones the subjects translated correctly most frequently, for example to do something behind someone’s back; to keep one’s fingers crossed/to cross one’s fingers; to lose one’s head; to twist somebody round one’s (little) finger; to lose one’s face, to have something at the tip of the tongue are more familiar to the students and probably used more frequently both in L1 and L2.

Conclusion I hypothesised that if advanced Polish learners of English are asked to render Polish idiomatic expressions in English, they would primarily resort to exact translation from L1 to L2 as a strategy to cope with the task. In addition, Polish idioms were selected in such a way as to show total formal similarity with English idioms. Such a way of idiom selection is based on cognitive semantics, especially on the idea that idioms are conceptually motivated by general world knowledge shared by people. The results of the study show that direct translation is neither the major nor the only way of coping with idiomatic expressions. The subjects used a variety of ways (exact translation, paraphrase, partial translation, other idioms) to render L1 meanings in L2, which points to their flexibility in language use. Moreover, it appears that the kind of eliciation test influences the kind of responses. A contextualised test (Task 1) seems to motivate subjects to complete it successfully more than a discrete item test (Task 2), which is an important implication for testing the use of language forms. Note 1. A detailed discussion of fixed expressions and idioms can be found in my paper entitled ‘Sensitivity to fixed expressions and idioms – an aspect of lexical competence’ in J. Arabski (ed.) (2002) Time for Words. Studies in Foreign Language Vocabulary Acquisition. Peter Lang: Frankfurt am Main

9612Arabski.indb 256

13/01/2006 08:51:55

‘Don’t Lose Your Head’

257

References Benson, C. (2002) Transfer/Cross-linguistic influence. ELT Journal 56 (1), 68–70. Cook, V.J. (1992) Evidence for multicompetence. Language Learning 42 (4), 557–91. Gass, S. (1988) Second language acquisition and linguistic theory: The role of language transfer. In S. Flynn and W.O’Neill (eds) Linguistic Theory and Language Acquisition (pp. 384–403). Dordrecht: Kluwer. Hornby, A.S. (2000) Oxford Advanced Learners’ Dictionary of Current English (6th edn). Wehmeier S. (ed). Oxford: OUP. Hammerly, H. (1991) Fluency and Accuracy. Clevedon: Multilingual Matters. Irujo, S. (1986) Don’t put your leg in your mouth: Transfer in the acquisition of idioms in a second language. TESOL Quarterly 20 (2), 287–304. James, C. (1988) Perspectives on transfer and translation. The Linguist 27 (1), 45–8. James, C. (1994) Don’t shoot my dodo: On the resilience of contrastive and error analysis. International Review of Applied Linguistics XXXII (3), 179–200. Kellerman, E. and Sharwood-Smith, M. (1986) Crosslinguistic Influence in Second Language Acquisition. New York: Pergamon. Kövecses, Z. and Szabó, P. (1996) Idioms: A view from cognitive semantics. Applied Linguistics 17 (3), 326–55. Laufer, B. (1992) Native language effect on confusion of similar lexical forms. In C. Mair and M. Markus (eds) New Departures in Contrastive Linguistics: Innsbrucker Beiträge zur Kulturwissenschaft. Vol. 2 (pp. 199–209). Innsbruck. Laufer, B. (2000) Avoidance of idioms in a second language: The effect of L1–L2 similarity. Studia Linguistica 54 (2), 186–96. Moon, R. (1998) Fixed Expressions and Idioms in English. A Corpus-based Approach. Oxford: Clarendon Press. Nattinger, J. and DeCarrico, J.S. (1992) Lexical Phrases and Language Teaching. Oxford: OUP. Nemser, W. (1991) Language contact and language acquisition. In V. Ivir and D. Kalogjera (eds) Languages in Contact and Contrast: Essays in Contact Linguistics (pp. 345–64). Berlin: Mouton De Gruyter. Nickel, G., Nakamura, Y., Davies R., Mizuno, M. and Lutjeharms, M. (1999) Multidimensional contrastive studies: From microlinguistics to macrolinguistics. Symposium held at the 12th AILA World Congress. Tokyo, Japan. Odlin, T. (1989) Language Transfer. Cambridge: CUP

Appendix: Translation tasks In TASK 1 the subjects were asked to translate selected expressions from Polish into English. Target idioms are given in bold. Last evening my best friend dropped in. (1) I welcomed her with open arms. I like her very much. Besides, when things get nasty, (2) she’s always at hand. But yesterday she was very upset. ‘What happened?’, I asked ‘I’m fed up with it! This time my man really (3) got under my skin. You know, (4) he always has an eye on some business. Now he has found a lousy joint and he wants to take it over’, she said.

9612Arabski.indb 257

13/01/2006 08:51:56

258

Cross-linguistic Influences in the Second Language Lexicon

‘What’s wrong with it?’ I enquired. (5) ‘He always falls on his feet’ ‘Yes, but this time he wants me to do some of the job. You know. I have accounting (6) at my finger tips and I know what’s up.’ ‘Perhaps it will work this time. Maybe you will (7) stand on your feet at last’. ‘Maybe you’re right but you know that (8) I have my hands full’, my friend complained. (9) ‘Don’t lose your head . I’m sure you’ll manage. (10) You can twist him round your (little) finger’. ‘Well, if you say so. We may try. But when he comes to you, don’t tell him what I told you. He doesn’t like when (11) I do something behind his back. ‘Don’t worry.’, I told her. ‘I shall (12) hold my tongue’, I promised. And (13) keep your fingers crossed for us / cross your fingers for us’, my friend added. In TASK 2 the subjects were asked to translate Polish idiomatic expressions into English. The English idioms that were the target are given below. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

a drop in the ocean the law of the jungle a bone of contention to stick/cling like a leech to make a martyr of oneself to shed crocodile tears to dice with death to lose face to have something at the tip of one’s tongue over my dead body

9612Arabski.indb 258

13/01/2006 08:51:57

Chapter 19

Phrasal Verb Idioms and the Normative Concept of the Interlanguage Hypothesis PRZEMYSŁAW OLEJNICZAK Introduction In the history of foreign language learning theory and research, the generativist or cognitivist ‘interlanguage hypothesis’, based on the internalised concept of a system or systems of significant errors (Corder, 1967; Selinker, 1972) can be compared with a non-dogmatic view of interlanguage, including the concept of language transfer. It is my intention to comment on the data on one type of idiomatic lexical units and the results of a research sample in this area of English lexis. As regards language competence, this paper focuses on the product of ‘social’ interaction or an instance of communication in which advanced students are to approximate the target language lexical code. Initially, we need to clarify the concept of the idiomatic phrasal verb, as well as some second language acquisition (SLA) issues.

Phrasal Verbs as Phraseological Units The institutionalised, semantically specified combinations of root verbs with adverbial particles and/or prepositions, such as come over, get by, turn down, sink in, go at, look forward to, come up with, take-off, have always troubled analysts, especially those who deal with English syntax, lexicography, phraseology, etc. Along with such terms as ‘separable compounds’, ‘polyword verbs’ (Sroka, 1972), idiomatic ‘B-sheet formations’ (Makkai, 1972), scholars have proposed hierarchies of fixedness (Fraser’s, 1976, structural ‘cases’), and the theories of metaphorical compounds, where phrasal verbs are unique but grammatically analyseable combinations, rather than units of vocabulary. The meaning of the particle or the whole verb phrase (as in make up a face) can be metaphorically extended. The more ‘figurative’ it becomes, the closer it is to the ‘second-level stereotypes’ (compare: come 259

9612Arabski.indb 259

13/01/2006 08:51:57

260

Cross-linguistic Influences in the Second Language Lexicon

across a field and come across a word), whereas ‘idioms’ are restricted to fixed expressions containing phrasal verbs (turn over a new leaf), called ‘thirdlevel stereotypes’ (Bolinger, 1971: 113–15). In a number of dictionaries (Longman Dictionary of Phrasal Verbs, Oxford Dictionary of Phrasal Verbs, Chambers Dictionary of Phrasal Verbs, Collins COBUILD Dictionary of Phrasal Verbs, NTC’s Dictionary of Phrasal Verbs), and such handbooks as J. Seidl’s (1999), and Scheffler’s (1994), phrasal verbs, however the boundary of this peculiar verbal phrase in context is understood, have always been considered as a special class of vocabulary. Nevertheless, their frequency of occurrence, varying degrees of idiomaticity, and syntactic ‘boundary’ problems (discussed below) have turned out to be uncomfortable facts to some analysts, and, as remains to be seen, language educators, from theoretical, systematic, and practical points of view.

The Notion of Idiomaticity In order for the terminology not to be used casually, it has to be based on a few, possibly weak, assumptions. The notion of idiomaticity has traditionally comprised such ideas, concepts or criteria as non-compositonality (in the weakest sense of a special meaning within a context – compare Hockett, 1958; Makkai, 1972), accompanied by the ‘not-the-sum-of-itsparts’ slogan (Fernando & Flavell, 1981; Makkai, 1995; Strässler, 1982), nonproductivity (Clausner & Croft, 1997), institutionalisation and conventionality (Nunberg et al., 1994), or the relative notion of fixedness (Moon, 1998). A ‘working definition’ of idiom would describe it as a multi-word linguistic unit which is often reproduced, in a fixed or variable form, conventionally understood to have a specific sense or senses different from the component meanings, their stylistic or communicative function, etc. It seems, however, that few of the above criteria, apart from conventionality, have ever been successfully vindicated. To explain the present treatment of phrasal verbs as phraseological units, it is worth noting here that the lexemic approach to phraseology brings back Makkai’s original lexemic idiom notion in his stratificational view of language (a lexeme being a basic meaningful element of the tactics on the lexemic stratum of the language; lexemes can be composed of lexons – materially, minimal free forms or words – Makkai, 1972: 90) and the abstract idea of ‘phraseme’ (Chlebda, 1991) – a communicative phraseological unit, a pragmatic ‘lemma’, dominating all potential reproducible variants used in similar contexts or situations to communicate or negotiate a meaning.

9612Arabski.indb 260

13/01/2006 08:51:58

Phrasal Verbal Idioms

261

Lexicographic Description The phrasal verb idiomatic unit – neither a compound word (such as prefixed verbs or compounds with literal derivational history), nor a syntactically described ‘verb phrase’ – can be presented in a hypothetical dictionary entry form which is based on the sources described below and on the recorded, published translation equivalents. In such an entry or lexicographic description, the criteria for both headphrase variance (including nominalisations such as pick-up, takeover) and ‘sense families’ (groups of senses under the same ‘lexeme’) are typically semantic. For example: COME THROUGH (WITH sth) 1. (cf. PULL THROUGH) ‘live, survive in spite of difficulties’; 2. (NOT FORMAL) ‘do or produce what is needed; help provide money or information’; 3. ‘appear; arrive, be delivered; show clearly, be communicated’. They had come through untouched. Przeszli przez to cali i zdrowi. (Closing Time – The Sequel to Catch 22, p. 271, sense 1.); ‘I came through when it counted, didn’t I ?’ – Nie zawiodÙem w odpowiedniej chwili, prawda? (God Knows, p. 52, sense 2.); Your examination results will come through by post. (Longman Dictionary of Ph.Vbs, p. 96). Francine replied in blubbering gibberish that she had never wanted the franchise for herself… Some of the words came through… Francine szlochaj­c beÙkotaÙa, Če nigdy nie myïlaÙa o tej filii dla siebie… Niektóre jej sÙowa docieraÙy do Hoovera. (Breakfast of Champions, p. 160, sense 3.).

Syntactic Restrictions Apart from, or within, such entries, one has to review the following semanto-syntactic issues typical of phrasal verb structure. Whereas in idioms (per se) the notion of fixedness is relative, phrasal verbs are syntactically or transformationally restricted as regards, for example: the obligatory (or usual) conditional mood; negation or question transformations (including dependent clauses), as in the case of put up with something or someone; obligatory passive transforms – in specific senses (e.g. be headed for something), or, contrarily, the prevention of the passive transformation; conventional compounding (hand-me-downs); fixed institutionalised derivatives (a rip-off); restrictions on the adverbial particle position (take in ‘receive’); the restriction on the position of the adverbial particle after the verb in action nominalisations, in the case of idiomatic phrasal verbs, where the particle cannot be used as a grammatical constituent (adverbial) in a gapped or ellipted sentence (compare the ungrammatical: *his throwing of the dinner immediately up). Moreover, restrictions on selection or collocability patterns include animate or inanimate subjects or objects taken (e.g. put away an animal), deictic features (stay put), the obligatory it or one in nominal slots (ask for it, put one over on someone).

9612Arabski.indb 261

13/01/2006 08:51:59

262

Cross-linguistic Influences in the Second Language Lexicon

The Significance of Phrasal Verbs On the basis of the material and frequency data acquired (a sample corpus of 10 novels written by American authors, newspaper and magazine articles, numerous dictionary sources, etc.), the following three crucial aspects of the phrasal verb have been taken into consideration: (1) The observed frequency of occurrence, as the present author’s data in the unpublished PhD thesis show, is a significant fact (99 identified idiomatic headphrases entering separate ‘sense families’ or sense groups of lexical units of at least 20 occurrences each, including transferred or derived forms; the three most common phrasal verb roots are: GO, COME and BE, with 599, 548, and 480 idiomatic occurrences, respectively). The reliability of corpus data in terms of ‘secondary genre’ composition (Bachtin, 1986) and the specific features of fictional prose phraseological material (e.g. ‘spokenness’ – ‘mówionoï°’, mentioned by Wawrzyniak, 1991), as well as the high rates of lexical choice of phrasal verbs in comparison to non-idiomatic synonyms – all seem to corroborate Denison’s (1984) statement on the centrality of the phrasal verb in the English lexical and grammatical system. (2) In spite of relative morphotactic instability or ‘dispersal’ of the phrasal verb predicate in the English sentence patterns (Hopper, 1997), the lexical status of non-literal, institutionalised phrasal verb combinations is not diminished or weakened in normative terms, as research has indicated. Of course, such a lexicographic description as above has little theoretical linguistic background in the sense of abstract rules and mental lexicon representations dominating the headphrases. (3) The idiomaticity of phrasal verbs as lexical units is to be understood in terms of phraseological non-compositionality and limited productivity, instead of the automatisation of ‘word-built’ compounds.

A Semantic Classification In terms of particular senses or occurrences of non-literal phrasal verbs, it seems reasonable to propose four tentative textual categories of phrasal verb units (as in Olejniczak, 1998), naturally based on semantic criteria, degrees of metaphoricalness or opacity, as well as institutionalisation and mutual selection of component subsenses (compare: Weinreich, 1966/1969). It must be emphasised here that the phrasal verbs described under the same lexical entry are not necessarily to be categorised in the same way from the point of view of figuration or degree of opacity, fixedness, institutionalisation, or selection for analogous or similar contexts. What we are

9612Arabski.indb 262

13/01/2006 08:52:00

Phrasal Verbal Idioms

263

dealing with in the case of the following categorisation is not lexicographic description but the occurrences of particular meaningful, paraphraseable linguistic units in context. Compare the following: (1) transparent metaphors (peripheral to both lexicalised idioms and regular collocations), for example: fall into (a frenzy), cling to (a belief); (2) lexicalised combinations in which the particles are used in one of the particular lexical senses, such as: ‘continuity’ (walk ON, drink AWAY), ‘influence’ or ‘interaction’ (turn ON someone), ‘activity’, ‘operation’, or ‘currency’ (switch something ON, you’re ON), ‘exhaustive’ (wither AWAY), ‘completion’ (sober someone UP, calm DOWN); (3) highly lexicalised (lexically fixed) combinations with questionable transparency or hardly detectable figurative core: tune out, sink in, carry on ‘continue’, come from something, look up to someone, get away with it or something; and (4) totally ‘opaque’ lexicalised idioms without a clear metaphorical core, such as act up, put up with someone, do someone in, be done for.

Phrasal Verbs: A Quiz The unitary, lexical treatment of phrasals has determined the formulation of the following questionnaire, for which the text material was taken, with one exception, from the ‘sample corpus’ mentioned above (compare Gläser’s, 1998, ‘sample analysis’). The questionnaire is basically a vocabulary quiz, not a translation exercise or a test in the active use use of structure or collocation, or a prescription for statistical data. It does, however, seem to have tested a small portion of the advanced students’ near-native active vocabulary – that which has undergone, to a certain extent at least, the process of gradual automatisation. The solutions are open-ended (the list of ‘adverbial particles’, an example, and the Polish rubric are meant to ensure clarity of instruction and avoid terminological misunderstanding, rather than provide ‘multiple-choice’ options); the units have clearly identified contexts, with a direct formulation of the type of structure required, which the students (subjects) identify as phrasal verbs, hard as it might be for them to define the latter. In both versions (1* and 2*) of the quiz, each of the above categories (1) to (3) of non-literal, institutionalised and/or idiomatic phrasal verbs is represented by three sentences or passages; idioms assigned to Category (4) are shown in five sentences. The phrasal verbs selected for the questionnaire represent different ‘sense families’ or complex dictionary entries in which the numbers of occurrences of verbs in all subsenses are over 10.

9612Arabski.indb 263

13/01/2006 08:52:01

264

Cross-linguistic Influences in the Second Language Lexicon

The answers consist in filling out the two word-positions (gaps) with a lexical verb and one adverbial particle (also in the case of so-called phrasalprepositional verbs), which constitute the translation equivalent of the Polish verb (verb phrase) printed in brackets at the end of each sentence. In a few cases, some idiomatic synonyms were accepted which were not used in the original. None of the Polish paraphrases representing Categories (2) to (4). (sentences 4 – 14) have English equivalents correctly transferrable in these particular contexts. The paraphrases are all, with one exception, original target language (TL) translation equivalents in the published Polish translation of each source novel. Below is the target form of the first paper (version 1*) of the quiz, where the correct items, with possible lexical synonyms in brackets, are given in bold type (the phrasal verbs used in version 2* are shown in the table below; in the original, the sources and authors’ names were omitted in order for the students to concentrate better on the text): PHRASAL VERBS – QUIZ (version 1*) W kaČdym z podanych niČej zdaÚ wpisz w puste miejsca czasownik w odpowiedniej formie oraz partykuÙ¿ b­dĊ przyimek, tak aby uzupeÙni° kontekst i sens zdania angielskiego. Polskie tÙumaczenie kaČdego ze zwrotów podane jest w nawiasie. W kaČdym zdaniu naleČy wpisa° dwa wyrazy. PartykuÙ¿ lub przyimek moČesz wybra° spoïród nast¿puj­cych form: about, across, after, along, apart, around, aside, at, away, back, behind, by, down, forth, forward, in, into, off, on, out, over, past, round, through, to, under, up, upon, with, without. PRZYKRAD: He’s nice to … put … …up… with us, though it costs him his reputation. (tolerowa°) [J. Irving, The World According to Garp] (1) Rabbit hates to put any more money into Japanese pockets but he knows from Consumer Reports that in this particular line they can’t be touched for quality. (wypycha°) [J. Updike, Rabbit is Rich] (2) Pru waves her husband’s quavery words and all their worried faces away. ‘I’m fine. I love all of you so much.’ Her hair streams outward as she waits to sink into (fall into) sleep. (pogr­Čy° si¿ w) [J. Updike, Rabbit is Rich] (3) Up till then it was all these strange little Arabs wanting to wipe out (wipe away) little Israel. (zetrze° z powierzchni ziemi) [P. Roth, The Counterlife] (4) I adored her, and love brought out my deepest peculiarities. (wydobywaÙa na jaw) [S. Bellow, Humboldt’s Gift] (5) It was a very restless country, with people tearing around all the time. Every so often, somebody would stop to put up a monument. (wznieï°) [K. Vonnegut Jr., Breakfast of Champions]

9612Arabski.indb 264

13/01/2006 08:52:01

Phrasal Verbal Idioms

265

(6) Forty years this went on, with God wrathful and fulminating and the people recalcitrant, stiff-necked and disobedient. (ci­gn¿Ùo si¿) [J. Heller, God Knows] (7) He wondered what they hoped to find out about him that he would not be willing to tell them outright. (dowiedzie° si¿) [J. Heller, Closing Time. The Sequel to ‘Catch 22’] (8) If people like Brenda Parsley can take over these men’s jobs I don’t see why I can’t. (przej­°) [J. Updike, The Witches of Eastwick] (9) She had dropped out of college almost as soon as she’d begun. (porzuciÙa) [J. Irving, The World According to Garp] (10) It was the hate coming out of her mouth that did her in, not a few harmless feathers and pins. (zabiÙa) [J. Updike, The Witches of Eastwick] (11) Don’t put on with me. Don’t be superior. (udawaj) [S. Bellow, Humboldt’s Gift] (12) I used to weep and wail about people shooting bald eagles with automatic shotguns from helicopters and all that, but I gave it up. (daÙem spokój) [K. Vonnegut, Breakfast…] (13) ‘Who says that ? I’m not sure of anything anymore’.‘Come on (get away, get off), Yo-Yo. On our floor, they’re saying you’ve even got a good thing going with one of the nurses’. (daj spokój) [J. Heller, Closing Time …] (14) She walks out with her dog before she turns in, but she didn’t show up yesterday (kÙadzie si¿ spa°) [U.S. movie].

Results The test was administered in April and May 2003, to four groups (three regular, one ‘evening group’) of the Sosnowiec Teacher Training College students in the second year, at the advanced level of EFL education (in the ‘Practical English’ course). The results were submitted in 57 copies. As a control group, the quiz was also written by 10 students in the fifth year of advanced EFL education at the English Language Faculty of the University of Silesia. Two versions of the quiz were used in the second year (for different groups) to reduce the possibility of information exchange between students in the same year. Some feedback on the results was provided for students interested at a special session on phrasal verbs and idioms at the end of the academic year. The results for both quiz versions, with examples of different response types, are presented in Table 19.1 (‘correct’ responses there include original translation equivalents, as well as acceptable, contextually appropriate phrasal verb synonyms).

9612Arabski.indb 265

13/01/2006 08:52:03

266

Cross-linguistic Influences in the Second Language Lexicon

Table 19.1 Data counted for two versions of the phrasal verbs questionnaire in the 2nd year of the Teacher Training College and the 5th year group Total responses 798

Version 1* (27 students)

Version 2* (20 students)

Version 1* (5th year Group – 10 students)

9612Arabski.indb 266

Correct idiomatic responses (35.8% of total) Examples (all target items are shown)

378

Data

125 (33.1%) 1. Put (money) into, 2. Sink into (fall into), 3. Wipe out (wipe away);

25 (20% of correct responses)

4. Bring out, 5. Put up (a monument), 6. Go on:

26 (20.8%)

7. Find out, 8. Take over, 9. Drop out;

45 (36%)

10. Do someone in, 11. Put on (with someone), 12. Give up, 13. Come on, 14. Turn in

29 (23.2%)

280

88 (31.4%) 1. Put (faith) in something, 2. Sweep someone away (wipe away, wipe out), 3. Come back;

28 (31.8% of correct responses)

4. Dress up (as__), 5. Go into (details), 6. Put (money) away (put aside, set aside);

32 (36.4%)

7. Drop out (of school), 8. Find out, 9. Put someone through something;

8 (9.1%)

10. Do someone in, 11. Work out, 12. Come on, 13. Give up something (for something), 14. Turn in

20 (22.7%)

140

73 (52.1%)

1. Put (money) into, 2. Sink into (fall into), 3. Wipe out (wipe away);

13 (17.8% of correct responses)

4. Bring out, 5. Put up (a monument), 6. Go on:

18 (24.7%)

7. Find out, 8. Take over, 9. Drop out;

23 (31.5%)

10. Do someone in, 11. Put on (with someone), 12. Give up, 13. Come on, 14. Turn in

19 (26%)

13/01/2006 08:52:03

267

Phrasal Verbal Idioms Table 19.1 cont. Incorrect responses (numbers 1–14 refer to respective target items/contexts) Literalisations (data and examples)

Metaphors (data and examples)

Phrasal verbs with questionable acceptability in context (data and examples)

Incorrect, nonexistent phrasal verbs, zero responses, other (data and examples)

29 (7.7%), of which 19 in sentences 1–6

23 (6.1%), of which 14 in sentences 10–14

17 (4.5%)

184 (48.7%)

1. Put in 5. Raise up, 6. Drag along

10. Put down, 13. Calm down

10. Knock down, 11. Act out

1. Fill with 5. Take up 7. Get into 13. Give up

27 (9.6%), of which 19 in sentences 1–6

13 (4.7%), of which 7 in sentences 10–14.

15 (5.4%)

140 (50%)

1. Put on, 3.Sweep off, blow away

10. Put down, 12. Come off, 13. Leave behind

8. Make out 10. Wear out

3. Return over 6. Put (money) out 7. Give away 11. Say out

5 (3.6%), of which 4 in sentences 1–3

7 (5%), of which 5 in sentences 10–14

5 (3.6%)

51 (36.4%)

3. Wash away 14. Lie down

3. Blow away 10. Put down

3. Tear down, 4. Bring forth 11. Play around

2. Plunder into 4. Take out 9. Get away 14. Lie in

9612Arabski.indb 267

13/01/2006 08:52:05

268

Cross-linguistic Influences in the Second Language Lexicon

The results recorded were as unsatisfactory in terms of ‘correct’ answers as expected (errors in tense forms were not taken into consideration); the highest individual score in the second year was nine correct items – 64% – and the lowest – one item (in the fifth year, half of the group scored above 50%). On the other hand, it was not a second year student’s ‘command’ which they ‘should’ have of English lexis that was tested. A good number of successful responses have been recorded on ‘common’ phrasals such as come back, find out, take over. The results for sentences 1–6, at least in the case of version 2* (nearly 70% of all correct answers) seem to corroborate, through receptive criteria, the above tentative categorisation of phrasal verb lexical units into metaphors, decomposeable idioms, and idiomatic lexemes, with the demarcation line between more and less analytic multiword units. Moreover, in view of no control over phraseology, students tend to give more metaphorical (thus, SL–TL tranferrable) responses in sentences 10–14 (category 4), as they are aware of the high idiomatic status of those phrasal verbs. However, the most advanced (fifth years) achieved the best results in the more idiomatic categories (over 57% – see Table 19.1).

The Notion of Interlanguage Regardless of the evaluation of the TL lexemic idiom performance, such as above, it is never possible for such evidence to provide an insight into the mental system of the interlanguage of the advanced student of English. One cannot deny that ‘interlanguage’ does exist as a ‘product’ and that certain regularities do emerge in the erroneous – or lack of – performance of foreign language learners in the same nationality (L1): learners ‘do’ something (or a lot) by way of strategically building up, or trying out, their TL competence (Hamilton, 2001), including testing their own errors (Lee, 1986). On the other hand, the generativist-cognitivist interlanguage hypothesis itself (Selinker, 1972; Ellis, 1998), as part of psychological mechanisms, merely re-describes errors and maps them unto abstract rules and representations, without explaining them. Hamilton (2001: 75) calls it an internalisation of the ‘language-like behaviour’ or the ‘Gypsy violinist’ syndrome, combined with the unjustified systemic view of learner ‘language’ and abstract ‘reconstruction’ of the speaker’s meaning outside of the situational context or social exchange. Such interesting studies as Dilin Liu’s (1995) weaken the traditional assumption of interlanguage as a psycholinguistic or linguistic aspect of learner (participant?) behaviour, and point to social values (‘deep structure’ mentality) or pragmatic factors (discourse, communicative interaction) as part of negative transfer.

9612Arabski.indb 268

13/01/2006 08:52:06

Phrasal Verbal Idioms

269

However abstract our view of learners’ competence or ‘mental systems’ may be, it is beyond doubt that the die-hard logical assumptions about negative NL–TL transfer remain unshaken, e.g. the fact that the two languages need to be genetically close enough for the linguistic transfer to occur (compare: English and Chinese). Moreover, there has to be an analogue of a TL structure in the mother tongue (Arabski, 1997). Systematic studies of the acquisition of such grammatical systems as determiners (Simons, 2001) neatly fit into the interlanguage theory, as long as they trace the ‘sequence’ of particular rules being learned as the ‘order’ of specific parts of the system or types of structures.

The (In)significance of the Error Gap In the present case, however, we can hardly assume that there is a ‘system’ of idiomatic phrasal verbs or a ‘sequence of rules’ for its acquisition. The crucial surface phenomenon observed is the error ‘gap’ or lack (irrelevance) of performance (as in Simons’ s, 2001, worst performance of Korean children – compared to Spanish – on the English article, to which there is no equivalent structure in Korean); the only cases of transfer, occasionally positive, occur at the level of literal translation of prefixed forms or metaphorical sequences. The simple and mundane, but most reasonable explanation is a lack of ‘knowledge’ – a gap in competence, which would say nothing by itself about any psycholinguistic systems; a gap which would not help reconstruct a mystified learner’s meaning; a gap which can only be looked at in terms of a situated practical achievement or failure. It is a failure in terms of a norm; Hamilton (2001: 83–4) offers a normative explanation, instead of a mentalistic one: ‘To make an error is not to know what a proficient speaker of a language would say or write to fulfil a particular purpose in a given situation. The criteria for deciding what counts as an error can only be based upon what speakers of a language do when they comunicate successfully.’ The role of the ‘brain’ is obvious here, but it is not ‘explanatorily relevant’. To be true to laboratory fact, we might also say that a highly marked phrasal verb which has a peculiar form and no analogue in NL (L1), with, in many cases, high degrees of idiomaticity or opacity (‘disinformation’), blocks and neutralises NL–TL transfer, which is inversely proportional to markedness (cf. Arabski, 1997; Ellis, 1998). However, in the present account of a research sample, the gap of ‘not knowing’ is related to the ‘active vocabulary’ mentioned and tested above. Kelly’s (1991: 136) data on active vocabulary in an advanced EFL student show c. 2000 up to 5000 words which s/he is ‘capable of incorporating into

9612Arabski.indb 269

13/01/2006 08:52:07

270

Cross-linguistic Influences in the Second Language Lexicon

his own speech’ (including selection and non-figurative collocation), in contrast to 20,000 in an ‘average native speaker’. Kelly admits that idioms are not included as a ‘particular and peripheral lexical zone’, which only strengthens our present issue.

Concluding Remarks If we were to conclude that the above ‘error gap’ in the case of advanced students was a breach of a norm, it would not be of any particular normative statement, but of a norm of frequency, a norm of casual use, or preferred lexical choice, considering the status, frequency and centrality of the phrasal verb in the English lexical and grammatical system. All of this, as well as the structural complexity (dispersal, collocability) of phrasal verbs, does not necessarily correspond to the opportunities of corrective feedback or input that a foreign student may need. That is confirmed by the better results in the fifth year: the area under consideration is understood to be synonymous with very proficient levels of language learning. From a psycholinguistic point of view, however, we might inverse the statement made by Marton (1986: 1361) that ‘the use of communication strategies is always a symptom of linguistic deficiency’. Deprived of the communicative option of paraphrasing or expressing the sense by means of singlelexon verb equivalents, the students reveal a deficiency of their productive (not necessarily to mean ‘creative’) repertoire of knowing what a proficient speaker of the language would say or write to fulfil a particular purpose in this situation. By no means is that intended as a critical statement on the students’ competence, knowledge of vocabulary or communicative capabilities in general, as the questions asked do not constitute a grammar or vocabulary exercise or a set of test items. However, a student with a low or medium level of aspiration, deprived of a large portion of meaningful input feedback (especially naturalistic or quasi-naturalistic), tends to rely on copying and simplification strategies, which do not necessarily forward him/her to the level of proficiency s/he is expected to achieve at particular stages of language education in an artificial environment. References Arabski, J. (1997) Przyswajanie j¿zyka obcego i pami¿° werbalna. Katowice: Wydawnictwo gl­sk. Bachtin, M. (1986) Estetyka twórczoïci sÙownej (translated from Russian by D. Ulicka). Warszawa: PIW. Bolinger, D. (1971) The Phrasal Verb in English. Cambridge, MA: Harvard University Press. Chlebda, W. (1991) Elementy frazematyki. Opole: WSP.

9612Arabski.indb 270

13/01/2006 08:52:07

Phrasal Verbal Idioms

271

Clausner, T.C. and Croft, W. (1997) Productivity and schematicity in metaphors. Cognitive Sciences 21 (3), 247–82. Corder, S.P. (1967) The significance of learners’ errors. International Review of Applied Linguistics 5, 161–70. Denison, D. (1984) On ‘get it over with’. Neuphilologus 68, 271–7. Ellis, R. (1998) Second Language Acquisition. Oxford: OUP. Fernando, C. and Flavell, R. (1981) On Idiom: Critical Views and Perspectives (Exeter Linguistic Studies 5). Exeter: University of Exeter. Fraser, B. (1976) The Verb-Particle Combination in English. London, New York: Academic Press. Gläser, R. (1998) A plea for phraseo-stylistics. In P. Kakietek (ed.) Topics in Phraseology Vol. 1 (pp. 22–32). Katowice: Wydawnictwo Ug. Hamilton, R.P. (2001) The significance of learners’ errors: A philosophical investigation of the interlanguage hypothesis. Language and Communication 21, 73–88. Hockett, C.F. (1958) A Course in Modern Linguistics. New York : The Macmillan Co. Hopper, P.J. (1997) Discourse and the category ‘verb’ in English. Language and Communication 17 (2), 93–102. Kelly, P. (1991) Lexical ignorance: The main obstacle to listening comprehension with advanced foreign language learners. IRAL 19 (2), 135–49. Lee, W.R. (1986) The correction and non-correction of errors. In D. Kastovsky and A. Szwedek (eds) Linguistics Across Historical and Geographical Boundaries. Vol. 2: Descriptive, Contrastive and Applied Linguistics (pp. 1321–32). Berlin, New York, Amsterdam: Mouton de Gruyter. Liu, Dilin (1995) Sociocultural transfer and its effect on second language speakers’ communication. International Journal of Intercultural Relations 19 (2), 253–65. Makkai, A. (1972) Idiom Structure in English. The Hague: Mouton. Makkai, A. (1995) A Dictionary of American Idioms (Introduction, pp. VI–X). Hauppaugue, NY: Barron’s. Marton, W. (1986) Language learners’ errors in a pedagogical perspective. In D. Kastovsky and A. Szwedek (eds) Linguistics Across Historical and Geographical Boundaries. Vol. 2: Descriptive, Contrastive and Applied Linguistics (pp. 1357–76). Berlin, New York, Amsterdam: Mouton de Gruyter. Moon, R.E. (1998) Fixed Expressions and Idioms in English. A Corpus-Based Approach. Oxford: Clarendon Press. Nunberg, G., Sag, I.A. and Wasow, T. (1994) Idioms. Language. Journal of the Linguistic Society of America 70 (3), 491–538. Olejniczak, P. (1998) Phrasal verbs as idioms. A case in point: Translation. In P. Kakietek (ed.) Topics in Phraseology Vol. 1 (pp. 52–90). Katowice: Wydawnictwo Ug. Scheffler, P. (1994) An English-Polish Dictionary of Phrasal Verbs with Exercises. WÙocÙawek: Altravox Press. Seidl, J. (1999) English Idioms: Exercises on Idioms. Oxford: OUP. Selinker, L. (1972) Interlanguage. International Review of Applied Linguistics 10 (2), 209–31. Simons, G.V. (2001) The acquisition of the English determiner system: Sequence, order and transfer (pp. 1–165). EdD thesis, Temple University: UMI ProQuest Digital Dissertations. Sroka, K. (1972) The Syntax of English Phrasal Verbs. The Hague, Paris: Mouton.

9612Arabski.indb 271

13/01/2006 08:52:08

272

Cross-linguistic Influences in the Second Language Lexicon

Strässler, J. (1982) Idioms in English. A Pragmatic Analysis. Tubingen: Narr Verlag. Wawrzyniak, Z. (1991) Praktyczne apekty translacji literackiej. Na przykÙadzie tekstów niemieckiego i angielskiego. Warszawa: PWN. Weinreich, U. (1966/1969) Problems in the analysis of idioms. In J. Puhvel (ed.) (1969) Substance and Structure of Language (pp. 23-81). Berkeley, Los Angeles: University of California Press.

9612Arabski.indb 272

13/01/2006 08:52:09