268 46 2MB
English Pages 304 [303] Year 2022
Advanced Students’ Knowledge of Vocabulary in a First and Second Language
SECOND LANGUAGE ACQUISITION Series Editors: Professor David Singleton, University of Pannonia, Hungary and Fellow Emeritus, Trinity College, Dublin, Ireland and Professor Simone E. Pfenninger, University of Zurich, Switzerland This series brings 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 is thus interpreted in its broadest possible sense. The volumes included in the series all offer in their different ways, 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 is privileged in the series; nor is any relevant perspective – sociolinguistic, psycholinguistic, neurolinguistic, etc. – deemed out of place. The intended readership of the series includes final-year undergraduates working on second language acquisition projects, postgraduate students involved in second language acquisition research, and researchers, teachers and policymakers in general whose interests include a second language acquisition component. All books in this series are externally peer-reviewed. Full details of all the books in this series and of all our other publications can be found on http://www.multilingual-matters.com, or by writing to Multilingual Matters, St Nicholas House, 31–34 High Street, Bristol, BS1 2AW, UK.
SECOND LANGUAGE ACQUISITION: 158
Advanced Students’ Knowledge of Vocabulary in a First and Second Language Monica Karlsson
MULTILINGUAL MATTERS Bristol • Jackson
DOI https://doi.org/10.21832/KARLSS5249 Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress. Names: Karlsson, Monica, author. Title: Advanced Students' Knowledge of Vocabulary in a First and Second Language/Monica Karlsson. Description: Bristol; Jackson: Multilingual Matters, [2022] | Series: Second Language Acquisition: 158 | Includes bibliographical references and index. | Summary: "This book offers a comprehensive investigation into advanced students' knowledge of vocabulary in their L1 and L2. This cross-sectional study examines the qualitative and quantitative aspects of students' vocabulary knowledge in subfields such as suffixation, idioms and proverbs, multi-word verbs, polysemous words and inferencing skills"-- Provided by publisher. Identifiers: LCCN 2022027091 (print) | LCCN 2022027092 (ebook) | ISBN 9781800415249 (hardback) | ISBN 9781800415263 (epub) | ISBN 9781800415256 (pdf) Subjects: LCSH: Vocabulary--Study and teaching (Secondary) | Language and languages--Study and teaching (Secondary) | Second language acquisition. Classification: LCC P53.9 .K37 2022 (print) | LCC P53.9 (ebook) | DDC 418.0071/2--dc23/eng/20220822 LC record available at https://lccn.loc.gov/2022027091 LC ebook record available at https://lccn.loc.gov/2022027092 British Library Cataloguing in Publication Data A catalogue entry for this book is available from the British Library. ISBN-13: 978-1-80041-524-9 (hbk) Multilingual Matters UK: St Nicholas House, 31–34 High Street, Bristol, BS1 2AW, UK. USA: Ingram, Jackson, TN, USA. Website: www.multilingual-matters.com Twitter: Multi_Ling_Mat Facebook: https://www.facebook.com/multilingualmatters Blog: www.channelviewpublications.wordpress.com Copyright © 2023 Monica Karlsson. 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. The policy of Multilingual Matters/Channel View Publications is to use papers that are natural, renewable and recyclable products, made from wood grown in sustainable forests. In the manufacturing process of our books, and to further support our policy, preference is given to printers that have FSC and PEFC Chain of Custody certification. The FSC and/or PEFC logos will appear on those books where full certification has been granted to the printer concerned. Typeset by Nova Techset Private Limited, Bengaluru and Chennai, India. Printed and bound in the UK by the CPI Books Group Ltd.
Contents
Acknowledgements
ix
1 Introduction 1 1.1 Teaching and Learning Vocabulary in a Second Language 1 1.2 The Present Study 2 The purpose and outline of the book 2 Informants, material and method 4 1.3 What Does It Mean to Know a Word? 8 1.4 The Chapter That Follows 12 Notes 12 2 Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary and Advanced Vocabulary 13 2.1 Introduction 13 2.2 Vocabulary Taught at Upper Secondary School Level 15 Receptive versus productive vocabulary knowledge 15 The parallel tests used for the present chapter 16 Results and discussion 19 2.3 Specialised Uses of Vocabulary 29 A continuum of specialised uses of vocabulary 29 The parallel tests used for the present chapter 30 Results and discussion 32 2.4 Advanced Vocabulary 41 What makes a word advanced? 41 The parallel tests used for the present chapter 44 Results and discussion 45 2.5 The Students’ Vocabulary Size on the Three Parallel Tests Used for the Present Chapter 53 2.6 Conclusions and Pedagogical Implications 54 2.7 The Chapter That Follows 55 Notes 55 3 Word Formation: With a Focus on Derivational Suffixation 56 3.1 Introduction 56 3.2 Theoretical Background and Previous Research 58 Do learners decompose morphologically complex words or not? 58 v
vi Advanced Students’ Knowledge of Vocabulary in a First and Second Language
A pedagogical approach to affixes 64 3.3 A Comparison of Affixation Rules between English and Swedish and the Parallel Tests Used for the Present Chapter 69 A comparison of affixation rules between English and Swedish 70 The parallel tests used for the present chapter 73 3.4 Results and Discussion 77 The students’ results 77 Results on Part A (word pairs, context-based and gap-filling) 79 Results on Part B (word families, decontextualised) 90 Individual L1 and L2 affixation knowledge and the correlation between affixation knowledge and vocabulary size 95 3.5 Conclusions and Pedagogical Implications 96 3.6 The Chapter That Follows 97 Notes 97 4 Idioms and Proverbs 99 4.1 Introduction 99 4.2 Theoretical Background and Previous Research 100 Idioms and proverbs defined 100 The processing and comprehension of L1 and L2 idioms 102 Idiom comprehension: The effects of contextual support, transparency, frequency and (cultural) sameness 109 Pedagogical approaches to idioms 112 4.3 The Parallel Tests Used for the Present Chapter 116 4.4 Results and Discussion 121 The students’ results 121 Results on the individual idioms and proverbs 123 Individual L1 and L2 knowledge 131 4.5 Conclusions and Pedagogical Implications 133 4.6 The Chapter That Follows 134 Notes 134 5 Idiomatically Used Prepositions and Multi-Word Verbs 136 5.1 Introduction 136 5.2 Theoretical Background and Previous Research 139 Idiomatically used prepositions and multi-word verbs defined 139 L2 learners’ mastery of idiomatically used prepositions and multi-word verbs 143 Pedagogical approaches to idiomatically used prepositions and multi-word verbs 149 5.3 The Parallel Tests Used for the Present Chapter 155
Contents vii
5.4 Results and Discussion 159 The students’ results 159 Results on the individual idiomatically used prepositions and multi-word verbs 162 Individual L1 and L2 knowledge 174 5.5 Conclusions and Pedagogical Implications 174 5.6 The Chapter That Follows 175 Notes 176 6 Polysemous Words, Lexical Fields of Near Synonyms and False Friends 178 6.1 Introduction 178 6.2 Polysemous Words 181 Introduction 181 Theoretical background and previous research 183 Part A of the parallel tests used for the present chapter: Polysemous words 191 Results and discussion 194 6.3 Lexical Fields of Near Synonyms 203 Introduction 203 Theoretical background and previous research 204 Part B of the parallel tests used for the present chapter: Lexical fields of near synonyms 208 Results and discussion 210 6.4 False Friends 215 Introduction 215 Theoretical background and previous research 216 Part C of the parallel tests used for the present chapter: False friends 218 Results and discussion 220 6.5 Individual L1 and L2 Knowledge 223 6.6 Conclusions and Pedagogical Implications 224 6.7 The Chapter That Follows 225 Notes 225 7 Inferencing 7.1 Introduction 7.2 Theoretical Background and Previous Research L2 Inferencing strategies L1 and L2 inferencing compared 7.3 The Parallel Tests Used for the Present Chapter 7.4 Results and Discussion 7.5 Inferencing in Relation to Vocabulary Size as Achieved on Some of the Previous Parallel Tests Used in the Present Study 7.6 Conclusions and Pedagogical Implications
227 227 228 228 230 233 238 242 246
viii Advanced Students’ Knowledge of Vocabulary in a First and Second Language
8 Summing Up and Directions for Future Research
247
252 255 257 266
Appendix 1: Tables Discussed in Chapter 3 Appendix 2: Tables Discussed in Chapter 5 Appendix 3: Tables Discussed in Chapter 6 Appendix 4: Tables Discussed in Chapter 7
References 272 Dictionaries 286 Newspaper and Magazine Articles 287 Websites 287
Author Index Subject Index
288 293
Acknowledgements
This book consists of eight chapters, five of which have previously been partly published in different international journals. Chapter 3 partly appears in: Karlsson, M. (2015) Advanced students’ L1 (Swedish) and L2 (English) knowledge of suffixation. International Journal of English Studies 15 (1), 23–49. Chapter 4 partly appears in: Karlsson, M. (2013) Quantitative and qualitative aspects of L1 (Swedish) and L2 (English) idiom comprehension. Studies in Second Language Learning and Teaching 3 (2), 293–319. Chapter 5 partly appears in: Karlsson, M. (2014) L1 (Swedish) versus L2 (English) mastery of free combinations of noun/verb + preposition as compared to multi-word verbs. International Journal of Language Studies 8 (3), 27–54. Chapter 6 – the part on polysemous words – appears in: Karlsson, M. (2013) Quantitative and qualitative aspects of advanced learners' L1 and L2 mastery of polysemous words. Hermes 51, 79–112. Chapter 6 – the part on lexical fields of near synonyms – appears in: Karlsson, M. (2014) Advanced students’ L1 and L2 mastery of lexical fields of near synonyms. World Journal of English Language 4 (3), 1–17. Chapter 7 partly appears in: Karlsson, M. (2014) Advanced learners’ L1 (Swedish) versus L2 (English) inferencing. Australian Review of Applied Linguistics 37 (1), 3–23. These articles have been edited for the volume to enhance their consistency with the rest of the book. I would like to thank the editors of the journals for allowing me to make my work available to an even larger international audience.
ix
x Advanced Students’ Knowledge of Vocabulary in a First and Second Language
I would also like to express my deepest gratitude to the students who so willingly contributed to the present study. With your help, the world of linguistics will have gained a much better understanding of advanced students’ mastery of vocabulary in a first and second language.
1 Introduction
1.1 Teaching and Learning Vocabulary in a Second Language
The introduction to an article giving a brief overview of the fields of vocabulary teaching and learning states that traditionally, research in language teaching and learning methods devoted less attention to vocabulary than to other aspects of language as a communication system. Other language elements took precedence, and vocabulary tended to be presented in a way that favoured the introduction of grammatical elements. There was a general consensus that grammar should be taught and that in due time learners would ‘acquire’ the vocabulary necessary to deal with specific communication situations through their exposure to the target language. (Chacón-Beltrán et al., 2010: 1; see also O’Dell, 1997)
Along the same lines, Boyd Zimmerman writes that ‘the teaching and learning of vocabulary has been undervalued in the field of second language acquisition’ and that ‘SLA researchers and teachers have typically prioritised syntax and phonology’ (1997: 5; see also Richards, 1976). In fact, around 1980, there was a consensus among scholars that vocabulary was indeed a neglected field of investigation (Haastrup, 1991: 44). In addition, in those days many researchers thought of language ‘in terms of morphosyntactic rules generating acceptable sentence structures into which lexical items’ were ‘then slotted’ (Larsen-Freeman, 2012); that is, ‘[l]anguage was generally conceived as a grammar-lexis dichotomy with grammar “rules” on the one hand and “lists” of individual words on the other’ (Boers et al., 2004: 54). Giving precedence to grammar in this way, teaching and learning vocabulary as if it were a mere slot-filling system, very often produces L2 errors (Martin, 1984). For example, while both concede and yield may take nouns as objects (i.e. concede/yield something to someone), only the former can be constructed with a that-clause as in, for example, She conceded that we were right. Thinking that concede and yield can occupy the same slots at all times may thus induce a learner to produce the incorrect sentence *She yielded that we were right (1984: 132). However, [s]ince the mid-1980s there has been a renewed interest in the role of vocabulary in second language learning, and this has seen a growing body of empirically based studies of such issues as the nature of the 1
2 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
bilingual lexicon, vocabulary acquisition, lexical storage, lexical retrieval, and use of vocabulary by second language learners. (Coady & Huckin, 1997: ix)
The main part of these studies appears to have focused on measuring quantitative knowledge rather than investigating achievements from a qualitative perspective. In 1998, Schmitt then writes that ‘there is a virtual explosion of vocabulary studies’ (1998: 282; see also Meara, 1995), a development that has continued ever since. Finally, in a doctoral dissertation completed in 2010, the author states that ‘the importance of vocabulary knowledge in mastering a second language is by now well established in second language research literature’ (Lemmouh, 2010: 1; see also Alderson, 2005 and Milton, 2009). The main reason for this dramatic increase in vocabulary studies is that researchers slowly came to realise, rather than ascribing vocabulary a secondary role as described earlier, that ‘utterances are formed from the interplay of the two [i.e. grammar and vocabulary] in a lexicogrammar’ (Larsen-Freeman, 2012; see also Celce-Murcia & Larsen-Freeman, 1999; Halliday, 1994; Malvern et al., 2008; Nattinger & DeCarrico, 1992; Pawley & Syder, 1983; Sinclair, 1991). There are many examples of this interplay. For instance, while the two verbs come and go may be used in V + Adj. constructions, as exemplified by come alive and go crazy, the verbs keep and leave can be found in V + N + Adj. resultative constructions such as keep someone happy and leave someone crippled. The verbs’ semantic profiles reveal, however, that it is only go and leave in the two sets that are usually used in negative contexts. That is why it is incorrect to say *keep unpleasant things behind (instead of leave unpleasant things behind) and *go to us (instead of come to us) (Liu, 2012). Put simply, these four verbs cannot be used indiscriminately in the grammatical frames discussed above; their semantic make-up must be taken into consideration. While the pendulum has now swung in the opposite direction completely for some scholars, who thus believe that vocabulary always takes priority over grammar, that is, that meaning always governs form, most researchers seem to think that the two systems are equally interdependent. Bell (2009: 126), for instance, states that a key question is whether and how far the developing lexicon can push structural development, or whether grammatical control pushes lexis. I suggest that the two systems are intertwined and that the answer depends on the structural/lexical item in question; it can work both ways. 1.2 The Present Study The purpose and outline of the book
Even though, as described in Section 1.1, there has been a considerable increase in the number of vocabulary studies during the last few decades,
Introduction 3
there are still comparatively few that simultaneously deal with more than one or two dimensions while focusing on advanced learners. In fact, since ‘any single measure of it [i.e. vocabulary] will give only a very minimal impression of the overall lexical knowledge constellation’ (Schmitt, 2010c), there is a need for more research projects that consider learners’ knowledge of vocabulary from a variety of perspectives concurrently. The present investigation, which is a cross-sectional study examining a group of learners at a single point in time, and whose overarching aim is to make a contribution to the understanding of advanced learners’ L1 and L2 mental lexicons so that necessary changes can be made to the L1 and L2 teaching syllabi, seeks to do just that by addressing the following research questions: In quantitative and qualitative terms, what are the similarities and differences between advanced students’ L1 (Swedish) and L2 (English) knowledge of
(a) vocabulary taught at upper secondary school level (Chapter 2), (b) specialised uses of vocabulary (Chapter 2), (c) advanced vocabulary (Chapter 2), (d) word formation – with a focus on derivational suffixation (Chapter 3), (e) idioms and proverbs (Chapter 4), (f) idiomatically used prepositions and multi-word verbs (Chapter 5), (g) polysemous words, lexical fields of near synonyms, and false friends (Chapter 6), and to what extent can students’ comprehension of vocabulary be attributed to L1/L2 inferencing (Chapter 7)?
The study begins by trying to gain insights into the size of the learners’ L1 and L2 lexicons, focusing on their comprehension of words from upper secondary school level, specialised uses of vocabulary, and advanced vocabulary. It continues by exploring the learners’ knowledge of derivatives, idioms, proverbs, idiomatically used prepositions, and multi-word verbs, with those of a more frequent and transparent nature providing further insights into the learners’ vocabulary size and those of a more infrequent and opaque nature providing insights into the depth of their vocabulary knowledge. The study then goes on to investigate the learners’ mastery of polysemous words, lexical fields of near synonyms, and false friends, which will primarily offer further insights into the informants’ knowledge depth, both in terms of individual word depth and relations between words. The final main chapter (Chapter 7), supplying a frame to the entire study, will explore to what extent the learners’ knowledge is based on their inferencing skills. To the present author’s knowledge, there is no other equally comprehensive investigation comparing advanced students’ knowledge of vocabulary in a first and second language.
4 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Informants, material and method
In each of the areas listed above, a group of 15 Swedish university students (age range: 19–49; mean: 23.47; SD: 7.81) was given two parallel tests, one in English and one in their mother tongue, each yielding 100 points. On each occasion the English test was given first, followed by the corresponding Swedish one. The subjects were able to sit with each test as long as they liked; no time constraints whatsoever were put on any of the tests in either language. All the material was collected during a three-week period, so the testing was comparatively intense. It needs to be pointed out here that, as the study only includes 15 informants, it has its limitations in terms of the generalisability of the results. On the other hand, the comparatively large size of the parallel tests administered, each comprising 100 test items, may partly compensate for the relatively small number of testees. The statistical analyses in each chapter will provide information about the results’ validity. In addition, one native speaker was subjected to the English tests. Since his results cannot be statistically tested in relation to the results of the informant group, he will primarily be used as a point of reference. Table 1.1 presents an overview of the informants. At the time of testing, all the Swedish testees were first-term university students taking English as a single-subject course. (Originally there were over 30 learners that were supposed to take the tests but, in the end, since some did not show up for all the tests and/or did not take all the tests at the appointed times, only 15 remained.) Of the 15 subjects, 12 are female (80%) and only three are male (20%).1 The overwhelming majority (10) of the students were, at the time of testing, in their late teens or early twenties; a few (four) were in their late twenties and one informant was middle-aged. (This 49-year-old student had not used his English actively for many years, and he brought this up on many occasions in connection with the tests, expressing time and time again that he did not think that his English was very good.) This appears to be a fairly normal distribution of age for a first-term group taking English as a single-subject course at the university where the present author works. Furthermore, four of the participants, all of whom were female, indicated that they had spent more than three consecutive months in an English-speaking country, something that may have influenced their knowledge of vocabulary in a positive way. Before the testing took place, the students were also asked to selfassess their current L1 and L2 vocabulary knowledge, to answer to what degree they felt that acquiring vocabulary is an important part of learning a language, and to indicate how they preferred to learn new words. 2 Additionally, the participants were asked if they had studied any other languages at university before and/or if they intended to study any other language(s) at tertiary level in the future.
Introduction 5
The native English speaker, a 33-year-old male, was, at the time of the testing, a second-term teacher-trainee student aiming to become an upper secondary school instructor, teaching English and a second subject (not chosen at the time) to learners in the Swedish school system. This student took several courses for the present author – proficiency-oriented (e.g. essay writing and grammar) as well as more theoretically oriented courses (e.g. didactics and second language acquisition) – and his performance was always average or below average. That this informant was a low achiever in almost all respects was also confirmed by his other teachers. He took the eight administered tests one year before the Swedish test group. What has been said above about the informants is summarised in Table 1.1. As a first step, not presupposing any single factor to be more important than another, the test items were – unlike in a more conventional approach, where commonality would have been implemented – selected randomly from a number of different sources, written as well as spoken, thus letting all factors relevant to the students’ knowledge in a specific area of vocabulary emerge on more equal terms. Even though, as evidenced in the research literature, frequency plays a role, it is not the only factor that determines what words learners remember from the input they receive, be it authentic or scripted text. Not even in texts put together for teaching purposes does word frequency appear to be taken into consideration to any great extent, content instead being the priority (Nordlund & Norberg, 2020). The aim, by initially selecting test items randomly, was hence to imitate the input received by many learners. There were also two other main reasons for such an unconventional approach. Firstly, a practical issue became apparent: in some cases, there were so many variables that matching the L1 and L2 tests/test parts in all respects became an impossible task. This was the situation with the second part of the suffixation test in Chapter 3, for instance. Here the total frequencies of the word families, the relative frequencies of the different family members, the number of members, and their word class affiliations, as well as the frequency, predictability, productivity, and regularity of the affixes themselves, would have to have been matched, which did not prove doable. Secondly, in Karlsson (2019), great care was taken to select idioms to be tested along three continuums, one of which was commonality. The present author therefore thought it would be interesting to explore whether a random approach to selecting test items in this area would yield similar results. As the reader will become aware in Chapter 4, this was indeed the case, which adds legitimacy to the approach taken in the present study. For all the other tests/test parts, versions and/or combinations of the explanations offered above together made the present author decide on a random approach as a first step.
Gender
F
F
M
F
F
F
F
Students
Student 1
Student 2
Student 3
Student 4
Student 5
Student 6
Student 7
20
21
25
19
19
20
19
Age
Yes
No
No
No
No
No
No
Good
Average
Average
Very good
Very good
Good
Average
Very important
Very important
Very important
Very important
Important
Very important
Important
R/L/S
R
R/W/L
R/L/S
R/L
R/L
R/W/S
How do you learn new Swedish words?
How is your knowledge of Swedish words?
Have you spent more than three consecutive months in an Englishspeaking country? How important is it to learn Swedish words?
English vocabulary knowledge
Experience of English
Good
Good
Poor
Very good
Good
Good
Good
How is your knowledge of Swedish words?
Important
Very important
Very important
Very important
Important
Important
Important
How important is it to learn Swedish words?
R/L
L
R/W
R/L
R
R/L/S
R/L/S
How do you learn new Swedish words?
Swedish vocabulary knowledge
No (Continued)
Swedish (in the past)
No
No
Maybe
Spanish (in the future)
Swedish (in the future)
Have you studied/ Are you studying/ Will you study any other languages at university level?
Studies in other languages
Table 1.1 The students included in the present investigation (F = Female, M = Male, R = Reading, W = Writing, L = Listening, S = Speaking)
6 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
F
F
F
F
M
F
F
M
M
Student 8
Student 9
Student 10
Student 11
Student 12
Student 13
Student 14
Student 15
Native speaker
33
20
19
27
49
20
19
27
28
No
No
Yes
No
No
No
Yes
Yes
Very good
Good
Good
Good
Poor
Good
Average
Good
Good
Very important
Important
Important
Important
Very important
Very important
Very important
Very important
Very important
R/L/S
R/W/L
R/W/L
S
R/L
R/L
Average
Good
Good
Good
Good
Good
Good
R/S L/S
Very good
R/W/L/S
Important
Important
Average
Important
Important
Important
Important
Very important
R/W/L
-
R
R
R/L
R/L
R/L
R/L/S
Possibly Swedish
Not sure
Maybe German
No
German (in the past)
Spanish (in the future)
No
No
French (in the future)
Introduction 7
8 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
What has been said above does not, however, mean that commonality was not considered at all. As a second step, the total frequencies of the L1 and L2 test items respectively were indeed ascertained. While the frequencies of the English test items are based on the British National Corpus (henceforth the BNC) (http://www.natcorp.ox.ac.uk), those of the Swedish test items are based on Språkbanken (http://spraakbanken.gu.se). The BNC consists of 100,000,000 running words, of which 90% is written language (60% books, 30% newspapers and periodicals, and 10% miscellaneous) and 10% is spoken language (conversation, chat shows, lectures, meetings, speeches, sports commentaries, and TV programmes). While the vast majority of the written language incorporates informative text, there are also sections that include imaginative writing (Nation, 2004: 11). The parts of Språkbanken that were available when the material was put together consist of 69,762,402 running words, all of which is written language taken from one daily newspaper. The effects of the different make-ups of these two corpora will, when relevant, be addressed in the different chapters. In addition to the BNC and Språkbanken, the present study also makes use of A General Service List of English Words (West, 1953), designed entirely for the teaching syllabus. Based on written material consisting of 5,000,000 words, this corpus lists 2000 head words, selected by considering a word’s frequency, range, ease/difficulty of learning, necessity, cover, and stylistic and emotional level (West, 1953: ix). Even though it was developed as early as the 1950s, it still has great validity (Nation & Waring, 1997: 13). Regrettably, no corresponding corpus was found for the Swedish language. 1.3 What Does It Mean to Know a Word?
In the research literature, there are two main ways of conceptualising vocabulary knowledge. On the one hand, learners’ mastery of the various components of which a word is made up may be investigated. According to Nation (2001), there are three such main components: form, meaning, and use. Form is concerned with knowledge of pronunciation, orthography, and the morphological structure of a word. The second component, meaning, involves knowledge of links between form and meaning, what the concepts refer to, and semantic relationships between words, such as hyponymy, polysemy, synonymy, antonymy and meronymy. Use, finally, deals with knowledge of a word’s grammatical pattern, the contexts in which it can be implemented, and its collocational profile and frequency. Furthermore, in the research body, a distinction is most often made between vocabulary mastered receptively and productively (Milton, 2009). Essentially, ‘[r]eceptive carries the idea that we receive language input from others through listening or reading and try to comprehend it, productive that we produce language forms by speaking and writing to convey messages to others’ (Nation, 2001: 24). This is also the approach
Introduction 9
taken in Nation’s conceptual framework discussed above, where the three main components are divided into subcomponents that deal specifically with these two types of knowledge. In accordance with a compositional approach, Schmitt (2010a) divides word knowledge into eight main components – spoken form, written form, meaning, grammar, collocations, register, frequency and associations – and discusses how the knowledge of a word may develop in relation to different stages of acquisition: early knowledge of a word, developing knowledge of a word, and advanced knowledge of a word. In the early stages of the developmental trajectory, Schmitt suggests that it is likely that the learner will know only some form of the word (spoken or written) and one of its meanings. Which of these two aspects is developed more quickly depends on the situation. The difficulty in learning the meaning of a word is substantiated by the fact that research has shown that learners often possess other kinds of knowledge even though they cannot demonstrate any semantic knowledge (Schmitt, 1998). However, as L2 learners usually approach a new language with ready-made concepts, it is normally the form rather than the meaning that creates more problems. This is especially true when the learner’s mother tongue and the L2 are closely related. In these initial stages, the learner also appears to have some very basic knowledge of a word’s grammatical framework, such as which word class it belongs to. After a few additional encounters, the knowledge of a word’s form (written and/or spoken) and the first sense learned will become even more consolidated in a learner’s lexicon, with form very often preceding meaning. However, since there are many words that have more than one sense, some of which may be unrelated to each other, the semantic profile of a word is still not fully mastered at this stage. Nor is its grammatical profile completely understood, since it is highly likely that a learner still does not know all of a word’s derivative forms (inflectional as well as derivational). What is interesting at this stage is that, even though there are limitations as to learners’ semantic and grammatical knowledge, these two components appear to develop at approximately the same pace (Schmitt, 1998). Also, other types of knowledge that are context-dependent – that is, collocational knowledge, knowledge of the word’s register, frequency and associations – now begin to develop. Depending on the type of exposure, learner motivation, and the distance between the L1 and the L2, research has shown that it takes between five and 16 encounters until a word is mastered so well that it can be considered to be known to the L2 learner (Nation, 1990). At this advanced stage, the word’s written and spoken form and most of its meanings will have been acquired. However, due to the intricateness of all the various components, it is highly unlikely that any learner will gain complete knowledge of any single word. Also, a great deal of research claims that it is the context-dependent components, especially collocation, that create
10 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
most problems for learners, lagging behind the form and meaning components (e.g. Bahns & Eldaw, 1993). The other way to conceptualise vocabulary knowledge is to see words as part of a lexical network, referred to as a learner’s mental lexicon. Within a connectionist framework, whether it is concerned with an L1 or an L2, a developing vocabulary network – that is, a network in which some items are acquired whereas others are lost – is described in terms of the strengthening and weakening of neural links or so-called network nodes. Moreover, Albrechtsen et al. (2008: 22) claim that [t]he many lexical entries in our L1 mental lexicon are not stored randomly, but are structured in a well-organized web with connections or pathways between the words; a structural system that enables us to retrieve words rapidly and with ease, because the access routes in the lexical store are varied and well-established.
They also wonder if the same [can] be said about foreign language learners’ mental lexicon at different stages of their interlanguage development. By nature, the L2 learner has had less exposure to the target language; so, one would naturally expect the L2 lexicon to differ from the L1 lexicon – not only in relation to size, but also in relation to the structural properties of the word store. But how different is the L2 lexicon? (2008: 22)
Trying to find out about differences between L1 and L2 vocabulary networks, researchers have engaged learners in so-called vocabulary association tests, of which there are two main types: productive association tests and receptive association tests. In the former case, subjects are most commonly offered prompt words and asked to say/write the first word that springs to mind (Namei, 2002, 2004; Singleton, 1999; Söderman, 1993; Wolter, 2001). Productive association tests can thus be said to deal primarily with the types of links seen in the mental lexicon. In receptive association tests, informants are most commonly given a stimulus word (e.g. edit) and asked to indicate which items in a set of words (e.g. arithmetic, film, pole, publishing, revise, risk, surface and text) they believe it is linked to (Read, 1993). Hence, in such cases, researchers focus on the number of links and/or the density of links. Interesting results have sprung from both types of testing. Conclusions have been drawn from productive association testing in four main areas. Firstly, form-related responses have been contrasted with meaning-related responses, and there are clear indications that both L1 and L2 beginners give more so-called clang responses than adult L1 learners and advanced L2 learners. Young L1 and L2 learners are hence much more likely to give a word like land when offered the stimulus word hand, while adult L1 learners and advanced L2 learners are more likely to produce the word finger when faced with the same prompt word (Cohen & Aphek, 1981; Meara, 1978; Piper & Leicester, 1980; Singleton, 1999; Söderman, 1993;
Introduction 11
Wolter, 2001). Secondly, syntagmatic responses have been compared with paradigmatic responses. Most research in this area shows that whereas beginners (in L1 as well as in L2) tend to offer words that are linked collocationally to the stimulus word, such as the word butter as a response to the prompt word bread, more advanced learners (again both in L1 and in L2) give words that are linked predominately in hierarchical ways to the prompt word, such as the word animal as a response to the stimulus word dog. Thirdly, prototypical responses (which can be either syntagmatic or paradigmatic) – that is, answers that a majority of informants tend to give to certain prompt words (Murphy, 2003; Singleton, 1999; Wilks & Meara, 2002) (e.g. ice, hot, freezing and blue as responses to the stimulus word cold) – have been contrasted with more infrequent responses (e.g. shoulder, hankie and dreary as responses to cold) (Albrechtsen et al., 2008: 33). Results of such studies show that native speakers generally produce many more prototypical responses than do non-native speakers, and that advanced L2 learners produce more such responses than do low- proficiency L2 learners. This is especially interesting in view of the fact that prototypical associative links appear to ‘play a central role in the structuring of the mental lexicon, perhaps functioning as bridges or pointers between different parts of the net’ (Albrechtsen et al., 2008: 34). Lastly, the frequency of the response word has been investigated. Research here shows that high achievers in an L2 are not only more likely to offer prototypical answers than low achievers, as described above, but are also more likely to produce more low-frequency responses than low achievers (e.g. Namei, 2002, 2004). From the above, it is evident that there is a difference in the number of form-related/meaning-related, syntagmatic/paradigmatic, prototypical/non-prototypical, and high-frequency/low-frequency responses given between L1 learners and L2 learners on the one hand and advanced L2 learners and low-proficiency L2 learners on the other hand. Nevertheless, Albrechtsen et al. (2008: 35) emphasise that [t]he whole lexicon as such cannot be described as either form-driven versus meaning-driven or syntagmatically versus paradigmatically structured. The structural properties of the lexicon will rather be determined by the language learner’s degree of knowledge of the individual lexical item. In the initial phases of learning, the semantic specifications in the lexical entry are relatively weak, and we thus find that formal factors play a significant role. When dealing with less familiar vocabulary items with weaker semantic specifications mapped onto the word the more advanced learner may still rely more on the form-driven, phonological information in the lexical entry. With increased word knowledge, the words become more and more meaning-driven. Paradigmatic or syntagmatic response types will be given, but the response type will be related to the degree of word knowledge.
Receptive association testing, focusing on, as mentioned above, the number and density of word links, has also produced some interesting results that help shed light on differences in the structure of the L1 and L2
12 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
lexical network. In Greidanus et al. (2005), for instance, native speakers clearly outperformed L2 learners in identifying association links between words. Moreover, Greidanus and Nienhuis (2001) and Greidanus et al. (2004) showed that the more advanced the L2 learners were, the more links they could identify. This difference was especially pronounced with high-frequency words. Albrechtsen et al. (2008: 38) conclude that [n]ative speakers are able to identify a significantly higher proportion of network links than are L2 learners. In other words, the results indicate that the mental lexicon of a language learner is qualitatively different; that is, less dense than that of a native speaker. An implication of this may be that an L2 learner’s retrieval paths are different as to the number of paths and their length; a fact that may well affect his ability to access words efficiently in L2 reception.
The receptive/productive dichotomy discussed in connection with the compositional approach can also be related to the connectionist framework. Meara (1990), for instance, describes words that are known productively as words that have many links to other words, while items that are only known receptively have few such association links, if any. Finally, within the compositional as well as the connectionist frameworks it is recognised that knowledge of a lexical item is never constant. In Wolter (2001), for instance, a model that places mastery of a word on a continuum from well-known core words to peripherally known words is put forth, allowing knowledge to advance or regress or even be lost. This aligns with Aitchison (1987), who considers a learner’s mental lexicon to constantly be changing. 1.4 The Chapter That Follows
In Chapter 2, setting the stage for the rest of the study, we will begin by exploring the 15 university students’ vocabulary breadth. Their results on three of the eight parallel tests administered – vocabulary taught at upper secondary school level, specialised uses and advanced vocabulary, the items within which range from very common to very infrequent – will therefore be investigated. Notes (1) This appears to be normal for a first-term university student group taking English as a single subject, as both Thagg Fischer (1985) and Karlsson (2002), focusing on the same type of informants, observed a largely similar distribution of gender. (2) Some linguists make a distinction between ‘learning’ and ‘acquiring’, where the former refers to knowledge having been assimilated consciously in, for example, a classroom setting, and the latter to knowledge having been assimilated subconsciously by, for instance, spending time in an English-speaking country, communicating with natives. In the present investigation, no such distinction is made and the two terms will therefore be used interchangeably.
2 Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary and Advanced Vocabulary
2.1 Introduction
An L2 student of English is faced with the no mean task of learning the language that contains more items of vocabulary than any other language in the world, the main reason for this cornucopia of course being the great influx of loanwords into English throughout history (Schmitt & Marsden, 2006). Estimates based on Webster’s Third International Dictionary of the English Language (1963) (Dupuy, 1974; Goulden et al., 1990) show that if compound words, obsolete words, proper nouns, abbreviations, dialect forms and alternative spellings are excluded, present-day English contains about 54,000 word families, a word family here incorporating the stem and inflected and transparent derivative forms (Nation & Waring, 1997). It is only if compound words, which form a very prolific area in the Swedish language, are included that Swedish is here on a par with the English language (Allén et al., 1989). Of these 54,000 word families, an educated native speaker knows about 15,000– 20,000 (Goulden et al., 1990; Nation, 1990; Zechmeister et al., 1995), having acquired 4000–5000 word families when starting school at the age of five and adding an average of 1000 word families each year (Nation, 2001: 9; Nation & Waring, 1997: 7). Even though the vocabulary of L2 learners appears to tally with that of native speakers in growth, the initial acquisition of 4000–5000 word families is very difficult to recover (Nation & Waring, 1997: 8). Attaining 15,000–20,000 word families is therefore almost impossible for any non-native speaker, especially if acquisition is to occur in a classroom setting only. Including L2 learners of different linguistic backgrounds, Laufer (2010) offers an overview of studies for which the number of hours of instruction is related to the number of word 13
14 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
families acquired, with the figures amounting to an average of two to three acquired words per hour of instruction. In one of Laufer’s own investigations (1998), high school students who studied English for 1500 hours acquired mastery of 3500 word families. In Miralpeix (2007), studying Spanish university students of English, one substudy, in which the informants had received 1100 hours of instruction, yielded a knowledge of 3750 word families, while another that included 900 hours of instruction yielded a knowledge of 2470 word families. In Shillaw (1995) and Barrow et al. (1999), who instead focused on Japanese college students, and where instruction involved between 800 and 1200 hours, the informants gained mastery of between 2000 and 2300 word families. In Nurweni and Read (1999), lastly, incorporating Indonesian university students, 900 hours of instruction were offered, yielding a knowledge of 600 word families. All these studies point to the fact that not even the highest number of hours yields the mastery estimated to have been acquired by a five-year-old native speaker of English. This can be contrasted with the number of word families required to listen to a radio interview without the repeated support of a dictionary, which is estimated to be 6000–7000 word families, and to be able to read newspapers and novels, which corresponds to having knowledge of about 8000–9000 word families (Nation, 2006). Encouragingly enough, the situation becomes more positive if learning takes place in an L2 environment. In Milton and Meara (1995), for instance, incorporating advanced European students, it was shown that there was an average growth of 2500 words per year. An L2 learner may also be helped further by the fact that a small number of reasonably frequent words make up a considerable part of the English language; that is, by acquiring these words. a learner will be able to understand large proportions of texts, both spoken and written. For instance, it has been shown that around 80% of any text in English is made up of around 1000 of the most frequently used words (of which only 9% are academic words, 5% are technical words, and another 5% are low-frequency words (Nation, 2001: 11–12)), the next 3–8% of another 1000 words of a slightly lower degree of frequency, and so on (Nation, 2006). Highfrequency items thus constitute an important foundation for any learner’s linguistic repertoire (Nation, 2001, 2006; Schmitt, 2008; Strömqvist, 2009), which means that commonality has a crucial place in the planning of any L1/L2 teaching syllabus (Hargevik, 1998; Milton, 2009). Research has shown that mastery of 3000–5000 high-frequency word families gives an L2 learner a reasonably good foundation for comprehension (Nation & Waring, 1997), with the number of word families required varying slightly according to the learner’s L1. An even smaller number of word families appears to be sufficient for production. For instance, in order to be able to take part in daily conversation, estimates have shown that knowledge of around 3000 word families will suffice (Adolphs & Schmitt, 2003; see also Nation & Waring, 1997). However, to be able to
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 15
approach written texts with minimal disturbance from unknown vocabulary appears to require 95–98% coverage, calculated to correspond to about 6000–7000 word families (Nation, 2006), and, if an L2 learner wants to be able to use their English in a wide variety of contexts, around as many as 10,000 word families need to have been acquired (Hazenberg & Hulstijn, 1996). To set the stage for the rest of the study, the 15 university students’ knowledge in three different areas – vocabulary taught at upper secondary school level, specialised uses of vocabulary, and advanced vocabulary – will be presented and analysed in the following subsections. With the items of vocabulary included ranging from very common to very infrequent, the subjects’ achievements in these areas are primarily intended to give a first indication of the size of the learners’ vocabulary knowledge in their L2 as compared to their L1. It is therefore interesting to note here that learners’ results on tests focusing on vocabulary size are usually also thought to be able to offer reasonably accurate indications of students’ general L2 proficiency (Chacón-Beltrán et al., 2010). 2.2 Vocabulary Taught at Upper Secondary School Level
In the three subsections that follow, the students’ mastery of words taught at upper secondary school level will be in focus. While the second subsection will present the tests implemented, the final subsection will be dedicated to presenting and analysing the informants’ results. As these tests include one receptive and one productive part, investigations focusing on this dichotomy will be offered in the first subsection. (For more about receptive versus productive vocabulary knowledge, the reader is also referred to Section 1.3.) Receptive versus productive vocabulary knowledge
According to Bogaards and Laufer-Dvorkin (2004), there are only five experimental studies that focus on comparing receptive and productive knowledge of vocabulary in a second language; brief summaries of three of these are offered here. In Griffin and Harley’s (1996) investigation, American high school students taking their first term of French were in focus. While half of the subjects were asked to learn the test items from an English–French perspective, the other half learned the words from the opposite perspective. Half of the informants from each of these two groups were then tested receptively, whereas the other students were tested productively. Quite a few conclusions could be drawn based on the results. Two of the most important findings were that the learners scored significantly higher on the receptive task than the productive one and that not only did receptive learning help improve productive learning but also the link was reciprocal.
16 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
In Waring’s (1997) study, Japanese students of English were asked to learn two sets of words, one receptively and one productively. Half of the subjects started with the test items to be learned receptively, while the other half started with those that were to be learned productively. Ten minutes after the informants had completed the session, they were tested on both sets of words. The participants were then tested three additional times: the day after the learning session, one week later, and then, for the third and last time, three months later. As in Griffin and Harley’s investigation, the results showed that the learners did considerably better on the receptive task than the productive one, and that both types of knowledge boosted each other. The results also showed that mastering items productively generally took more time than acquiring them receptively. In Schneider et al.’s (2002) research, American college students doing their first term of French were tasked with learning a set of L2 words. While half of the informants learned the words receptively and took a test structured to focus on this knowledge type immediately after the learning session, the other half went through the same procedure from a productive perspective. One week later a second test was administered. On this test, half of the subjects in each of the two groups were given a receptive test, whereas the other informants were given a productive test. As in Griffin and Haley’s and Waring’s investigations, the results of the study showed that the learners did considerably better on the receptive tasks, both on the test administered immediately after the learning session and on the delayed retention test, than on the productive tasks. Also, as in the other two studies, Schneider et al. were able to demonstrate that receptive knowledge increased the learners’ productive knowledge. However, this transference of knowledge was not as great as in the other two studies. Productive knowledge also helped increase the students’ receptive knowledge, to an even greater extent than that observed in Griffin and Harley’s and Waring’s work. The findings of these three experiments are confirmed in two older studies by Schuyten (1906) and Stoddard (1929). In fact, among researchers, there appears to be a consensus that receptive knowledge of vocabulary normally precedes productive knowledge, resulting in a lexicon where learners’ comprehension of words encountered receptively is greater than their ability to produce them. The parallel tests used for the present chapter
Both tasks implemented to test vocabulary taught at upper secondary school level (one in English and one in Swedish) were divided into two parts. In Part A, which tested receptive knowledge (50 test items), the informant group were asked on the L2 test to give Swedish translations of the English words offered, while on the corresponding L1 part they were requested to explain the meaning of the test items in their own words (in Swedish or English). In Part B, which focused on productive vocabulary
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 17
knowledge (also 50 test items), the learners were asked on the L2 test to produce the corresponding English words for the Swedish ones, whereas on the L1 part they were given definitions of the words sought. In this part on both tests, in cases where more than one answer was possible, the subjects were provided with the initial letter so as to force them to search for a particular word in their mental lexicon. While this may primarily be perceived as an advantage, it may also, if the student has come up with an acceptable synonym, be a stumbling block in their quest for the word starting with the requested letter. On both parts the students were given 1 point per correct answer, so a total of 100 points could be achieved. While the English test items were picked randomly from Weinius and Whitlam (1997), a booklet designed to test vocabulary appropriate for third-year upper secondary school students, the Swedish words were picked, also randomly, from Alstertun (1996a, 1996b), which focuses solely on vocabulary judged by the author to be appropriate to teach/learn in upper secondary school. All the test items were presented to the students in order of frequency based on their base forms, starting with the most frequent word. The frequencies were ascertained from the BNC and Språkbanken respectively (see Section 1.2). Furthermore, all of the items, whether tested receptively or productively, were context-based, with the context also adopted from the BNC and Språkbanken. An example of an item tested receptively from Part A is offered by (1). (1) Only the professor in a hospital department is allowed to do private practice and this must be done on the hospital premises.
In (2) an example of an item found in Part B (i.e. the productive task) is presented. (The word sought here was the adjective lukewarm.) (2) ljummen (the word sought starts with an l)
Mix all the ingredients in a bowl; add about 300 ml water to make to make a soft dough.
Moreover, the subjects were asked to evaluate their knowledge of each test item. How this appeared in Part A is exemplified in (3). (3) A second ballot will be held within fourteen days and a simple majority of those voting will be sufficient.
□ I don’t know this word. □ I’m guessing the word’s meaning from the context of the sentence. □ I recognise this word, but I don’t know what it means. □ I recognise this word and I’m guessing its meaning from the context of the sentence. □ I recognise this word and I think I know what it means. □ I am sure I know what the word means.
18 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
The learners could here indicate that they were sure they knew the word, that they thought they knew the word, that they recognised the word but did not know what it meant, or that they did not know its meaning at all. In addition, they could also indicate to what extent, if at all, they had used the word’s context to interpret its meaning. Here they could choose between not knowing the word at all but having tried to guess its meaning with the help of the context or recognising the word but still having used its context to arrive at an answer. The students’ self-evaluation of an item in Part B is exemplified in (4). (The word sought here was the noun invoice.) (4) faktura (the word sought starts with an i)
I attach a copy of a standard contract which I would be grateful if you could sign and return to me with the _____________________ for your fee.
□ I don’t know the English word. □ I think I know the English word. □ I am sure I know the English word. The informants could here indicate either that they were sure that they knew what word to insert, that they thought they knew what word to insert, or that they simply could not produce the word sought. Finally, the subjects were also asked to evaluate each part of the test as well as the tests as a whole. Here they could choose between ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. From the description above, it can be seen that the English and Swedish tests are constructed in almost exactly the same way. However, in order to be able to make comparisons between the students’ results on the two tests, the frequencies of the L1 and L2 test items need to be compared. As for the receptive part of the tests, the total frequency of the Swedish test items is somewhat lower (constituting ≈0.013% of Språkbanken) than for the English words (making up ≈0.040% of the BNC). From a frequency perspective, this makes the Swedish test part more difficult than the English one. The opposite is the case with the productive part, where the total frequency for the L1 items (≈0.025% of Språkbanken) is marginally higher than for the L2 items (≈0.022% of the BNC), rendering the English test part slightly more difficult than the Swedish one. As the total frequency for all the English items is higher (≈0.062%) than the total frequency for the Swedish items (≈0.038%), the L2 test is, in terms of frequency, more difficult than the L1 test overall. None of the L1–L2 differences in frequency were confirmed statistically, which means that comparisons may be made between the parallel tests/test parts. Still, the differences will be taken into consideration when discussing the results.
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 19
Results and discussion
Table 2.1 presents the students’ results on Part A (receptive part) and B (productive part) separately as well as on the tests as a whole. Achieving a total score of 81, the native speaker, as expected, easily outperformed the Swedish learners, who only received a mean score of 39.87; that is, the native speaker received more than double the number of points accumulated by the L2 learners. The Swedish participants’ meagre result was mainly due to their low mean score on the productive part (15.27), which tallies with the results seen in the investigations focusing on comparing receptive and productive knowledge discussed earlier, but it was also to some extent due to the surprisingly low score on the receptive part (24.60). It may be that, while the items incorporated were categorised as words suitable to teach at upper secondary school level by the authors of the booklet from which they were chosen, the words had not, due to not having been encountered enough times, been ingrained into long-term memory. As anticipated, and confirmed statistically at a significance level of 5%, the Swedish subjects also achieved better results in their mother tongue (mean: 74.73) than their second language (mean: 39.37). This occurred despite the fact that the total frequency of the Swedish test items was lower (constituting ≈0.038% of Språkbanken) than the total Table 2.1 The students’ results on the tests on words taught at upper secondary school level used in the present study Test part(s)
Students
N
Correctness rate
Mean
SD
Standardised scores
Parts A + B (receptive + productive knowledge)
Native speaker of English
1
81.00% (=81/100)
–
–
–
Swedish students – English test
15
39.87% (=598/1500)
39.87
16.96
Highest: 1.72 Lowest: −1.41
Swedish students – Swedish test
15
74.73% (=1121/1500)
74.73
10.37
Highest: 1.86 Lowest: −3.35
Native speaker of English
1
96.00% (=48/50)
–
–
–
Swedish students – English test
15
49.20% (=369/750)
24.60
10.84
Highest: 1.79 Lowest: −1.53
Swedish students – Swedish test
15
75.20% (=564/750)
37.60
8.86
Highest: 1.06 Lowest: −2.21
Native speaker of English
1
66.00% (=33/50)
–
–
–
Swedish students – English test
15
30.53% (=229/750)
15.27
7.08
Highest: 1.80 Lowest: −1.59
Swedish students – Swedish test
15
74.27% (=557/750)
37.13
6.55
Highest: 1.51 Lowest: −2.31
Part A (receptive knowledge)
Part B (productive knowledge)
20 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
frequency of the words on the L2 test (making up ≈0.062% of the BNC). The more difficult nature of the English test is further substantiated by the higher SD (compare 16.96 with 10.37). (The L1–L2 difference was also confirmed statistically at a significance level of 5% for the two test parts separately.) It is also interesting to note that, whereas there is no statistically confirmed difference between what the students achieved on the receptive part as compared to the productive part on the L1 test, which perhaps may partly be explained by the fact that the productively tested items were more frequent (≈0.025% of Språkbanken) than the receptive ones (≈0.013%), there is such a statistically confirmed difference (at a significance level of 5%) in the L2. This tallies with a connectionist framework, as discussed in Section 1.3, in which it is believed that more links need to have been created to achieve productive knowledge than for receptive knowledge. However, even though students’ knowledge is not only based on word frequency, this difference needs to be interpreted with some caution, as the items making up the L2 productive part were less frequent (≈0.022% of the BNC) than those making up the receptive part of the same test (≈0.040%), though no statistical difference was confirmed. On the other hand, as the reader will see in Tables 2.2 (L2 receptive knowledge) and 2.4 (L2 productive knowledge), more of the items tested productively (eight out of 50) were found on West’s (1953) general service list than those tested receptively (only one out of 50). All in all, this might indicate that the students’ L2 developmental trajectory is lagging behind their L1 curve, even for these advanced students dealing with a comparatively frequent subset of words. Furthermore, the students’ knowledge is mirrored accurately in their evaluation of the difficulty level of the test parts, with the learners indicating that Parts A and B were both easier on the Swedish test than on the English one, and that the productive part on the L2 test was more difficult than the receptive part on the same test. Tables 2.2 and 2.3 present the subjects’ results for each tested item in Part A on the English and Swedish tasks respectively. In both tables the words are presented in order of accuracy, with the item that received the highest score given first. (When more than one test item received the same score, the one for which the learners displayed most accuracy in their selfevaluation is given first. For test items A29 and A16, for instance, both of which received 14 points, gravy is placed before outskirts since 10 learners indicated accurately that they already knew gravy, while only eight wrote the same for outskirts.) Also, in Table 2.2, the one test item indicated in bold can be found on West’s (1953) general service list. To give an example, of the 15 subjects, 10 were able to translate the word cunning (test item 21) correctly into Swedish. The majority (five participants) indicated that they knew the word (column KN, highlighted in grey), whereas three wrote that they only thought they knew it (column TK). Furthermore, one student, even though they recognised the word,
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 21
Table 2.2 The students’ results on and evaluation of Part A (receptive knowledge) of the English test, with test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
English word given in context (correct answers)
KN
TKR
RC 2
NKR
NKC
NK
1
1
27
shove (down) (15)
7
6
29
gravy (14)
10
4
1
16
outskirts (14)
8
4
1 (1)
36
renounce (14)
4
4
5
23
aisle (12)
10 (2)
3
1 (1)
28
scatter (12)
7
5 (1)
2 (1)
35
bead (12)
8 (1)
3
5
poll (noun) (12)
5
3 (1)
3 (1)
41
wither (12)
4
6
1
9
vigorous (12)
3
4 (1)
4
6
thrust (into) (12)
3
3
3 (1)
37
stoop (verb) (12)
2
1
6
3
3
8
ballot (noun) (11)
2
1
2
2
6
1
21
cunning (10)
5
3
2 (1)
1
1
3
14
weary (adj.) (10)
5 (2)
2
5 (1)
1
1
1
50
brood (on) (10)
3
1
4
1
4 (2)
1
24
wrath (9)
4
4
2 (1)
1 (1)
4
18
merge (9)
5 (1)
4 (1)
3 (2)
2
49
conjurer (9)
1
3
2
2
expenditure (9)
26
rinse (verb) (8)
7
gravel (noun) (8)
3
2 (1)
1 4 (2) 2
2 (1)
1
2 (1)
1
1
2
1
1
5 (1)
2
1
9 (3)
2
1 (1)
6
1
2 (1)
2
premises (8)
3 (1)
9 (3)
2 (2)
1
30
detergent (8)
2
4 (1)
3
2
42
skid (verb) (8)
1
3
2 (1)
48
chirp (noun) (8)
1
2
2 (1)
38
allege (8)
5 (1)
3 (2)
3
20
tedious (7)
3
1
5 (4)
1
rural (7)
2
4 (1)
3 (1)
46
waver (verb) (7)
1
1
2
1 (1) 4
4
7 (1)
5 3
1 (1)
3
1 (1)
3
2 (1)
5
7 (3)
3
2
1
1
4 (2)
1
2
3 (3)
1
4 (1)
7 (Continued)
22 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.2 (Continued) No.
English word given in context (correct answers)
47
implore (7)
32
pretext (7)
2 (2)
1
5
15
pasture (6)
3
2 (1)
3 (3)
25
malicious (5)
4 (1)
3 (1)
5 (5)
17
vicinity (5)
2
4 (2)
34
toil (verb) (5)
1
45
dejected (adj.) (4)
12
brook (noun) (3)
3 (1)
31
miscellaneous (3)
2 (2)
39
zealous (3)
10
blaze (noun) (2)
2 (1)
11
sinister (2)
22
gale (2)
43
impertinent (2)
44
abominable (2)
4
revenue (2)
13
prone to (1)
19
apprehension (1)
40
bough (0)
33
venerable (0)
KN
TKR
RC
3 (1)
NKR
NKC
NK
1
8 (4)
2
3
1
3
1
3 (1)
3
1 (1)
2
3 (2)
6
3 (1)
2
5 (3)
3
2
3
6 (5)
4 (1)
2 (1)
1 (1)
2
3 (3)
4
2 (1)
1 (1)
4
2 (1)
3
1 (1)
3 (1)
2
9
1
1 (1)
6
1 (1)
3
1
6 (5)
2 (2)
3
1
2 (1)
1 (1)
2 (1) 1 (1)
3 3 (3)
7
1
3 (2)
9
2
3
2 (2)
7
3 (3)
3 (1)
3
2 (2)
4
3 (3)
2 (1)
5 (5)
2 (2)
2 (2)
2 (1)
1 (1)
1 (1)
Total
152 (24)
Correctness rate
84.2%
2 (2)
3
7
1 (1)
1
3
3 (3)
7
2 (2)
2 (2)
2
2 (2)
7
133 (33)
113 (47)
72 (2)
122 (59)
146 (3)
75.2%
58.4%
51.6%
97.9%
97.2%
could not offer a translation (column NKR), while three other students wrote that they did not know the word at all (column NK). Another three students used the context to figure out its meaning (columns RC and NKC). Two of these (one of whom offered an incorrect translation) indicated that, even though they recognised the item, they had based their answers on the context given (column RC), while the third student indicated that they did not know the word but had tried to infer its meaning from the context offered (and succeeded in doing so). If a comparison is made between Tables 2.2 and 2.3, the students’ mean scores are clearly mirrored here, with the learners having a considerably better grasp of the majority of the test items in their L1 than their L2 and being more certain of that knowledge. On the whole, a rather scattered picture is painted by the students’ self-evaluation of their mastery of the receptively tested L2 words; that is, the learners do not appear to have a very solid
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 23
Table 2.3 The students’ results on and evaluation of Part A (receptive knowledge) of the Swedish test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
Swedish word given in context (correct answers)
KN
3
avvakta (15)
(=wait and see)
15
5
illegal (15)
(=illegal)
15
18
donation (15)
(=donation)
14
TKR
RC
NKR
NKC
NK
1
47
utmärglad (15)
(=emaciated)
12
8
empati (15)
(=empathy)
12
3 1
1
2
konsument (15)
(=consumer)
11
3
1
40
enhällig (15)
(=unanimous)
9
4
2
7
naiv (14)
(=naïve)
14
1
28
temporär (14)
(=temporary)
13 (1)
2
17
disponera (14)
(=utilise)
10
4
4
kompetent (14)
(=competent)
10
3 (1)
1 1
39
verifiera (14)
(=verify)
9
6 (1)
11
absurd (14)
(=absurd)
9
4
1
1
26
åkomma (14)
(=complaint)
9
4
1
1
1
1
etablera (14)
(=establish, start)
8
5 (1)
37
emigrera (13)
(=emigrate)
12
3 (2)
12
integrera (13)
(=integrate)
8
3
3 (1)
1
1
1
32
anekdot (13)
(=anecdote)
7
6 (1)
20
amortera (13)
(=pay off by instalments)
7
6 (1)
1
41
tvetydig (13)
(=ambiguous)
6
7
1 (1)
1
27
profit (13)
(=profit)
6
5 (1)
3
1
30
förhala (13)
(=delay)
5
6
4 (2) 2 (2)
19
diffus (12)
(=diffuse)
10
2
10
misär (12)
(=destitution)
10 (1)
2
36
sporadisk (12)
(=sporadic)
9
1
14
motto (12)
(=motto)
8
5 (1)
23
febril (12)
(=frantic)
7 (1)
6 (1)
45
klenod (12)
(=treasure)
7 (1)
5
29
autentisk (11)
(=authentic)
10
3 (2)
1
1 1
2 (1)
1
1
1
1 1 (1)
2
1 1 (1)
1 (Continued)
24 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.3 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
2 (2)
1
25
frekvent (11)
(=frequent)
8
4 (1)
6
repertoar (11)
(=repertoire)
11 (3)
3
44
spartansk (11)
(=Spartan)
7
3
NKC
NK
1 (1) 2
49
utarmad (11)
(=depleted)
5
2
3 (1)
22
bjärt (11)
(=glaring, stark)
1
3
4
42
begeistrad (10)
(=enthusiastic)
9 (1)
4 (2)
1
2 3 (1)
1
3
4
1
1
16
utkast (10)
(=draft)
8 (1)
4 (1)
1 (1)
2
9
lågmäld (10)
(=quiet, unobtrusive)
9 (3)
3
1
1
48
stagnera (10)
(=stagnate)
6 (1)
6 (1)
1 (1)
21
medioker (10)
(=mediocre)
7 (2)
6 (2)
1
1
38
trivial (9)
(=trivial)
6
4 (2)
1
4
33
fadd (9)
(=insipid)
4 (1)
3
1
46
reprimand (8)
(=reprimand (noun))
6 (1)
1
4 (2)
1
3
43
underfundig (8)
(=subtle (humour))
1
6
2 (1)
3
3
31
lukrativ (7)
(=lucrative)
7 (2)
2 (1)
3
2
50
inkognito (7)
(=incognito)
5 (2)
4 (1)
3 (2)
2
34
jävig (5)
(=challengeable)
3 (1)
2
2 (1)
2
1 (1)
5
35
paroll (5)
(=watchword)
2
4 (2)
1 (1)
3 (1)
1 (1)
3
1
1 1 (1)
1
4 (2)
3
1
15
sympatisk (4)
(=likeable)
8 (5)
4 (3)
1 (1)
24
indignation (4)
(=indignation)
1
3 (1)
4 (4)
3 (2)
4
13
eftergift(er) (2)
(=concession(s))
4 (4)
3 (1)
2 (2)
2
1 (1)
3
Total
400 (31)
173 (30)
61 (26)
45 (1)
20 (11)
40 (0)
Correctness rate
92.2%
82.7%
57.4%
97.8%
45.0%
100%
knowledge of these items. While some, but far from an overwhelming majority, are categorised as known items, many others are either only thought to be known or not known at all. Still others were approached and interpreted by using their contexts. This then stands in stark contrast with the picture painted in connection with the Swedish test items, the vast majority of which most of the learners indicated were already familiar to them. Tables 2.4 (English test items) and 2.5 (Swedish test items) present the results for the 50 words that were tested productively (Part B).1 As in Tables 2.2 and 2.3, the items here are also presented in order of accuracy, starting with the item that received the highest score. (The items in bold are again found on West’s (1953) general service list.)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 25
Table 2.4 The students’ results on and evaluation of Part B (productive knowledge) of the English test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TK = the item is thought to be known, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
English word produced by the student (correct answers)
KN
TK
7
aerial (0) (BrE), antenna (14) (AmE) (14)
6
8
NK
9
criminal (12), culprit (1) (13)
9
3 (1)
1
28
cinnamon (12)
8
4 (2)
1
5
trolley (9) (BrE), cart (3) (AmE) (12)
5
7
3
6
divine (11)
8 (1)
4
3
44
tweezers (11)
6
2 (1)
4 (1)
17
beak (11)
4
6
4 (1)
6
23
travel plan (8), itinerary (3) (11)
1
15
almond (10)
10
38
autopsy (10)
6
5 (3) 3
5 (1)
2
29
detour (9)
7
2 (1)
3
35
asparagus (9)
5
6 (3)
2
22
hedgehog (8)
6
3 (2)
4
42
inflate (7)
4
3 (1)
5
49
mitten (7)
3
2
7
19
melt (5), thaw (2) (7)
2
6 (3)
5 (1)
37
cauliflower (6)
4
5 (4)
4
48
smirk, sneer (6)
4
5 (3)
3
21
kerb (3) (BrE), curb (3) (AmE) (6)
2
5 (1)
6
45
eavesdrop (5)
4
5 (4)
4
36
yolk (5)
3
5 (3)
5
25
(adjustable) spanner (0) (BrE), wrench (4) (AmE) (4)
4 (1)
1
7
10
invoice (4)
2
3 (1)
8
31
lukewarm (4)
2
3 (1)
8
20
beetle (4)
1 (1)
4 (1)
6
24
furnish (3)
1
8 (7)
2
1 (1)
3
temple (body part) (2)
3 (1)
16
umpire (noun) (2)
2
50
embezzle (2)
1
2 (1) 6 (4)
9
1 (1)
3 (1)
9
1
basin (2) (BrE), bowl (0) (AmE) (2)
27
perpendicular (2)
10 10 9
(Continued)
26 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.4 (Continued) No.
English word produced by the student (correct answers)
KN
TK
NK
41
gooseberry (2)
1 (1)
12
birch (1)
1
1 (1)
12
11
4
directory (1)
1
3 (3)
11
33
magpie (1)
1
2 (2)
10
26
(ring) spanner (0) (BrE), (box end) wrench (1) (AmE) (1)
2 (1)
5 (5)
6
13
rectory (1)
1
14
46
speedometer (1)
6 (5)
7
34
oblong (1)
2 (2)
8 (7)
2
1 (1)
32
dandelion (1)
1 (1)
9
18
pulpit (0)
3
10
8
pike (0)
2
13
2
40
perjury (0)
47
teetotaller (0)
11
beech (0)
10 13
1 (1)
2 (2)
12
39
truncheon (0)
12
43
hyphen (0)
12
2
turnover (0)
3 (3)
11
30
larder (0)
1 (1)
11
14
midwife (0)
1 (1)
6 (6)
6
Total
135 (12)
169 (83)
344 (6)
Correctness rate
65.7%
50.9%
98.3%
Table 2.4 indicates the more difficult nature of productive knowledge versus receptive knowledge in the students’ L2, in that many more of the English items were indicated as unfamiliar to the students (column NK). This is confirmed in the correctness rates for items indicated as known (84.2% for receptive knowledge (Table 2.2) versus 65.7% for productive knowledge (Table 2.4)) and items that were thought to be known (75.2% for receptive knowledge (Table 2.2) versus 50.9% for productive knowledge (Table 2.4)). Also, there is again a difference between the English and Swedish test items, for the latter of which most students indicated in an overwhelming majority of cases that they were sure of what word to produce. However, the reader is again reminded that the fact that the L2 items were in total slightly less common (≈0.022% of the BNC), though more frequently part of West’s (1953) general service list, than the L1 items (≈0.025% of Språkbanken) may have played a part in these results.
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 27
Table 2.5 The students’ results on and evaluation of Part B (productive knowledge) of the Swedish test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TK = the item is thought to be known, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
Swedish word produced by the student (correct answers)
KN
TK
NK
4
valuta (15)
(=currency)
15
20
rubrik (15)
(=headline)
15
25
ignorera (15)
(=ignore)
14
1
43
immun (15)
(=immune)
14
1
8
genre (15)
(=genre)
13
2
24
kontinent (15)
(=continent)
12
3
36
synonym (15)
(=synonym)
12
3
22
jubileum (15)
(=jubilee)
12
2
45
deodorant (15)
(=deodorant)
11
4
2
repris (14)
(=rerun)
15 (1)
12
sekel (14)
(=century)
13
1
1
30
megafon (14)
(=megaphone)
13
1
1
19
diktator (14)
(=dictator)
13
6
prioritera (14)
(=prioritise)
12
28
labyrint (14)
(=labyrinth)
12
26
nyans (14)
(=nuance)
11
3
1
48
reptil (14)
(=reptile)
11
3
1
23
vanilj (14)
(=vanilla)
9
4
2 (1)
5
permanent (13)
(=permanent)
13
40
patriot (13)
(=patriot)
12 (1)
2
1
27
kapitulera (13)
(=surrender)
9
4
2
13
neutral (13)
(=neutral)
9
3
1
15
dominera (13)
(=dominate)
8
4
3 (1)
18
monter (12)
(=display case)
12 (1)
2 (1)
1
42
staffli (12)
(=easel)
13 (2)
2 (1)
31
kuliss (12)
(=side-screen)
10
2
3
44
kranium (12)
(=skull)
10
2
3
3
kritiker (12)
(=critic(s))
9
4 (1)
2
1
majoritet (11)
(=majority)
11
1 (1)
2
16
transportera (11)
(=transport)
9
5 (3)
1
46
dräktig (11)
(=pregnant)
10 (1)
1
3
11
fiasko (11)
(=fiasco)
8
5 (3)
2
2 (1) 3 (1) 1
2
(Continued)
28 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.5 (Continued) No.
Swedish word produced by the student (correct answers)
KN
TK
NK
10
kompromiss (11)
(=compromise)
7
4
3
17
abstrakt (10)
(=abstract)
6
4
4
33
karikatyr (10)
(=caricature)
6
3
4
14
införsel (10)
(=import)
6
5 (2)
2
9
citat (9)
(=quotation)
9
41
analfabet (9)
(=an illiterate)
9
2 (2)
3
35
sil(h)uett (9)
(=silhouette)
7
6 (4)
2
32
spekulant (8)
(=prospective buyer)
6
2
5
21
objektiv (8)
(=unbiased)
5
5 (2)
5
37
steril (8)
(=sterile)
5
4 (1)
5
38
kaross (7)
(=body (car))
6
3 (2)
6
49
transfusion (6)
(=(blood) transfusion)
8 (4)
4 (2)
3
39
audiens (6)
(=audience (with king etc.))
2
5 (2)
8 (1)
7
dispens (5)
(=exemption)
3 (1)
3
7
47
monogram (5)
(=monogram)
1
8 (4)
6
29
hypotes (4)
(=hypothesis)
3
5 (4)
6
50
diligens (3)
(=stagecoach)
8 (6)
7 (6)
34
kortege (0)
(=procession)
5
6 (6)
7 (7)
2
Total
473 (23)
145 (49)
111 (4)
Correctness rate
95.1%
66.2%
96.4%
As for commonality on the receptive part, by comparing the total number of points accumulated by all the informants for the first 25 items (the most frequent test items) with the remaining 25 test items (the least frequent ones), no frequency effect could be detected on this test part, with the students even receiving a slightly higher score on the former part. In cases where learners know comparatively infrequent words better than more common ones, researchers have found that it may have to do with the fact that there is a focus on relatively advanced vocabulary in school settings, while words more common in spoken language are neglected; that is, some learners may have gaps in their basic vocabulary (Viberg, 1993). On the other hand, there is a slight frequency effect on the productive part. This effect is strengthened further by the fact that the frequency span was not as great in Part B as Part A. (In Part A, the most frequent item occurs 6178 times in the BNC, while the least frequent occurs 13
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 29
times, the corresponding figures for Part B being 3283 and 5, respectively.) It is quite logical that frequency is more important for productive vocabulary knowledge than receptive, as making it possible to use a word correctly probably requires more encounters in a variety of contexts than making it possible to understand a word. On the Swedish test, frequency had an effect on Part A as well as Part B. The students’ individual scores will be discussed in the last part of this chapter, together with their results for specialised uses and advanced vocabulary. (Their implementation of context in the disambiguation process will be investigated thoroughly in Chapter 7.) We will now turn to the students’ knowledge of specialised uses, exploring their vocabulary size even further. 2.3 Specialised Uses of Vocabulary
In the subsections that follow, the learners’ knowledge of specialised uses of vocabulary will be investigated. The first subsection aims to give a description of the degree of ‘technicalness’ of the items at hand. The second subsection will present the tests, while the last subsection will offer the reader insights into the participants’ mastery of the specialised uses tested. A continuum of specialised uses of vocabulary
The specialised vocabulary in focus here is that which is referred to as technical vocabulary, that is, words that are ‘recognisably specific to a particular topic, field or discipline’ (Nation, 2001: 198). Within each area there are, however, words that are more technical than others, with the degree of ‘technicalness’ spanning along a continuum. Nation (2001) offers such a continuum. Words found in Nation’s Category 1 are truly technical words and can therefore really only be acquired and used appropriately by thoroughly studying the specific field to which they belong. Examples of this type within applied linguistics would be morpheme, hapax legomena and lemma. Words belonging to Category 2 are slightly less technical. The reason for this is that these items also exist outside the field in question, albeit not with the same meaning. Examples here, again within the field of applied linguistics, are sense, reference, type and token. Items belonging to Categories 1 and 2 are not only typically low-frequency items but also will only occur in a very limited number of genres and disciplines (Nation, 2001: 200). In Francis and Kučera (1982), for instance, the word bargain could be seen in eight different types of texts and seven different disciplines, while the word styrene was only found in two different types of texts in two different disciplines, thus indicating that styrene is a truly technical word whereas bargain is not. There is
30 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
research that suggests words belonging to Categories 1 and 2 usually make up as much as 10% of a technically oriented text (Sutarsyah et al., 1994). Lastly, words belonging to Categories 3 and 4 are even less technical as neither their form nor their meaning exists in a specific discipline only, nor are they highly restricted to a specific discipline. Of the two, Category 4 words are the least technical in nature since their meaning is almost always the same, regardless of whether they occur inside or outside the relevant topic, field or discipline. Examples of words within applied linguistics belonging to these categories are range and frequency (Category 3) and word and meaning (Category 4). (For more examples in fields other than applied linguistics, the reader is referred to Nation, 2001: 198–199.) The parallel tests used for the present chapter
The two tests concerned with specialised uses of vocabulary (again one in English and one in Swedish) were both divided into four parts, with Part A dealing with school- and work-related words (40 test items), Part B focusing on words in the areas of business and commerce (20 test items), Part C testing vocabulary that pertains to medicine (another 20 test items), and the last part, Part D, dealing with words that are particularly used in the area of technology (also 20 test items). The majority of the test items belong to Categories 1 and 4, according to the continuum outlined by Nation (2001) discussed in the previous subsection. All four parts tested the students’ receptive knowledge by offering them emboldened words in context (taken mainly from the BNC and Språkbanken respectively) and asking them to give correct translations, each resulting in one point. On the Swedish test, the subjects were asked to explain the meanings of the items in their own words (in English or Swedish). Furthermore, in each part the items were presented to the students in frequency order based on the words’ base forms, always starting with the most frequent item. While both the English and Swedish terms dealing with school and work were picked on a random basis from a list put together by the Swedish Ministry of Education (http://www.regeringen.se/utbildning), the L1 and L2 terms included in Part B were, again randomly, taken from an online course in business English. For Part C, the test items were selected randomly from Hebbe (1999), the English items being translations of the L1 ones brought up in the book. Finally, while the English terms making up Part D were picked randomly from a word list of technical terms put together by Karlsson (1987), the Swedish test items were, again on a random basis, picked from another word list created by Engström (1995). An example from each part on the English test is offered in (5)–(8).
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 31
(5)
If you came from a poor family the only way you could get education at university was by being awarded a scholarship.
(6)
We got a very small discount considering the fact that we bought such a huge quantity.
(7)
He was lying on his back on the stretcher and appeared to be badly wounded.
(8)
If you lubricate a bit it will run more smoothly.
The informants were also asked to offer information about their knowledge of each word. This is exemplified in (9). (9)
Statistics already show that black women, less susceptible to the street violence and drug problems that undermine their brothers, are now more likely than black men to matriculate and graduate from university.
□ I don’t know this word. □ I’m guessing the word’s meaning from the context of the sentence. □ I recognise this word, but I don’t know what it means. □ I recognise this word and I’m guessing its meaning from the context. □ I recognise this word and I think I know what it means. □ I am sure I know what the word means. As on the receptive part of the test on vocabulary taught at upper secondary school level, the informants could here indicate to what degree they thought they knew the word and whether they had made use of its context. At the end of each test part, the informants were also asked to give their opinions on how difficult it was, the options being ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. Lastly, in order to be able to draw accurate conclusions about differences between the students’ L1 and L2 results, a comparison was also made between the frequencies of the test items. With the exception of the test items forming Part A (the L1 items making up ≈0.0242% of Språkbanken and the L2 items constituting ≈0.0145% of the BNC), the total frequencies of the test items on the other parts are lower on the Swedish task than on the English one (Part B: L1: ≈0.0024%, L2: ≈0.0208%; Part C: L1: ≈0.0017%, L2: ≈0.0031%; Part D: L1: ≈0.0022%, L2: ≈0.0057%), making the total frequency of all the test items lower on the L1 test (≈0.0305%) than the L2 test (≈0.0440%). Thus, from a frequency perspective, this renders the Swedish test slightly more difficult than the English one.
32 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
None of the L1–L2 frequency differences above were confirmed statistically, which means that comparisons may be made between the parallel tests/test parts. Still, the differences will be taken into consideration when discussing the results. Results and discussion
Table 2.6 presents the students’ results on the test as a whole as well as their results on the four test parts in both languages. As with the test on vocabulary taught at upper secondary school level (see Table 2.1), the native speaker here too scored considerably higher Table 2.6 The students’ results on the tests focusing on specialised vocabulary used in the present study Test part(s)
Students
N
Correctness rate
Mean
SD
Standardised scores
Parts A + B + C + D
Native speaker of English
1
95.00% (=95/100)
–
–
–
Swedish students – English test
15
52.80% (=792/1500)
52.80
13.91
Highest: 1.52 Lowest: −1.42
Swedish students – Swedish test
15
73.93% (=1109/1500)
73.93
13.24
Highest: 1.36 Lowest: −1.88
Native speaker of English
1
95.00% (=38/40)
–
-
-
Swedish students – English test
15
57.50% (=345/600)
23.00
6.35
Highest: 1.58 Lowest: −1.26
Swedish students – Swedish test
15
80.00% (=480/600)
32.00
3.42
Highest: 1.46 Lowest: −2.05
Native speaker of English
1
100.00% (=20/20)
–
–
–
Swedish students – English test
15
47.67% (=143/300)
9.53
3.78
Highest: 1.71 Lowest: −1.46
Swedish students – Swedish test
15
62.00% (=186/300)
12.40
3.36
Highest: 2,27 Lowest: −1.61
Native speaker of English
1
95.00% (=19/20)
–
–
–
Swedish students – English test
15
63.00% (=189/300)
12.60
3.81
Highest: 1.68 Lowest: −1.73
Swedish students – Swedish test
15
74.33% (=223/300)
14.87
3.46
Highest: 0.91 Lowest: −2.56
Native speaker of English
1
90.00% (=18/20)
–
–
–
Swedish students – English test
15
38.33% (=115/300)
7.67
3.56
Highest: 1.78 Lowest: 1.87
Swedish students – Swedish test
15
73.00% (=219/300)
14.60
4.64
Highest: 1.16 Lowest: 1.85
Part A (school/work)
Part B (business)
Part C (medicine)
Part D (technology)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 33
than the L2 learners, again achieving a total score of almost double that produced by the Swedish learners (95 points versus a mean score of 52.80), hence emphasising the greater size of his mental lexicon as compared to the non-native speakers. In addition, as the reader will see in connection with Table 2.7, parts of the Swedish students’ achievements with the English test items have to be attributed to their ability to infer the meanings from context. This was not the case with the native speaker who was able to give almost all of his correct answers based on previously acquired knowledge, at least as indicated by his self-evaluation. (The L2 learners’ inferencing skills will be explored in detail in Chapter 7.) Percentage-wise the Swedish subjects did best in Part C, followed by Part A, Part B and, lastly, Part D. That the students did comparatively well in Part A (school- and work-related words), which made up the largest part of the test, is not surprising. Since the vast majority of the participants were in their late teens/early twenties when they took these tests, they had probably been working, possibly abroad, and/or travelling for a few years before coming back to take part in tertiary education. It may also be that they chose to study English because they intended to go abroad to work or study in the near future. Either way, the items forming Part A were highly relevant to the developing stage at which most of these students were at the time of the testing. This result needs to be interpreted with some caution, however, as this was the only part of the test where the Swedish test items were more frequent (≈0.0242% of Språkbanken) than those making up the English test part (≈0.0145% of the BNC). (In the L2 data, statistical differences were confirmed (at a 5% significance level) between all subparts of the test except Parts A and C, while in the L1 data no such confirmation could be gained between Parts A and C and between Parts C and D.) As expected, the informants also did much better in their L1 than their L2 (73.93 as compared to 52.80). This was statistically confirmed at a significance level of 5% and given additional validity due to the fact that the English test items were more frequent on the whole (≈0.0440% of the BNC) than the Swedish ones (≈0.0305% of Språkbanken). (The L1–L2 difference was also confirmed for all four test parts separately, at the same significance level as the test as a whole.) They did not, however, reach the L1 level that the native speaker of English achieved in his first language. Part of the reason for this may be that, as mentioned above, the Swedish items were more difficult than the English ones in terms of frequency. No clear trends according to the categories of degree of ‘technicalness’ outlined in Nation (2001), as discussed in the theoretical background, could be detected. It is also worth mentioning that the students’ L2 mean score on specialised uses is considerably higher (52.80) than their L2 mean score on vocabulary taught at upper secondary school level (39.87), despite the fact that the former test items were in total less frequent (≈0.0440% of the
34 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.7 T he students’ results on and evaluation of Parts A + B + C + D on specialised uses on the English test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
English word given in context (correct answers)
KN
TKR
RC
NKR
NKC
C19
dental floss (15)
10
5
B3
discount (15)
8
5
B1
go on sale (15)
8
4
1
2
B6
bankruptcy (15)
5
4
4
1
A5
credit(s) (15)
3
7
3
1
A9
work experience (14)
10
4 (1)
1
C8
braces (14)
7
6
A40
upper secondary school teacher(s) (14)
5
7
1
1 (1)
A37
parent-teacher meeting (14)
2
7 (1)
2
3
A39
single-subject course (14)
2
3
1
6 (1)
C17
collar(-)bone (13)
6
5
1 (1)
1
D15
exhaust(-)pipe (13)
5
4
2 (1)
1
D4
radiator (13)
4
6
2
1 (1)
C11
stretcher (13)
4 (1)
4
3
A2
higher education (13)
3
4
6 (1)
2 (1)
A7
trainee (13)
2
6 (1)
4
2 (1)
A23
permanent position (13)
2
4 (1)
4
A4
scholarship (13)
10
2 (1)
1
A3
lecturer (12)
6
4
3 (2)
C10
health insurance (12)
6 (1)
6 (1)
2 (1)
C3
anaesthetic (12)
5
3
2
A13
academic year (12)
4 (1)
1
6 (2)
D7
circumference (12)
3
3
2
A20
compulsory school (12)
3
2
7 (1)
1(1)
D6
gearbox (12)
2
6 (1)
2
3 (1)
C9
congested (nose) (12)
2
1
1
A17
unemployment benefits (12)
2 (1)
4
6 (1)
C6
forefinger (12)
5
4 (2)
A27
compulsory attendance (11)
3
7 (2)
1
1
1 (1)
NK
1
1 (1)
1
3
3 2 2 (1) 1
1
2 (1)
1
4 1
5 (1)
8 (1) 1 (1)
4 1 (1)
2
1 1
3 (2) (Continued)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 35
Table 2.7 (Continued) No.
English word given in context (correct answers)
KN
TKR
RC
NKR
NKC
NK
C15
chicken(-)pox (11)
5 (2)
4
1
3 (1)
A38
probationary employment (11)
3
1
3 (1)
7 (3)
A22
visiting professor (11)
1
2
1
9 (2)
A26
language of instruction (11)
1 (1)
4
2
C14
appendicitis (10)
5
4
1 (1)
3
B2
recession (10)
4
4
3 (2)
1 (1)
2
A1
curriculum (10)
3
3 (1)
5 (1)
2 (1)
1
A21
supply teacher (10)
3 (1)
2
3 (1)
5 (1)
2
B5
the stock exchange (10)
2
3
3 (1)
6 (3)
1
C4
physiotherapist (109
2
3 (2)
4 (1)
A8
university college (10)
3 (2)
5 (1)
5 (1)
D2
indicator (car) (10)
1
3 (1)
3 (1)
6 (2)
1
A28
doctoral dissertation (10)
1
2
1
6
5
C12
spleen (9)
7 (2)
2
3 (1)
1 (1)
2
C16
resuscitate (9)
4
2
2
3 (2)
3
B7
retailer (9)
4 (1)
3
3 (1)
3 (3)
1
A12
the humanities (9)
3 (1)
4 (2)
5 (1)
2 (1)
1
B20
statement of account (9)
2
3 (1)
1 (1)
1
3
1
A15
physical education (9)
2 (1)
5 (1)
3
2
1
1
A32
school-leaving certificate (9)
2
1 (1)
1 (1)
8 (2)
2
A34
pass with distinction (9)
2 (1)
2
3 (1)
1
4 (2)
1
A31
doctoral studies (9)
3 (1)
3 (1)
1 (1)
6 (3)
1 (1)
C13
paediatrician (8)
5 (1)
4 (1)
1
3 (2)
1
D16
lubricate (8)
4 (1)
2 (1)
2
2 (1)
D8
goggles (8)
3 (2)
6 (1)
2
2
B11
instalment (8)
1
1
2 (1)
1
A6
supervisor (7)
4 (2)
1
8 (5)
C1
general practitioner (7)
1
2 (1)
1
B13
window(-)dressing (7)
1 (1)
4 (1)
2 (2)
B14
gross income (7)
1 (1)
2
2
D19
spark(ing)(-)plug (6)
3
1
C5
pelvis (6)
2
5 (2)
B19
time of prosperity (6)
2
1
A18
leave of absence (6)
1
5 (3)
2 (1)
A10
head of department (6)
3 (2)
4 (3)
4 (2)
1 (1)
3 (2)
A36
general eligibility (6)
2 (1)
1 (1)
3 (1)
1
3 (1)
4
B8
budget deficit (6)
2 (1)
1
2
7 (3)
3
2
7 (2) 1
5 (2) 1
1
1
1 1 6 (2)
3
2 (1) 1
8 (4)
2
6 (2)
2
2
3 (2)
3
1
3 (2)
5 (1)
1
1
1 (1)
5
5 (3)
2 (1)
4 7 (5)
(Continued)
36 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.7 (Continued) No.
English word given in context (correct answers)
KN
TKR
RC
NKR
NKC
NK
A30
the Higher Education Act (6)
1 (1)
1 (1)
1
1
6 (3)
3 (2)
C2
molar(s) (5)
2 (1)
6 (3)
1 (1)
5 (4)
1
B16
down payment (5)
1
3 (1)
3 (2)
1
6 (5)
1
D18
direct current (5)
2 (1)
1
2 (2)
2
3 (1)
5 (1)
B17
golden hello (5)
1
8 (4)
4
B10
overdraft (5)
1 (1)
D10 B9 C18
mumps (4)
A11 C7 D11
weld (4)
D14
carburettor (4)
1 (1)
B12
wholesaler (4)
1 (1)
2 (1)
D17
alternating current (4)
1
2 (1)
D12
solder(ing) (3)
1
2
C20
geriatric care (3)
1
1
D1
vice (noun) (2)
1
D20
wheel hub (2)
2 (1)
B15
golden handshake (2)
1
A29
Bachelor of Arts (2)
1 (1)
D5
alloy (noun) (2)
A35
course coordinator (3)
A19
interdisciplinary research (2)
A16
throughput (2)
1 (1)
D3
valve (1)
1
D9
piston (1)
A25
Master of Arts (1)
A24 A33
3 (1)
2
2
5 (4)
1
chisel (5)
1
1
2
4 (2)
6
Chancellor of the Exchequer (5)
1 (1)
2
2
5 (2)
4
2
1
1 (1)
1 (1)
2 (2)
7
deputy head (4)
2
2 (2)
6 (2)
3 (3)
2
syringe (4)
3 (1)
1
1 (1)
1
3 (2)
6
4 (1)
1 (1)
1
4 (3)
3
2
1
1
2 (1)
7
2 (1)
2 (1)
5
2
3 (2)
4
1
5 (5)
5
1
1 (1)
8
2 (1)
1
3 (3)
7
3 (3)
2
4 (4)
3
7
5
4 (3)
4
3 (2)
11
7 (5)
1
5
3 (3)
5 (1)
2
4 (3)
8 (1)
2 (2)
3
5 (5)
4
1
1 (1)
2
1
3 (3)
1
3 (3)
4
civics (1)
1 (1)
1 (1)
3 (1)
2 (2)
7
matriculate (0)
1 (1)
3
1 (1)
10
B18
interest yield (0)
1 (1)
2
3 (3)
7
B4
the Treasury (0)
2 (2)
2
4 (4)
6
D13
lathe (noun) (0)
1 (1)
1 (1)
12
A14
vocational training (0)
5 (5)
5
1
3 (2)
2 (1)
1 (1)
2
1 2 (2)
5 (4) 1
1 (1)
1 (1)
10
1 (1)
1 (1)
2 (2)
Total
254 (40)
286 (59)
201 (67)
86 (13)
330 (163)
254 (8)
Correctness rate
84.3%
79.4%
66.7%
84.9%
50.6%
96.9%
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 37
BNC) than the latter ones (≈0.062%). This again emphasises the difficulty in acquiring productive vocabulary knowledge, which made up half of the first test discussed in the present chapter. The students’ achievements are also reflected comparatively well in their self-evaluation, in which, for example, they indicated that they thought the words concerned with school and work formed the second easiest part of the English test, with Part C being considered the least difficult part. Unsurprisingly, the self-evaluation also shows that the informants generally thought that the Swedish test parts were easier than the corresponding English ones. In the two tables that follow, the students’ achievements for each item of specialised uses of vocabulary are presented, with Table 2.7 showing the results on the L2 test and Table 2.8 the results on the L1 test. The items belonging to the four different test parts (school/work (Part A), business (Part B), medicine (Part C), and technology (Part D)) have here been merged into one table for each language. As before, the items are presented in order of accuracy, starting with the one for which most of the students were able to offer a correct translation. (Again, when more than one item received the same score, the word for which the subjects showed the greatest accuracy in their evaluation comes first.) To exemplify, 14 of the 15 learners were able to offer a correct translation of the word braces (test item C8). While seven indicated that the item was already known to them (column KN, coloured grey since these students were in the majority), six were less sure, indicating that they only thought they knew the word (column TK), and one learner did not know the word at all (column NK). Yet another student indicated having used the word’s context to arrive at a translation (column NKC). The picture painted by the informants’ evaluation of their L2 knowledge is as scattered as the one seen with vocabulary taught at upper secondary school level, again not giving a very clear impression of their mastery of these English words. With the Swedish words, on the other hand, the students show considerably more certainty, as already indicated by their mean. This is confirmed in the correctness rates for items indicated as known (84.3% on the L2 test (Table 2.7) versus 96.9% on the L1 test (Table 2.8)) and items that were thought to be known (79.4% on the L2 test (Table 2.7) versus 97.5% on the L1 test (Table 2.8)). Also, the students’ knowledge displays a frequency effect. In fact, with the exception of Part C on the English test (focusing on medical terms), on all of the test parts, the first half of the items (the most frequent ones) received more points than those forming the latter part (the least frequent test items). This applies to both languages. The students’ individual achievements will be discussed in Section 2.5, together with their results for vocabulary taught at upper secondary school level and advanced vocabulary. (Their use of context will be investigated thoroughly in Chapter 7.) We will now continue by looking into their mastery of advanced vocabulary, exploring their vocabulary size even further.
38 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.8 The students’ results on and evaluation of Parts A + B + C + D on specialised uses on the Swedish test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
Swedish word given in context (correct answers)
KN
TKR
RC
A16 obehörig (15)
(=unqualified)
10
2
3
A17 dyslexi (15)
(=dyslexia)
9
4
2
A33 syo (15)
(= study and careers adviser)
9
4
2
A23 högskolepoäng (15)
(=credit (at university))
9
4
1
C10 senil (15)
(=senile)
9
4
1
A2
enkät (15)
(=questionnaire)
8
6
1
D2
expandera (15)
(=expand)
8
6
1
A21 särskola (15)
(=special school)
8
6
1
A19 internatskola (15)
(=boarding school) 8
4
3
C14 kuvös (15)
(=incubator (for children))
8
4
2
A10 bevilja (15)
(=grant (verb))
8
3
3
A3
rekrytera (15)
(=recruit)
7
8
B1
faktura (15)
NKR
(=invoice)
7
5
2
(=flake (off))
7
5
2
A1
(=managing director)
6
8
1
B14 förmyndare (15)
(=guardian)
5
8
1
C17 stetoskop (15)
(=stethoscope)
5
4
1
A18 varsla (14)
(=give notice of a strike)
9
4 (1)
2
A34 nationellt prov (14)
(=national test)
9
4
1
1 1
D4
stabilisera (14)
(=stabilise)
9
3
2
C1
kronisk (14)
(=chronic)
9
2
4 (1)
C5
gynekolog (14)
(=gynaecologist)
8
5
A40 SO-ämne(na)
(=social studies)
8
4
A8
(=unemployment benefit)
8
4
a-kassa (14)
A11 merit (14)
(=qualification)
7
6 (1)
A13 humaniora (14)
(=the humanities)
5
6
NK
1 (1)
D17 flagna (15) vd (15)
NKC
1
1 2 (1) 2 (2) 1 (1) 3 (1) (Continued)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 39
Table 2.8 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
NKC
A24 omskolning (14)
(=retraining)
6 (1)
5
4
A20 kriterium (14)
(=criterion)
5
4
5 (1)
A25 inskolning (14)
(=acclimatisation (at school))
4
7
2 (1)
1
A32 tentera (14)
(=be examined (at 4 school))
7 (1)
3
1
C11 logoped (14)
(=speech therapist)
4
6
1
D11 sekatör (14)
(=secateurs)
8
3
2
A9
(=compulsory school)
10 (2)
2
1
grundskola (13)
1 (1)
C3
kirurg (13)
(=surgeon)
8
3
1
3 (1)
disponera (13)
(=make use of)
7
5
1
2
A15 CV (13)
(=CV (curriculum vitae))
7
4
D1
konsistens (13)
(=consistency (material))
6
6
C6
reumatism (13)
(=rheumatism)
6
6
(=technical terminology)
7 (1)
4
1
1
1
2 2 (1)
1 1
1
2
1
A5
rehabilitering (13)
(=rehabilitation)
5
8 (1)
1
1
D3
vakuum (13)
(=vacuum)
5
6
2
1
A27 yrkeslärare (13)
(=vocational teacher)
5
4
1
2
C7
anatomi (13)
(=anatomy)
5
4 (1)
4
1
B2
kredit (13)
(=credit)
5 (1)
4
5 (1)
B19 ajournera (13)
(=adjourn)
4
5
2
B4
(=audit (noun))
4
2
6
revision (13)
1
1 (1)
B3
A22 fackspråk (13)
NK
3
2
1
1
1
1
1
A14 blockad (13)
(=boycott)
3 (1)
6
4
1
D15 ampere (13)
(=ampere)
2
8
1
2 (1)
A39 preparandkurs (13)
(=preparatory course of study)
2
2
1
7
3 (1)
D8
(=a water-level (tool))
9
3
1 (1)
1 (1)
1 1
vattenpass (12)
A6
dispens (12)
(=exemption)
7
2 (1)
2
2 (1)
1
C2
trauma (12)
(=trauma)
6
3
1
3
1
(=a jack (tool))
6 (1)
4
2
1
1
5 (2)
5
2
D12 domkraft (12)
A30 hemspråksundervisning (=L1 instruction in (12) an L2 setting)
1 1 (Continued)
40 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.8 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC 3 (1)
C20 dermatolog (12)
(=dermatologist)
4
4
C19 pediatriker (12)
(=paediatrician)
4
4
C9
farmakologi (12)
(=pharmacology)
1
5
D5
frekvens (11)
NKR
2 1
NKC
NK
1
2
4 4
4 (1)
(=frequency)
6 (1)
6
1 (1)
D19 armera (11)
(=reinforce (material))
6 (1)
3 (1)
4 (2)
A35 anpassad studiegång (11)
(=curriculum for weak student)
5 (1)
4 (1)
3
2
C8
ortoped (11)
(=orthop(a)edist)
5 (1)
3 (1)
3
1
B8
filial (11)
(=branch (business))
3
3
3
2 (1)
B7
klausul (11)
(=clause, proviso)
3
3
4 (1)
1 (1)
1
B18 nettolön (11)
(=net pay)
5 (1)
4 (1)
2
2 (1)
1
D9
(=cylinder)
6
5 (2)
1
1
1 (1)
1
D10 ekolod (10)
(=sonar)
5
5
3
1 (1)
1
D14 koagulera (10)
(=coagulate)
5
5 (1)
2
1 (1)
1
C4
fraktur(er) (10)
(=fracture)
5
6 (2)
D20 navkapsel (10)
(=hub-cap)
4
4
2
1 (1)
1
A7
1
1 (1)
cylinder (10)
stipendium (10)
2 2
1
1 2 2
2 (2)
(=scholarship)
6 (2)
6 (1)
A29 examinator (10)
(=examiner)
5 (1)
2 (1)
7 (2)
B12 prospect (10)
(=prospectus)
2
1
6 (1)
1
3 (2)
1
4 (1)
1
1
2 (1)
2
2
3
1
B15 bordlägga (10)
(=postpone)
2 (1)
5 (1)
A12 SFI (9)
(=Swedish for immigrants)
6
2 (1)
D16 kondenserad (9)
(=condensed)
4 (1)
8 (2)
1 (1)
2
D13 symmetri (9)
(=symmetry)
5 (2)
7 (3)
2
1
B6
(=quotation (price))
4 (1)
4
2 (2)
3 (1)
1
C15 pandemi (9)
(=pandemic)
3
5 (1)
1 (1)
4
D7
(=refinery)
4 (1)
3 (1)
3 (1)
1 (1)
2 (1)
2
B11 dödsbo (9)
(=estate (of a 4 (1) deceased person))
3 (1)
2
1 (1)
3
C18 obstetriker (9)
(=obstetrician)
1 (1)
2
2
5
5
A26 prorektor (8)
(=pro-vicechancellor)
3 (1)
2
4 (1)
5
B5
(=premium)
3
2
3 (2)
3 (1)
1 (1)
3
(=diagnostic test)
2
5 (2)
1
2
offert (9)
raffinaderi (9)
premie (7)
A36 diagnostiskt prov (6)
5 (Continued)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 41
Table 2.8 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
NKC
NK
B16 postförskott (6)
(=cash/collect on delivery)
2 (1)
7 (3)
3 (2)
2
A4
(=docent, associate professor)
1 (1)
6 (4)
3 (1)
1 (1)
(=allowance for expenses)
3
3 (2)
1
A38 lasa in4 (5)
(=see note)
3 (1)
2 (1)
3 (1)
1
1 (1)
3
A28 tvärvetenskaplig (5)
(=interdisciplinary) 3 (1)
3 (2)
2 (1)
1
1
4
B17 plenum (5)
(=plenary meeting)
1
2
1 (1) 2
docent (6)
B13 traktamente (5)
D18 etsa (5)
(=etch)
2 (1)
4 (2)
3 (1)
D6
(=radius)
6 (3)
5 (4)
3 (3)
A31 disputera (4)
(=defend a doctor’s thesis)
1
4 (2)
B9
(=devaluation)
2 (1)
1
1
B20 tariff (4)
(=tariff)
1
3 (2)
1
C13 odontologi (4)
(=odontology)
3 (1)
2 (1)
C16 konvalescent (3)
(=convalescent)
radie (4)
devalvering (4)
2
1
1 1
1
7
11 (1) 1 (1)
2 1
3
2 (1)
5
1
1
9
4 (3)
6
1
2 (1)
6
2
1 (1)
9
C12 artrit (3)
(=arthritis)
2 (1)
2 (1)
2 (2)
3
1
3
B10 likviditet (2)
(=liquidity)
1
3 (2)
1 (1)
5
3 (3)
2
A37 promovering(en) (1)
(=conferment of a doctorate)
1
1(1)
416 (52)
192 (41)
13
Total
508 (36)
92 (17)
82 (25)
146 (4)
Correctness rate
96.9% 97.5% 78.6% 81.5% 69.5% 97.3%
2.4 Advanced Vocabulary
The present section will be concerned with the 15 university students’ mastery of advanced vocabulary, with the first subsection offering the reader examples of items belonging to this type. The subsection that follows will give a detailed description of the tests used in the present investigation, while the final subsection will present and analyse the learners’ results. What makes a word advanced?
There are a number of reasons why words may fall into the category of difficult vocabulary, hence further increasing the possibility of such words being considered advanced vocabulary. Laufer (1997b: 142–153)
42 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
lists quite a few such factors. Firstly, whether a word is categorised as advanced or not may have to do with its pronounceability (Ellis & Beaton, 1993; Higa, 1965). In Celce-Murcia’s (1978) study, for example, it was shown that the researcher’s bilingual daughter always preferred the word that was easier to pronounce. The child thus chose to use the French word couteau instead of its English counterpart, since she seemed to feel that the f-sound in knife was too difficult. Similarly, the girl also preferred to use boy to garςon since the r-sound caused her difficulties. In fact, the phonological characteristics of a word have also been shown to have an impact on how well the word is remembered. The more difficult its phonological make-up, the more likely it is that the word will be forgotten (Rodgers, 1969). Ultimately, when it comes to the phonological structure of a word, it appears that it is the learner’s first language that determines which L2 words are conceived as more difficult than others. The spelling of a word may also create problems. This is especially true of English words since, whereas pronunciation has changed dramatically over the last few centuries, orthography has not, at least not on par with pronunciation (Moseley, 1994). The same incongruence is not present in Swedish, in which spelling appears to tally well with words’ pronunciation in the vast majority of cases. Some studies have also shown that long words are more difficult than short ones (Gerganov & Taseva-Rangelova, 1982). There are, however, other investigations that present the opposite conclusion (Stock, 1976), so no clear conclusions can be drawn in this respect. The morphological make-up of a word, especially derivative forms, may also have an effect on the ease or difficulty with which a word is understood and acquired. Here, a learner’s ability to decompose words into stems and affixes comes into play. It is generally thought that the better a learner is at decomposing, the more easily they will understand previously unknown morphologically complex words. (The reader is here referred to Chapter 3, in which the 15 students’ L1 and L2 mastery of derivatives and word families is explored in more detail.) In addition, some words that from a morphological perspective display deceptive transparency also seem to be difficult to gain mastery of (Laufer, 1989b). This is the case with the word infallible, which superficially may appear to be made up of the prefix in-, the stem -fall-, and the suffix -ible (Laufer, 1997a). One such example was also found in the present study, where a student believed that the word remonstrate consisted of the prefix re- and the stem -monstrate. Another area that may cause problems is that of synforms, that is, words that are similar in form. In Laufer (1991), 10 different types were found to be especially problematic to the L2 university students tested. The most troublesome were those in which the stem of the words was identical but the suffixes differed, as in, for instance, the word pair c onsiderable – considerate. Other examples of synforms causing
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 43
problems in Laufer’s study are consumption, assumption, and resumption, where the synforms have the same stem but different prefixes (none of which are used to form new words in present-day English); affect versus effect and quite versus quiet, which are identical except for the vowel and diphthong respectively; and ledge versus pledge, where the only difference is the presence of a consonant in one of the forms. For a more detailed discussion on synforms, the reader is referred to Laufer (1991). In connection with synforms, false friends – that is, words that are orthographically the same in a learner’s L1 and L2 but differ in meaning – need to be mentioned, as they too seem to cause a lot of extra problems for L2 learners. (False friends and L2 learners’ mastery of them will be discussed more thoroughly in Chapter 6.) Finally, semantic properties may also play a role in how difficult a word is to a learner. While the degree of abstractness appears to have comparatively little effect (Laufer, 1997b), the levels of specificity, transparency, and polysemy seem to play major roles. Regarding specificity, Blum and Levenston (1978), for instance, found that, whereas L2 learners have a proclivity to use more general terms (i.e. superordinates) that are able to cover a wider range of contexts, native speakers do not, where appropriate, shy away from using more specific terms (i.e. hyponyms). (The reader is again referred to Chapter 6, in which the students’ mastery of lexical fields of near synonyms will be explored.) As for transparency, Marton (1977) and Dagut and Laufer (1985) noticed that idioms and phrasal verbs, both of which may display high degrees of opaqueness, also cause L2 learners great problems. (For more in-depth analyses of these areas, the reader is referred to Chapters 4 and 5, the former dealing with idioms and the latter with multi-word verbs.) As for polysemy, findings indicate that words with more than one meaning are often more difficult to acquire (e.g. Bensoussan & Laufer, 1984). This is troubling from a second-language-learner perspective because many polysemous words also appear to be comparatively frequent (e.g. Gyori, 2002; Murphy, 2004). (The reader can gain a deeper understanding of learners’ knowledge of polysemous words in Chapter 6.) A great many of the word types outlined above may, due to their difficult nature from an L2 perspective, thus fall into the category of advanced vocabulary. However, the most common reason for categorising a word as advanced has to do with its degree of commonality. Low-frequency words make up more than 10% of the running words in a ‘normal’ text (Nation, 2008: 14) and more than 5% of an academic text (Nation, 2001: 12). As the reader will see in Chapter 7 on inferencing, research has shown that in order for someone to be a successful inferencer 95–98% of a text’s words have to be known to them. This means that L2 learners cannot avoid lowfrequency vocabulary altogether but instead have to take on the challenging task of starting to learn at least some of the over 100,000 highly infrequent words that exist in the English language.
44 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
The parallel tests used for the present chapter
Both tests on advanced vocabulary (one in English and one in Swedish) included 100 items that were tested receptively, and that were mainly chosen due to their infrequency and/or possible difficulties in their pronunciation and/or spelling. On the L2 test, the students were asked to translate the contextualised word in bold into Swedish, receiving one point for each correct translation, while on the L1 test, the subjects were asked to explain the items in their own words (in English or Swedish). Whereas the English words were, on a random basis, picked from a list of more infrequent words that the present author had put together while reading and watching TV, the Swedish test items were selected, also randomly, from Hebbe (1999) and Hebbe and Ek (2005), both of which focus on more infrequently used vocabulary in the Swedish language. As in the two previous tests described in the present chapter, the items were presented to the informants according to frequency, starting with the item with the highest frequency. Both the frequency of the words and the context in which the words were presented were taken from the BNC and Språkbanken respectively. Examples of items tested on the English task are given in (10) and (11). (10)
Snow looks less innocuous but is even worse than rain because it blots out visibility more.
(11)
The bed, with its cream lace counterpane, was huge and comfortable, and the antique wardrobe, original fireplace and generous sofas piled high with cushions immediately made us feel at home.
Also, as before, the students were asked to evaluate the degree to which they believed they knew the test items. This is exemplified in (12). (12)
Parliament has recognised the unsatisfactory results of this law and has been willing to abrogate or modify it.
□ I don’t know this word. □ I’m guessing the word’s meaning from the context of the sentence. □ I recognise this word, but I don’t know what it means. □ I recognise this word and I’m guessing its meaning from the context of the sentence. □ I recognise this word and I think I know what it means. □ I am sure I know what the word means. Again, here the students were able to indicate not only to what extent they knew the word but also to what degree the context in which it was offered helped them arrive at its meaning.
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 45
The informants were also asked to evaluate the test as a whole by describing it as either ‘very easy’, ‘easy’, ‘average’, ‘difficult’, or ‘very difficult’. Finally, to be able to make a fair comparison between the students’ L1 and L2 results, the total frequencies of the English items and the Swedish ones were calculated, the latter ones being marginally more frequent (≈0.007%) than the former ones (≈0.005%). From a frequency perspective, the L1 test may thus be considered slightly less difficult than the L2 one. The L1–L2 difference in frequency was not confirmed statistically, which means that comparisons may be made between the parallel tests. Still, the difference will be taken into consideration when discussing the results. Results and discussion
Table 2.9 offers the participants’ results. What is perhaps most surprising is the low score achieved by the native speaker (40) as compared to the mean score achieved by the L2 learners (30.00). However, as the reader will be able to see in Table 2.13, which provides information on the informants’ knowledge of each specific test item, the overwhelming majority of the correct translations offered by the Swedish learners may in fact be attributed to successful inferencing. This will be explored further in Chapter 7. As expected, the learners also did better in their first language than their second, which was confirmed statistically at a significance level of 5%, though it needs to be remembered that the Swedish test items were in total more frequent (≈0.007% of Språkbanken) than the English ones (≈0.005% of the BNC), which may have affected the results in favour of the L1 test. It is also worth noticing here that, in contrast to many of the translations offered on the English test, more of the L1 correct answers were based on actual knowledge rather than the subjects’ inferencing skills (see Tables 2.10 and 2.11). The students’ self-evaluation of the difficulty level of these tests largely reflects these results, with the learners indicating that the English task was substantially more difficult than the corresponding Swedish one. Table 2.9 The students’ results on the tests on advanced vocabulary used in the present study Students
N
Correctness rate
Mean
SD
Standardised scores
Native speaker
1
40.00% (=40/100)
–
–
–
Swedish students – English test
15
30.00% (=450/1500)
30.00
12.93
Highest: 1.16 Lowest: −1.93
Swedish students – Swedish test
15
50.40% (=756/1500)
50.40
16.55
Highest: 1.79 Lowest: −1.78
46 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.10 The students’ results on and evaluation of advanced vocabulary on the English test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
English word given in context (correct answers)
KN
TKR
RC
91
scaredy-cat (15)
1
1
1
98
deep-six (15)
60
gobsmacked (14)
3
7
100
putrefy (13)
2
6
77
pummel (13)
1
2
83
obfuscate (13)
87
lacerate (12)
82
inoculate (11)
10
reiterate (11)
71
festoon (11)
85
man Friday (11)
96
mendacity (11)
15
hiatus (10)
2
75
gatecrash (10)
1
23
decrepit (10)
40
demur (10)
19
stupendous (9)
67
abscond (9)
80
hoodwink (8)
2
62
evanescent (8)
1
73
mawkish (8)
22
auspicious (8)
1
63 74
NKR
NKC
NK
10
1 (1)
12
1 (1)
2
2
1
1
2 (1)
2 (1)
8
3 (1)
10
4 (2)
6
3
9 (1)
4 (2)
10 (3)
3 (2)
9 (1)
4 (2)
13 (2)
1
10 (1)
3 (1)
5 (2)
5 (2)
1
1
2
1 (1)
1 1 1 1 (1)
2 1
1 (1)
6 (1)
5 (1)
8 (2)
5 (2)
9 (1)
5 (1)
5 (2)
4 (2)
7 (1)
6 (1)
7 (3)
4 (1)
4 (1)
3
4 (1)
2
7 (1)
5 (1)
10 (4)
4 (2)
guffaw (8)
8 (1)
6
abrogate (8)
9 (2)
5
38
compunction (8)
8 (2)
6 (1)
21
intimation (8)
8 (2)
5 (1)
34
forbearance (7)
84
sarnie (6)
29
baleful (6)
64
castigate (6)
1
3
2 (1)
1
1 (1) 1 1
7
6
6 (2)
7
1
10 (6)
3 (1)
1
10 (5)
3
1 1
(Continued)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 47
Table 2.10 (Continued) No.
English word given in context (correct answers)
66 13
KN
TKR
RC
NKR
NKC
NK
disavow (6)
7 (3)
7 (1)
garish (6)
8 (4)
6 (1)
45
miffed (6)
8 (4)
5
50
slovenly (6)
6 (3)
8 (2)
65
serendipity (5)
1 (1)
3 (1)
2
3 (1)
6 (1)
27
ostentatious (5)
1 (1)
1
2
6 (3)
5 (1)
12
paraphernalia (5)
1
1
6 (4)
6 (1)
61
lassitude (5)
1
6 (3)
8 (1)
92
cast(s) aspersions on (5)
9 (4)
3
70
proclivity (5)
8 (4)
6
44
extradite (5)
11 (7)
3
24
emaciated (4)
1
6 (3)
5
57
importunate (4)
2
7 (5)
5 (1)
52
counterpane (4)
1
9 (5)
4
16
fastidious (4)
9 (7)
3 (2)
97
milksop (4)
11 (7)
4
25
extricate (4)
10 (7)
4
6
mitigate (4)
7 (5)
7 (1)
26
churlish (4)
7 (5)
7 (2)
99
nincompoop (3)
4
4 (4)
4
88
oxymoron (3)
1
7 (6)
5
5
impromptu (3)
8 (6)
3
11
ramshackle (3)
1
9 (8)
3
28
petulant (3)
1
1
7 (5)
6
7
impervious (3)
1
1 (1)
6 (5)
7
17
placate (3)
1
7 (5)
5
1
onerous (3)
6 (5)
9 (2)
58
lascivious (3)
8 (7)
5 (1)
95
contumely (3)
93
inebriate (adj.) (2)
51
impecunious (2)
89
sycophant (2)
35
refractory (2)
59
salubrious (2)
8
stalwart (2)
18
reticence (2)
2
1
1 (1)
1
1
1 (1)
1 (1)
1
2 1 (1)
1
2 (1) 2 (1)
1
2
4 (4)
8 (1)
1 (1)
4 (4)
7
2 (1)
5 (4)
8
2
9 (7)
4
2
7 (6)
5 (1)
6 (4)
9
8 (6)
6
8 (6)
6
1 (1)
(Continued)
48 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.10 (Continued) No.
English word given in context (correct answers)
49
punctilious (2)
41
indomitable (2)
68
expunge (2)
78
perspicacity (1)
46
debauchery (1)
33
feckless (1)
2
hapless (1)
32
demure (1)
56
rigmarole (1)
3
innocuous (1)
39
doleful (1)
30
nebulous (1)
81
rebarbative (1)
48
plimsoll(s) (1)
37
maudlin (1)
43
facetious (0)
53
germane to (0)
47
chimera (0)
54
cantankerous (0)
55
obsequious (0)
69
deprecating (0)
20
desultory (0)
31
temerity (0)
14
nimble (0)
42
fractious (0)
36
lugubrious (0)
86
uxorious (0)
4
pernicious (0)
79
pernickety (0)
72
mellifluous (0)
76
infraction (0)
90
equivocate (0)
9
raucous (0)
94
chichi (0)
KN
1 (1)
TKR
RC
NKR
NKC
NK
10 (8)
4
1 (1)
5 (4)
8 (1)
1 (1)
7 (7)
6 (1)
1
5 (5)
9
1
4 (4)
8
2
7 (6)
6
1
7 (6)
7
1
7 (6)
7
1
6 (5)
7
6 (5)
7
9 (8)
4
4 (4)
10
5 (5)
9
7 (7)
7 (1)
1 (1) 1 (1) 1 (1)
1 (1)
2 (2)
10 (10)
5 (1)
2
6 (6)
7
2
7 (7)
5
1
1 (1)
10
1
5 (5)
9
1
5 (5)
9
1
6 (6)
7
1
7 (7)
6
1
9 (9)
5
1
6 (6)
3
3 (3)
11
5 (5)
9
6 (6)
9
1 (1)
6 (6)
8
1 (1)
5 (5)
8
9 (9)
6
6 (6)
6
8 (8)
6
8 (8)
5
1 (1)
1 (1)
1 (1)
3 (3)
1 (1)
2 (2)
1 (1)
10 (10)
3
Total
16 (3)
48 (10)
1 (1) 45 (23)
54 (5)
708 (429)
556 (51)
Correctness rate
81.2%
79.2%
48.9%
90.8%
39.4%
90.8%
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 49
Table 2.11 The students’ results on and evaluation of advanced vocabulary on the Swedish test, with test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
NKC
NK
24
pseudonym (14)
(=pseudonym)
8
2
2
1 (1)
1
1
16
destruktiv (14)
(=destructive)
7
6
1
1
21
marinera (14)
(=marinate)
7
4
35
omen (14)
(=omen)
6
8 (1)
31
utopisk (14)
(=utopian)
4
7
99
antonym (14)
(=antonym)
1
4
64
panera(r) (13)
(=coat (food))
7 (1)
4
4 (1)
88
flambera (13)
(=flambé(e))
6 (1)
4
4
44
dupera (13)
(=dupe)
5
3
2
71
insinuera (13)
(=insinuate)
5 (1)
5
4 (1)
54
deportera (13)
(=deport)
3
6
5 (1)
4
monolog (12)
(=monologue)
7 (1)
7 (1)
6
ultimatum (12)
(=ultimatum)
7 (1)
5 (1)
3 (1)
49
apatisk (12)
(=apathetic)
6
3
4 (1)
15
symbios (12)
(=symbiosis)
5
5
40
hegemoni (12)
(=hegemony)
3
7 (1)
3
22
epitet (12)
(=epithet)
3
5 (1)
1
4
2
1
resolution (12)
(=decision, resolution)
1
6 (1)
2
1 (1)
3
2
27
virtuos (12)
(=virtuoso)
1
5
1
1 (1)
5 (2)
20
autentisk (11)
(=authentic)
7
7 (3)
11
dementera (11)
(=deny)
6 (2)
6 (1)
2 (1)
39
svinn (11)
(=loss)
5 (1)
5 (2)
2
29
gytter (11)
(=muddle)
4
3
1
67
eufemism (11)
(=euphemism)
4
1
61
krösus (11)
(=Croesus)
4 (1)
5
adekvat (11)
(=adequate, exact)
42
sovra (11)
19
pragmatisk (11)
68
3 3
1 7
3
2 1
2 2 (1)
1
1
1 (1)
2
1 1
1 2
1
2
3
1
5 (1)
3
4 (1)
4
2 (1)
1
2 (1)
5
4
2 (2)
1
(=cull, eliminate)
1
5 (1)
1
5 (1)
2
(=pragmatic)
1
1
5
3
2
monogami (10)
(=monogamy)
7 (1)
4
1 (1)
1 (1)
1
25
kliché (10)
(=cliché)
7 (1)
5 (2)
1
2
47
hädelse (10)
(=blasphemy)
5
2
4 (1)
1
1 (1)
2
3 (Continued)
50 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.11 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
NKC
NK
13
karismatisk (10)
(=charismatic)
5 (1)
5 (1)
3 (1)
2
80
saktfärdig (10)
(=slow)
5 (1)
3
1
72
kufisk (10)
(=odd)
4
2
6 (2)
1
2
3
76
marodör (10)
(=marauder)
3
4
4 (1)
1
2 (2)
52
vertikal (10)
(=vertical)
3
3
5 (3)
1 (1)
12
karsk (10)
(=cocky)
55
sinnrik (10)
(=ingenious)
3
3 (1)
2
3
4
2
4
3 (1)
1
2
86
konnässör (10)
3
(=connoisseur)
2
2
1
1 (1)
6 (2)
23
3
placebo (9)
(=placebo)
6
4 (1)
2
alter ego (9)
(=alter ego)
5 (2)
3
4 (1)
2
1 (1)
28
kognitiv (9)
(=cognitive)
2
4 (1)
3 (1)
2
1
36
fadäs (9)
(=faux pas)
3
5 (2)
3 (1)
32
selektiv (9)
(=selective)
4 (1)
5 (2)
2
60
bedarra (9)
(=die down)
2
3 (1)
4
14
revidera (9)
(=revise)
2
2 (1)
4 (1)
93
blanchera (9)
(=blanch (food))
18
kontext (8)
(=context)
4
4 (1)
9
metafor (8)
(=metaphor)
4 (1)
38
prekär (8)
(=precarious)
45
galghumor (8)
(=gallows humour)
59
eskapism (8)
65 97
3 1 2
1
4 1 3 1 (1)
2
5 (4)
1
1
4 (1)
2
2 (1)
7 (3)
1
3 (2)
2
1 (1)
1
3 (1)
4 (2)
3 (1)
3
1
1
1(1)
5 (3)
4
3 (1)
4
2
2
(=escapism)
1
5
1
1
imbecill (8)
(=imbecile)
3 (2)
2
8 (3)
konkludera (8)
(=conclude)
2
8 (3)
69
numismatiker (8)
(=numismatist)
3
veto (7)
(=veto)
4 (2)
34
torso (7)
(=torso)
48
vagabond (7)
(=vagabond)
33
armod (7)
77
4
4 3 (2)
4 2
3 (2)
1
6 (1)
6 (1)
1
1
5 (1)
3 (2)
3
7 (5)
4
3 (2)
1
4 (2)
5 (2)
5 (3)
(=poverty)
3 (1)
3
2
1
1 (1)
5
mellanhavande (7)
(=argument)
2
2
1
1
2
7
53
aber (7)
(=snag)
2
1
2
2
1
6
95
malign (7)
(=malignant)
1
2 (1)
2
2 (1)
5 (3)
3
89
tordön (7)
(=thunder)
1
1
1
1 (1)
5 (2)
6
82
desavouera (7)
(=disavow)
1
2
8 (2)
3
91
palindrom (6)
(=palindrome)
3
2
1
3
1 (1)
5
81
asket (6)
(=ascetic)
3
2
1
5 (5)
3
98
alliteration (6)
(=alliteration)
3 (1)
2
4 (2)
2
17
deponera (6)
(=deposit)
3 (2)
2
5 (2)
1
1 (1)
4 1 (1)
2
(Continued)
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 51
Table 2.11 (Continued) No.
Swedish word given in context (correct answers)
KN
TKR
RC
NKR
10
cynisk (6)
(=cynical)
3 (2)
3 (2)
6 (2)
2
58
diskrepans (6)
(=discrepancy)
1
2 (1)
1
5 (2)
4
37
pendang (5)
(=counterpart)
1 (1)
1
2 (1)
1
5 (3)
4
90
grassera (5)
(=run rampant)
1 (1)
2
2 (1)
3 (1)
6
7
subventionera (4)
(=subsidise)
3
1 (1)
5 (4)
1
2 (2)
3
30
tangera (4)
(=equal, touch (record))
2 (2)
7 (3)
1 (1)
3 (3)
1
75
tranchera (4)
(=carve)
1 (1)
2 (1)
3 (1)
3 (2)
5
79
ornitologi (3)
(=ornithology)
7 (4)
1 (1)
1
3 (3)
3
87
pochera (3)
(=poach (food))
4 (1)
4 (4)
1
2 (2)
73
månglare (3)
(=hawker)
2
1
2
1 (1)
8
51
sanktionera (3)
(=sanction)
2 (1)
3 (3)
3 (2)
1 (1)
4 (4)
2
74
rudimentär (3)
(=rudimentary)
1 (1)
3 (1)
1 (1)
5 (4)
5
4 (4)
NKC
NK 1
100 revalvera (3)
(=revaluate)
1
3 (2)
1
2 (1)
5
46
kongenial (3)
(=congenial)
1
2 (2)
1
2 (1)
7
92
petimäter (3)
(=nitpicker)
1
2
12
63
etymologi (3)
(=etymology)
3
3 (1)
6
96
fåvitsk (3)
(=foolish)
1
3 (1)
10
41
pamflett (2)
(=libellous pamphlet)
2 (1)
56
lakonisk (2)
(=laconic)
1
26
kommuniké (2)
(=communiqué)
62
obstinat (2)
(=obstinate)
85
harmsen (1)
(=indignant, vexed)
57
eklektisk (1)
(=eclectic)
83
efemär (1)
(=ephemeral)
8
avbräck (1)
(=setback)
1 (1)
78
opportun (1)
(=opportune)
1 (1)
66
esoterisk (1)
(=esoteric)
84
förmäten (1)
(=presumptuous)
43
subversiv (0)
(=subversive)
70
balustrad (0)
(=balustrade)
94
guckusko(n) (0)
(=lady’s slipper (flower))
50
stringent (0)
(=logical and to the point)
1 (1) 2 (2)
2 (1)
1
7
1 (1)
4
5 (4)
4
2
1 (1)
2
4 (4)
6
2 (2)
2 (2)
6 (4)
2
2 (2)
1
3 (2)
4 (4)
2 (2)
5 (5)
1
6 (5)
3
1 (1)
4
1
1
2 (2)
11
4 (3)
1
4 (4)
1
4
1
9
1 (1)
3
5 (4)
6
3 (3)
1 (1)
2
6 (6)
2
3
2 (2)
9
1 (1)
2 (2)
2
1 (1)
9
4 (4)
11
5 (5)
4
4 (4)
1 (1)
1 (1)
3 (3)
1 (1)
Total
289 (57)
302 (67)
232 (86)
108 (15)
226 (123)
296 (1)
Correctness rate
80.3%
77.8%
62.9%
86.1%
45.6%
99.7%
52 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
In Tables 2.10 and 2.11, the students’ results for each of the test items are presented, with Table 2.10 dealing with the English test items and Table 2.11 with those on the Swedish test. As before, the words are listed in order of accuracy, starting with the item that most of the learners were able to translate successfully. (Again, if more than one item received the same score, the word for which the informants were most accurate in their evaluation is presented first.) As an example, 13 of the 15 students were able to offer the correct translation of the word obfuscate (test item 83). However, the vast majority (10 students) were forced to use the context in order to do so (column NKC, highlighted in grey since this involved the majority of the students). Another four informants indicated that they did not know the word, two of whom actually offered a correct translation (column NK). Testifying to the difficulty level of the L2 test, it can be observed that as many as 41 of the 100 test items were not translated correctly by a single one of the testees. This can be contrasted with the results on the Swedish test, where only four of the test items received no correct answers. Furthermore, for the English test items, a totally different picture is painted than the scattered ones seen with vocabulary from upper secondary school level and specialised uses. Here the vast majority of the meanings were either indicated not to be familiar but guessed based on the context given (column NKC), or not known nor guessed (column NK). The use of context to infer meaning may thus have obfuscated any other trends that may have been discernible without such support. Moreover, in contrast to many of the test parts on vocabulary taught at upper secondary school level and specialised uses, no frequency effect can be detected here on the L2 test. This is not surprising since the frequencies of all the English words included are extremely low, the most frequent item occurring only 240 times in the BNC and the least frequent occurring only once. Considering their extreme infrequency, it is instead rather due to chance whether a learner will have come across one of these items and actually acquired it. However, as mentioned above, a great many of the meanings of these words were inferred from the context in which they were offered, which may have affected the role of frequency in unpredictable ways. On the Swedish test, on the other hand, if the number of points accumulated for the first 50 items (the most frequent ones) is compared with the number of points for the remaining test items (the least frequent ones), there is a clear frequency effect. The reason for this difference may be that the frequency span was greater on the Swedish test than the English one, with the most frequent L1 item occurring 541 times in Språkbanken and the least frequent one not occurring at all. It is worth pointing out here that if a comparison is made between the students’ results on vocabulary taught at upper secondary school level and advanced vocabulary, rather than contrasting results within the tests, frequency indeed emerges as an important facilitator.
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 53
2.5 The Students’ Vocabulary Size on the Three Parallel Tests Used for the Present Chapter
In Tables 2.12 (the English words) and 2.13 (the Swedish words), the students’ individual results on vocabulary from upper secondary school level, specialised uses, and advanced vocabulary are put together. Here L1–L2 correlations can be detected for quite a few informants, with Students 2, 9 and 11 (indicated in dark grey) accumulating among the top five scores in both languages. Their results on the English tests – 174, 169 and 150 points respectively – can be compared to the native speaker’s results, which amounted to 216 points. Whether these L2 learners’ stellar results are due to already acquired knowledge or good inferencing skills will be investigated in Chapter 7. A weaker correlation can also be observed with Students 3, 5, 8 and 14, who all did comparatively well in both languages, the former two especially so in their L2, while the latter two did better in their L1. The two tables also show that there were four learners – Students 1, 4, 10 and 13 (colour-coded in light grey) – that did poorly in both languages, accumulating among the bottom five scores in both languages and achieving only 52% or less of that which the best student achieved in her L2, and 62% or less of what the best student achieved in her L1. A weaker link can also be observed with Student 6, who also accumulated comparatively low scores in both languages, especially so in her L2. (For a more detailed descriptions of all the informants, the reader is referred to Section 1.2.) Table 2.12 T he students’ total individual scores on the three tests on English vocabulary included in the present chapter Individual scores on the English test (300 items) Student 2
174
Student 11
169
Student 5
166
Student 3
162
Student 9
150
Student 15
145
Student 14
141
Student 8
139
Student 7
115
Student 12
98
Student 13
91
Student 1
89
Student 10
75
Student 6
69
Student 4
57
54 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 2.13 T he students’ total individual scores on the three tests on Swedish vocabulary included in the present chapter Individual scores on the Swedish test (300 items) Student 11
263
Student 12
251
Student 9
233
Student 8
222
Student 14
222
Student 2
219
Student 5
218
Student 3
207
Student 7
196
Student 6
191
Student 15
186
Student 13
163
Student 4
153
Student 1
148
Student 10
114
Finally, it is interesting to note that, although the results are generally lower on the English tests, the span is almost the same in the two languages, the highest and lowest scores on the English test being 174 and 57 respectively, resulting in a difference of 117 points, and the corresponding scores on the Swedish test being 263 and 114 respectively, resulting in a difference of 122 points. (The fact that the total frequency of the 300 English test items (constituting ≈0.111% of the BNC) was found to be higher than that of the 300 Swedish test items (making up ≈0.076% of Språkbanken) gives additional legitimacy to these differences.) In terms of vocabulary size, this implies that the learners’ L1 and L2 developmental paths follow the same trajectory but that the students lag behind in their second language, to varying degrees, due to the late start. 2.6 Conclusions and Pedagogical Implications
Taking into account that native speakers have a considerable head start in the acquisition of vocabulary, since when students officially embark on their L2 journey (around the age of nine in Sweden) L1 speakers will already have mastered thousands of word families, it is not very surprising that even university students know fewer words than an L1 learner, particularly in areas such as specialised uses and advanced vocabulary. The fact that frequency crystallises as an important facilitator on the majority of the test parts in the present chapter as well as between tests
Vocabulary Taught at Upper Secondary School Level, Specialised Uses of Vocabulary 55
should therefore be taken into consideration when putting together syllabi for the teaching of L2 vocabulary (as well as vocabulary in an L1). Regrettably, in Sweden this does not appear to be the case, where instead the selection of texts is based on content, disregarding almost completely the commonality of words as an important facilitator, not only at the point of a first selection but also in the subsequent stages of learning (Nordlund & Norberg, 2020). The implementation of frequency might be especially important for productive vocabulary knowledge as while the university students in the present study did not appear to experience any knowledge gap between active and passive mastery of relatively high- frequency words in their first language, there might be such a gap in their L2. In order to close this possible gap in the learners’ overall L2 developmental trajectory, there may be a need to incorporate more writing and speaking exercises in a variety of topics and text types. Moreover, as the findings indicate that there are L1–L2 correlations for quite a few learners, with informants doing either consistently well or poorly in both languages, it also seems important to begin by exploring the L1 developmental level of each learner and, based on that, putting together L2 exercises customised to each individual student. 2.7 The Chapter That Follows
In the next chapter, the 15 university students’ L1 and L2 mastery of affixation rules and word families will be investigated extensively. This means that Chapter 3, in contrast to the present chapter, will no longer be focusing only on the learners’ vocabulary breadth but will begin to dig more thoroughly into their mental lexicon, exploring the extent of their vocabulary depth too. Notes (1) Even though precautions were taken by giving the initial letter of the word sought in cases where the present author could think of more than one possible answer, some students were inventive and came up with acceptable answers that the present author had not predicted. These correct alternatives all received points and are included in Tables 2.4 (English test items) and 2.5 (Swedish test items). (2) This means to hire according to the law of employment security. The meaning is supplied here for spatial reasons.
3 Word Formation: With a Focus on Derivational Suffixation
3.1 Introduction
In an episode of the American sitcom Friends, one of the main characters, Ross Geller, goes to a tanning salon. Since it is his first time, the manager explains to him how it all works: he is to go into the tanning booth and face the spray jets, and when the jets starts to spray he is to count to five, turn around, and count to five again while the jets spray his back. However, when Ross is in the booth he does not have time to turn around since he is counting one Mississippi, two Mississippi, three Mississippi, and so on, meaning he gets sprayed twice on the same side. This is repeated a few times until he finally leaves the booth to talk to the manager. The manager questions if Ross really only counted to five. Ross then stops talking, thinks for a few seconds, stares at the manager and shouts out of sheer desperation, ‘Mississippilessly?!’1 This is just one example of the endless possibilities when it comes to creating new words with the help of affixes. However, though it may be quite easy to create new morphologically complex words, as exemplified above, it may be equally difficult to remember the correct forms of already existing derivatives, at least when it comes to words consisting of comparatively infrequent stems and/or affixes. In one episode of the American sitcom Everybody Loves Raymond, for example, Ray, the main character, is dying to tell his wife Debra about what happened to him at the baseball game earlier that day. He paints a vivid picture of how the baseball almost hit his head and thus nearly killed him. However, his wife is very flippant about the whole thing, taunting him and letting fly several remarks indicating that she thinks he is exaggerating and being very childish. Ray is mortified by this and asks her, in a disappointed tone of voice, not to be so *mock…ula…tory, hesitating while forming the word, thus showing very clearly that he realises he is not using the correct form. There is also evidence in linguistic research that shows how difficult it is to master the correct forms of derivatives. In a study by Schmitt and Zimmerman (2002; 56
Word Formation: With a Focus on Derivational Suffixation 57
to be discussed in more detail in the theoretical background) dealing with advanced learners’ production of suffixed words, many incorrect forms were produced by the L2 learners. Examples are *releasement (instead of release), *minimizement (instead of minimization), *persistment (instead of persistence) and *survivation (instead of survival) (2002: 147). Moreover, experiencing difficulty in gaining mastery of derived forms is not only limited to non-native speakers (as exemplified by the incorrect derivative *mockulatory above). A significant amount of research shows that affixation is indeed no mean task even for L1 speakers. In Schmitt and Zimmerman’s (2002: 160) study, for example, where native subjects were included as a point of reference, it is concluded that ‘the performance of the native speakers indicates a high but less than complete productive knowledge of derivational morphology’. In fact, gaining derivative knowledge appears to be a slow, incremental process for native speakers of English, starting in elementary school (Carlisle, 2000), continuing through high school (Nagy et al., 1993; Tyler & Nagy, 1989), and, with more infrequent words, probably involving a lifelong learning situation (Tyler & Nagy, 1990). It also appears to be a universally difficult area of the lexicon to master. In a study on Dutch L1 speakers’ knowledge of word formation, Smedts (1988) found that the seven-year-old subjects only knew 14% of a number of different word formation rules, the 13-year-old subjects knew 51% of the rules, and, finally, even the 17-year-old subjects had only mastered 66% of all the rules tested. If mastery of word formation is such a slow, incremental process even for native speakers, is it worthwhile spending time as a second language learner trying to develop affixation skills? In Nation (2001: 264), there are three types of studies that point in the direction that it may indeed be beneficial to L2 learners to gain mastery of word formation rules. Firstly, there are quite a few investigations that have focused on exploring the sources of the English vocabulary. These studies show, among other things, that a large number of words are derived from a limited number of roots. Investigating the 7476 most frequent words in the Lancaster-Oslo/ Bergen (LOB) Corpus, for example, Bird (1987, 1990) was able to show that 97% of these words could be derived from 2000 different roots. Focusing on the affixes forming these words thus seems to be time well spent. Secondly, there are studies that have investigated the proportion of affixed words in different corpora. In a study by Cunningham (1998), it was shown that affixed forms (derivational and inflectional) outnumbered stems four to one. In another study by Nagy and Anderson (1984), the ultimate aim was to find out how many word families can be found in all printed school English. They were able to show that as much as 12.8% of the different word-family types contained derivational affixes (21.9% involved inflected forms), again indicating that derivatives make up a comparatively large part of the English lexicon. Lastly, there are a number of investigations that focus on the frequency of specific affixes. White et al.
58 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(1989), for instance, were able to show that 60% of all the words containing the prefixes un-, re-, in- and dis- could be understood once the learners knew the most common meaning of the base word. If the derived words were contextualised and knowledge of some of the less common meanings of the prefixes was acquired, as many as 80% of the derivatives could be understood. Other investigations focusing on the frequency of affixes (e.g. Becker et al., 1980; Harwood & Wright, 1956) generally show that a small number of affixes are indeed very frequent and make up a very large percentage of all affix use. Based on the above, there appears to be a great deal of evidence suggesting it may be useful to develop affixation skills when learning English as a second language. As a matter of fact, studies have shown that this is one of the three most important skills to possess for a learner aiming to enlarge their L2 vocabulary, explicit teaching and learning vocabulary from context being the other two (Nation, 2001: 263). This is of course especially the case with advanced learners who have already developed a comparatively rich vocabulary (Schmitt & McCarthy, 1997). In addition, the present author decided to focus on derivational suffixation, which is normally considered to be the most difficult aspect of word formation (Nagy et al., 1993: 156). This is, for example, mirrored in the fact that these affixes are generally acquired later than other types of affixes, by L1 as well as L2 learners (Berko, 1958). Since the present study focuses on university students, it seems justified to investigate the informants’ mastery of derivational suffixation as one important part of their entire knowledge of the English vocabulary. 3.2 Theoretical Background and Previous Research
In the present section, two issues that are still at the forefront of research on affixation will be addressed. In the first subsection, the discussion will be concerned with whether, in a learner’s mental lexicon, the stem and the attached affix are stored separately or whether learners see derivatives as unseparated wholes. In the second subsection, the discussion will explore, theoretically as well as practically, whether a structured approach to affixes may help learners in their endeavours to learn more words. Do learners decompose morphologically complex words or not?
For quite some time researchers have discussed whether the term word family may only be used as a linguistic term for a set of words that all have the same base (e.g. persist, persists, persisted, persisting, persistence, persistency, persistent, persistently) or if the concept paints a picture of how words are stored in a learner’s mental lexicon. This issue is addressed in Nagy et al. (1989). According to the article, there are several possible approaches to how morphologically complex words may be stored. One
Word Formation: With a Focus on Derivational Suffixation 59
possibility is that all the members of a specific word family can be found under the same lexical entry; that is, that, for instance, believe, believable, unbelievable, etc. can all be found under the stem believe. It could also be argued that all the members of a word family are listed under totally separate entries. Between these two extremes, there are yet other possibilities, one being that all the members of a word family have separate entries but that these entries are linked in some way, so that if one of the members is encountered the links between the family members will automatically heighten the learner’s awareness of the other members of that same word family. It is also suggested that it is likely that only words with regular inflections and those affixes that are semantically transparent can be found under the same entry as the stem, whereas those words with inflections that are irregular and affixes whose meanings are semantically blurred have separate entries. That is, from a student perspective, a distinction is made between those forms that are easy to learn and those that, for various reasons, are more difficult to learn (e.g. Caramazza et al., 1988). The ease or difficulty with which inflected and derivative forms are learned also appears to be related to many more aspects than regularity/ irregularity and the degree of semantic transparency (Marslen-Wilson et al., 1994). Factors that also need to be considered are, for example, the frequency of the stem, inflection, and derivational affix (Caramazza et al., 1988) and whether the spoken and written forms of the stems and affixes are easily recognisable. The age at which a derivative is acquired may also decide whether a word is decomposed or if it is approached as an unseparated whole (Gardner, 2007: 258). It seems that the earlier a word is learned, the more likely it is that it is not decomposed into parts later in life. The many factors listed here show the complexity of the issue. In order to test if words are stored in word families, Nagy et al. (1989) set up an experiment addressing two main research questions. The first question deals with whether the frequency of the stem alone affects the time it takes a person to recognise it or if this is also affected by the inflected and derivative forms of that stem. For example, is it only the frequency of the word quiet as the stem of the word family that affects the time it takes for a person to recognise that stem or do the frequencies of the word-family members quietness and quietly also play important roles? If it could be shown that the frequencies of inflected and derived forms play as great a part in recognising a stem as the frequency of the stem itself, it would indicate that the members of a specific word family can be found under the same entry. If, on the other hand, the frequencies of inflected and derived forms could be shown to have no part in the recognition of the stem, the members of that word family would probably be listed under different entries than the stem. The second research question addressed by Nagy et al., not focused on here, deals with how factors other than frequency may influence the speed of word recognition and how these may condition the effect of frequency. Examples are the word’s
60 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
part of speech, its abstractness, the number of distinct meanings the word has, and the length of the word. In addition to contributing to this much-investigated field, Nagy et al. (1989) also put forth a pedagogical reason for their study. It has been shown that one of the main differences between poor readers and good readers is that those belonging to the latter category are faster and more efficient when it comes to recognising words, a difference that becomes especially pronounced with words containing many syllables, which very often is the case with morphologically complex words. Since the parts (stem and affix) of morphologically complex words are usually more frequent than the inflected/derived forms and thus easier to understand, students would benefit from gaining morphological knowledge before reading texts that include these kinds of words. Also, considering that over 60% of the new words learners encounter when reading can be analysed morphologically, acquiring knowledge of affixation is of great importance (Nagy et al., 1989: 21). The experiment performed by Nagy et al. involved three types of word pairs, each consisting of 28 pairs in which the stems were matched for length and frequency. It is the second type of word pair, focusing on derivational family frequency, that is highly relevant to the present chapter. One example of a word pair belonging to this category is the adjective stems slow and loud. Whereas these stems are similar in frequency, the derivative forms tested differ greatly, slowly being much more common than loudly. The findings show that the accuracy with which the university undergraduates tested recognised the words was dependent on the frequency of the word pair as a whole. That is, the more frequent the word pair was, the more accurate the students were in their recognition. Also, frequency appeared to increase the speed with which the subjects recognised nouns and verbs, but not adjectives, thus indicating that adjectives form a more difficult word class. This, Nagy et al. conclude, rules out the possibility that every member of a word family has its own separate entry. The opposite position – that all the members of a word family share the same entry – also appears to be false in that the effect of the frequency of the morphologically related words, though significant, was not as strong as the effect of the frequency of the stems themselves in their experiment. Nagy et al. thus claim that there appears to be partial activation in that some members affect the accuracy and speed of recognition more than others. Regularity appears to be a factor to consider here. Bradley (1979), for example, observed that the frequency of the stem affected the reaction time for suffixed words in a positive way when the spelling and/or pronunciation of the stem involved few changes. When, on the other hand, the spelling and/or pronunciation involved major changes, the frequency of the stem had little or no measurable effect. Schmitt and Zimmerman (2002) also claim that there is psychological justification for the term ‘word family’ in that learners are able to
Word Formation: With a Focus on Derivational Suffixation 61
recognise unknown members of a word family based on a member they already know. That is, there is evidence that the organisation of words into word families helps learners’ receptive knowledge. What has not been investigated thoroughly though, according to Schmitt and Zimmerman, is learners’ productive knowledge of other members of a word family once one of the members becomes known. For example, if a learner demonstrates knowledge of the verb stimulate, can they then also automatically produce the inflected forms stimulates and stimulated as well as the derivative forms stimulation (noun) and stimulating (adjective)? The investigation performed by Schmitt and Zimmerman addresses three research questions. Firstly, they wanted to find out how many of the four word classes formed with the help of affixes (noun, verb, adjective and adverb) the learners knew productively. They also wanted to explore which of the word classes the subjects tended to know the best. Lastly, they wanted to investigate the relationship between productive derivational knowledge of a word and more general knowledge of that same word. Here the subjects were asked to give information concerning the degree to which they knew each word tested. Schmitt and Zimmerman’s (2002) study comprised non-native learners of English (both pre-university students and university students) who spoke languages that were not cognate with English. The study also included native speakers of English studying at university level. These were primarily made part of the study for comparative reasons. Sixteen words (as exemplified by stimulate below), all of which represented the most frequent word of the 16 word families tested (from 10 different frequency bands), were chosen as prompt words and distributed as evenly as possible among the four word classes tested. The derivatives sought were tested in a contextualised form, as in: Noun Verb Adjective Adverb
A massage is good . Massages can tired muscles. A massage has a effect. He massaged .
The results of the study show that being able to produce other members of a word family once one of the members has been learned is no easy task. The students in Schmitt and Zimmerman’s study managed to give 58.8% correct answers, with a difference between advanced students, who produced three to four forms, and less advanced students, who produced two to three of the four forms. Furthermore, all four forms were produced correctly in only a few cases. On the other hand, there were also only a few cases in which the subjects were not able to produce any correct forms. Finally, of the four word classes, verb derivatives were best known (67%), followed by noun derivatives (63%), adjective derivatives (54%) and, lastly, adverbs (52%), which proved to be the most difficult word
62 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
class. One reason why verbs form the easiest word class is most likely that the verb is quite often the stem of a word family and is thus the form encountered more frequently than the others. The results also showed that the subjective information given by the students concerning the degree to which they thought they knew the word in question gave a good indication of whether they would get the word right or not. This was especially true for verbs and nouns, separating the two more semantically transparent word classes from the two that are less so (adjectives and adverbs). In addition to producing incorrect solutions for which the subjects stated that they knew the derivatives, it is interesting to note that the learners also produced quite a few derivatives that they claimed were fully unknown to them. Finally, even if all the native speakers claimed they knew all the derivatives tested, they did not show complete mastery when asked to produce them. Nevertheless, they were able to offer the correct forms in 91.8% of cases, thus outperforming both non-native university and pre-university students, where the former in turn outperformed the latter group. This goes to show that learning how to form derivatives is a slow, incremental process regardless of whether you are a first or second language learner. There are a number of other studies that point in the same direction as Schmitt and Zimmerman’s (2002) investigation. For example, in Schmitt and Meara’s (1997) longitudinal study, the non-native university students tested were only able to produce 15% of all the derivatives asked for. In yet another longitudinal study, Schmitt (1998) included three postgraduate university students. Even these very advanced students who were undertaking PhD studies had clear gaps in their production of derivative forms, and when tested a second time they showed no great improvement. Finally, in Schmitt (1999) the subjects, also university students, could only produce 12 of the 180 derivatives (nouns, verbs, adjectives and adverbs) asked for. These three investigations differ in one respect from the Schmitt and Zimmerman study, however, in that the subjects were asked to provide the derivative forms out of context, which probably affected the results in a negative way. The Nagy et al. (1989) and Schmitt and Zimmerman (2002) studies discussed above both help shed light on the real issue at hand: When a learner wants access to a specific derivative, do they then take the decomposition route or the direct route? That is, do they analyse the derivative into parts or do they approach it as an unseparated whole? More and more research seems to indicate that learners start their search by doing both and that the route finally chosen is the one that takes the shortest time to gain the sought answer. There are several factors that decide which route will be the fastest one. As discussed above, one such factor is certainly the frequency of the stem in relation to the frequency of the derivative. Learners appear to have an innate knowledge of frequencies of words and can thus subconsciously decide if it is the stem or the derived form that
Word Formation: With a Focus on Derivational Suffixation 63
carries more memory strength. This means, for example, that if the derivative is more frequent than the stem it is very likely that the learner will end up using the direct route to gain lexical access to the word in question. An example of a derivative that probably works in this way is the adjective insane. This derivative is more frequent than the stem sane (i.e. insane has more memory strength than sane) and will thus not be decomposed when sought in the mental lexicon. The same can be said for the derivative security, which is more frequent than its base secure, hence rendering it likely that this derived form will be stored as a separate entry in the mental lexicon. If, on the other hand, the stem carries more memory strength than the derivative – that is, if the stem is more frequent than the derivative – it is very likely that the learner encountering such an item will take the decomposition route, dividing the word into stem and affix. An example of a word that probably works in this way is modernity, in that the stem modern is more frequent than its derivative form (Haspelmath & Sims, 2010: 123). Affixes that induce allomorphy in the base also affect which route a learner finally takes when trying to gain lexical access to a derived form. The English suffixes -ity and -ship may serve as examples of two opposites. Whereas -ity causes allomorphy to occur in the base, as in electric – electricity, involving changes in pronunciation, there is no such change in the base when -ship is added, as in ambassador – ambassadorship. This may mean that when these derivatives are sought in the mental lexicon electricity is decomposed, whereas ambassadorship may be accessed through the direct route. Put differently, since ambassadorship is more easily segmentable than electricity, a learner will most likely access ambassadorship via the direct route whereas electricity will be accessed via the decomposition route (Haspelmath & Sims, 2010: 72–74). Furthermore, an extension of the supposition that it is the relative frequencies of the stem and the derivative that decide whether a word is decomposed or not is that the affix itself is also affected by which of the two processes is eventually selected. If an affix tends to occur in words that are decomposed, the lexical entry of such an affix will be frequently accessed and such an affix will therefore carry greater memory strength. According to Haspelmath and Sims (2010: 124), this means that such an affix is likely to have high productivity; that is, it is more readily available for speakers when coining new words. From a second-language-learner perspective, it would be interesting to see if affixes with great memory strength are also the affixes with which learners have the least problems and/or make use of the most if they are uncertain of what affix to produce. (A great majority of the words that -ity occurs in (85%), exemplified above, appear to be stored as unseparated wholes; i.e. this affix has a low memory strength because its lexical entry is consistently bypassed.) Hay and Baayen (2002: 233–235) offer some examples of suffixes and their so-called parsing ratios. These are -ence (parsing ratio: 0.1), -ity
64 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(0.17), -ate (0.31), -dom (0.5), -ness (0.51), -ish (0.58), -like (0.68), -proof (0.8), and -less (0.86). The higher the parsing ratio, the more the affix is activated in the lexicon; that is, such affixes are very often seen in words that are decomposed. Thus, whereas -ence is a suffix that primarily occurs in words that are stored as unseparated wholes, -less is a suffix that mostly occurs in words that are decomposed. Hence, in theory this means that the suffix -less should be more easily acquired than the suffix -ence. Research in this area is still ongoing since no conclusive results have been achieved. In general, though, results for inflected forms appear to be more certain than results for derived forms, where again regularity seems to come into play. Whereas inflected forms, such as the -s used in the third person singular in the present tense and the past tense inflection -ed, show high regularity and can, to a high degree, be generalisable, derived forms are comparatively idiosyncratic in character. This is also the reason why inflected forms are usually acquired before derivatives by native speakers (Berko, 1958). Since the present chapter focuses on suffixation, it is hoped that it will contribute to the understanding of how students approach these less regular and less transparent types of affixes. A pedagogical approach to affixes
In order to provide teachers with a more systematic way of approaching affixes (both inflectional and derivational), Bauer and Nation attempt in a study from 1993 to divide affixes into increasing levels of difficulty. Using the LOB Corpus, there are four criteria upon which Bauer and Nation base their ‘difficulty levels’: frequency, predictability, productivity and regularity (1993: 255). In all cases, priority is given to enhancing students’ understanding of written texts; thus the four criteria are mainly based on the word-building process of the written form. Additionally, Bauer and Nation are careful to explain that their difficulty levels are not discrete or absolute but rather form a continuum. The first criterion, frequency, deals with the number of words in which a specific affix appears. Put simply, the more words an affix is seen in, the more useful it is to learn. This is then mirrored in the fact that such an affix would be placed at a low difficulty level and should thus be included early in the learning process. For example, whereas the suffix -ness is one of the most productive derivational suffixes in the English language, forming nouns such as cleverness and earliness (Bauer & Nation, 1993: 273; Ljung, 2003: 92), the suffix -al, also used to form nouns such as arrival and dis/ approval, is quite infrequent; that is, in terms of frequency it is more useful to learn the suffix -ness than the suffix -al (Bauer & Nation, 1993: 270). The second criterion is concerned with how predictable the meaning of a suffix is. The more predictable it is, the easier the suffix is to learn and the earlier it should appear in instruction. Such an affix would therefore be placed at a low difficulty level. For example, the meaning of the suffix
Word Formation: With a Focus on Derivational Suffixation 65
-ness is highly predictable in that the suffix only has two meanings, one of which (the quality, state, or condition of being X, e.g. clever and early) accounts for about 95% of its uses, while the other (X behaviour, e.g. brusque and kind) makes up the rest (Thorndike, 1941). In contrast to -ness, the meaning of the suffix -ee is considerably less predictable. The most common meaning is ‘one who is X-ed’ (e.g. employee and nominee), but this still only accounts for a comparatively small number of its uses (Nation, 2001: 271). The suffix is, for example, also used in words such as absentee and escapee, having an agentive meaning. It can also denote small size, as in bootee (Ljung, 2003: 91). This means that it would be more useful to learn the suffix -ness than the suffix -ee, at least at the earlier stages of acquisition. Productivity, the third criterion used by Bauer and Nation, has to do with the extent to which the affix in question is able to form new words. Thus, the more productive an affix is, the more important it is to acquire early in the learning process (and the lower it is placed on Bauer and Nation’s difficulty continuum). In addition to -ness, already discussed above, both -er (as in reader and runner) and -ly (as in gladly and quickly) are highly productive affixes. At the other end of the productivity scale there are suffixes, such as -th (as in breadth and depth), that do not produce any new words at all. Between these extremes there are of course varying levels of productivity (Bauer & Nation, 1993: 272, 278; Ljung, 2003: 81, 104). The fourth and final criterion focuses on regularity, both of the base to which the affix is added and of the affix itself. The more regular the spoken and written forms are, the easier they are to learn, and hence they should be dealt with at an early stage in the learning process. For example, verbs containing -mit change the final -t to -ss- before -ible, -ion, and -ive. Thus from permit one would get permissible, permission and permissive (Ljung, 2003: 83–84). Due to these changes in stem endings, the suffixes -ible, -ion and -ive can be found on the higher levels in Bauer and Nation’s difficulty continuum. A second example of a suffix causing major spelling changes is -ic. For instance, there is truncation of -y in geography when forming the word geographic and the adding of an -n- in embryo when forming the derivative embryonic. For this reason, the suffix -ic can, just like -ible, -ion and -ive, be found on the higher levels in the difficulty continuum (1993: 277). This criterion also deals with the regularity of the function of the affix in question; that is, it deals with whether a specific affix seems to attach regularly to a base of a certain word class to regularly form a specific word class or whether the affix in question is able to attach to several different word classes to produce several different word classes. The more regular an affix is in this respect, the easier it is to learn. The suffix -ess (as in heiress and tigress), for example, always attaches to nouns and always produces nouns (Bauer & Nation, 1993: 256). The suffix -ful, on the other hand, can form both nouns from nouns (e.g. hand – handful (signalling amount)) and adjectives from nouns (e.g. joy – joyful) (Ljung, 2003: 85, 99).
66 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Based on the four criteria discussed above, the continuum proposed by Bauer and Nation comprises seven levels in which the strictness of the inclusion of the criteria decreases incrementally as one approaches the higher levels. Levels 3–7 are of special interest to the present chapter since they deal with derivational suffixes. Level 1 deals with a situation in which learners are thought to be unable to recognise the existence of word families. Level 2 is concerned with inflectional suffixes. Since the present study focuses on derivational suffixation, neither the first nor the second level in Bauer and Nation’s continuum will be considered further. At Level 3, all of the four criteria are included quite strictly so that the affixes seen here are very frequent and predictable, are often seen in new derivatives, involve few irregularities in the spelling of the stems and the affixes, and tend to be regular when it comes to what word class(es) they are able to attach to and what word class(es) they are able to form. Examples of suffixes belonging to this level are -able (readable), -ly (charmingly), -er (writer), -less (clueless) and -ness (commonness). At Level 4, most criteria are still adhered to quite strictly, but emphasis has been placed on frequency and orthography rather than productivity and phonology. Examples of suffixes belonging to this level are -ity (reality), -ment (accomplishment), -ful (useful), -ation (realization) and -ous (adventurous). At the next level (Level 5), there is a decrease in the frequency of the affixes. Nevertheless, the affixes seen here are still fairly productive and regular, as exemplified by -age (leakage), -ance (clearance), -ary (revolutionary), -ence (emergence) and -ery (bakery). While the affixes included at the next level (Level 6) are almost as frequently used as those found at Level 5, they cause more orthographic and/or phonological irregularities. Examples are -ion (inclusion), -ee (employee), -ify (simplify), -ist (tobacconist) and -ive (supportive). Level 7, finally, includes those affixes that have to be taught explicitly, not only to L2 learners but to native speakers as well, that is, affixes that display a high degree of difficulty in connection with all four criteria. Examples of suffixes belonging to this level are -ar (circular), -et (packet), -ure (departure) -ate (captivate) and -some (troublesome). It should be noted that one and the same suffix, but with different meanings, may occur at more than one level. According to Bauer and Nation, the affixes belonging to Level 3 make up the majority of the word families (types and tokens) found in the LOB Corpus. It thus appears that these affixes are well worth learning before those placed at higher levels. A decrease in the number of types belonging to Level 4 and higher confirms the validity of the continuum set up by Bauer and Nation. For a more comprehensive overview of the various levels and the prefixes and suffixes included at each respective level, the reader is referred to Bauer and Nation (1993). To the present author’s knowledge, regrettably, no corresponding continuum has been put together for Swedish affixes.
Word Formation: With a Focus on Derivational Suffixation 67
One drawback with Bauer and Nation’s framework is that it does not at all address the role played by the stem (Gardner, 2007). As discussed previously, the frequency of the stem appears to be an important factor in relation to the ease or difficulty with which learners are able to produce derivatives. Other investigations point in the same direction. For example, there are studies that show that L1 students tend to focus entirely on the meaning of the stem when asked to define a derivative form (Freyd & Baron, 1982; Wysocki & Jenkins, 1987). There are two additional articles relevant to the present section, the second of which (partly mimicking the material used in the first one) makes an interesting comparison with the Affix Levels proposed by Bauer and Nation (1993). Schmitt and Meara’s pioneering empirical study from 1997 investigates students’ knowledge and development of inflectional and derivational verbal suffixes (selected to be the focus of attention due to their highly frequent nature) and word association, along with the interrelation between these. The reason for exploring a possible link is that it was considered likely that these two factors were connected, as the more affixes you know, the larger your vocabulary and the more associations you are able to provide. The results were also compared to the students’ overall vocabulary size, where both vocabulary breadth (i.e. how many words were known) and vocabulary depth (i.e. how well individual words were known) were considered, and to the students’ general language proficiency (measured by using the subjects’ scores on the Test of English as a Foreign Language (TOEFL)). The subjects in Schmitt and Meara’s study were Japanese students from three different educational levels – one high school group and two university groups (first and second year student). The high school students had studied English for approximately five years prior to the testing and the students in the two university groups had studied it for about six years. The majority of the students in all three groups were in their late teens. Furthermore, both the students’ receptive and productive skills were investigated during a one-year research period, with all three groups being tested twice, once at the beginning of their school year and once at the near end. On the test focusing on production, the students were asked to give three word associations and all possible suffixes for 20 prompt verbs. On the task focusing on reception, the students were to choose three out of four given associations and all possible suffixes out of a set of 14 provided by the researchers. The results showed that the subjects as a group improved in their knowledge of verbal suffixes during the test year, although the increase was not substantial. On the productive task, the students knew 5% more verbal suffixes by the end of the year, with a total score of 47%. However, if a comparison is made between inflectional and derivational suffixes, it appears that the informants found derivational suffixes to be considerably
68 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
more difficult to deal with than inflectional ones, offering only 15% correct answers in the former case. As Schmitt and Meara point out, this is not so strange since inflectional suffixes are rule-based whereas derivational suffixes are not to the same extent. Also, the students did slightly better on the receptive task, where they improved by 6%, ending up knowing 66% of all the verbal suffixes at the end of the test period. (No separate figures for inflectional and derivational suffixes are given here.) When it comes to the two word association tasks, 33 of the 86 learners who completed the pre- and post-tests improved in both their productive and receptive knowledge. However, whereas on average there was a statistically significant increase of three word associations on the productive test for all the subjects included, no statistically significant change could be seen on the receptive task. During the test year, the students also increased their general vocabulary knowledge, both by adding information to words they only knew partly and by gaining information about words they did not know at all at the beginning of the research period. They also generally showed improved scores on the TOEFL test. By comparing the students’ results on all four tasks – verbal s uffixation, word association, vocabulary size, and general language proficiency – Schmitt and Meara were able to detect connections between all four types of knowledge. However, clear differences in the strength of the relationships could be seen. For example, whereas the links between association knowledge (particularly the students’ productive skill), vocabulary size, and general language proficiency on the one hand and suffix knowledge and general language proficiency on the other hand were shown to be quite strong, the link between receptive derivative suffixation and vocabulary size was seen to be comparatively weak, as was the link between derivative suffixation knowledge and word association. The second study included here – Mochizuki and Aizawa (2000) – focuses on the relationship between L2 learners’ vocabulary size and their knowledge and acquisition of affixes, so its aims have strong affinities with those of the study performed by Schmitt and Meara (1997). The subjects included in the study were Japanese third-year high school students and university students (one second-year group and one third-year group), thus mimicking the nationality and distribution of the subjects involved in the Schmitt and Meara study. Two tests were administered: one vocabulary size test focusing on receptive vocabulary knowledge (developed in Mochizuki (1998)) and one affix knowledge test, constructed to be able to answer how many affixes a specific learner knew. Both tests were given at the beginning as well as at the end of term. The affixes chosen were selected from Bauer and Nation’s (1993) Affix Levels 3–6, as described in previous paragraphs. Furthermore, the affixes chosen had to have appeared in more than two words in Nation’s Vocabulary
Word Formation: With a Focus on Derivational Suffixation 69
Lists (1996). Thirteen prefixes and 16 suffixes passed both criteria. While the prefix test focused on lexical meaning, the test on suffixation dealt with the syntactic role of the suffixes, that is, which word class they form in a specific situation. Here three low-frequency words, transformed into pseudowords, were created with the help of the suffix -able (e.g. rombortable, quifable and slomitable). The subjects then had four different word classes to choose from – noun, verb, adjective or adverb – when stating which word class the suffix in question formed (adjective being the correct answer with the three pseudowords given above). The results show that the subjects’ affix knowledge increased proportionately to their knowledge of vocabulary in general. This means that, just like Schmitt and Meara (1997), Mochizuki and Aizawa (2000) were able to show a relationship between affixation knowledge and vocabulary size. They further demonstrated that the subjects followed certain routes in their affix acquisition. For example, the suffixes -ation, -ful and -ment produced more correct answers than did -ness, -ism and -able, which in turn produced more correct answers than -ish -y and -ity. Mochizuki and Aizawa regarded the accuracy order of the affixes as an acquisition order in that items that are known by more learners are generally acquired earlier than those known by fewer learners. Mochizuki and Aizawa (2000) list four factors that may be responsible for the specific order in which the affixes were acquired in their study, which did not adhere completely to the order proposed in Bauer and Nation (1993). For example, there are quite a few loan words from English to Japanese that contain prefixes, thus aiding comprehension of these specific prefixes. Also, instruction in Japan very often involves discussing the use of common affixes, making it likely that students have encountered certain more frequent affixes but not more infrequent ones. Furthermore, certain affixes and also the words to which they are attached are more infrequent than others, thus again affecting the possibility of encountering them. Lastly, Mochizuki and Aizawa point to the polysemous character of certain affixes, which may also affect students’ ability to acquire them (2000: 300–301). This may mean that the order in which affixes are acquired is not universal but culture-dependant. 3.3 A Comparison of Affixation Rules between English and Swedish and the Parallel Tests Used for the Present Chapter
In the present section, a comparison will first be made between the English affixation system and the Swedish one. Special emphasis will be given to derivational suffixation as the students involved in the present study were almost exclusively tested on that type of affixation. Secondly, the structure of the word formation tests used in the present investigation will be presented in detail.
70 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
A comparison of affixation rules between English and Swedish
In addition to making use of Nordic and Germanic affixes when forming derivatives, Swedish, just like English, includes a great number of affixes of French, Greek and Latin origin. 2 From a second-languagelearner perspective, this shared well of affixes naturally facilitates the comprehension of morphologically complex L2 words (Liljestrand, 1993: 55). Furthermore, in both languages, in the majority of cases, a change in word class occurs when a suffix is added to a stem, something that hardly ever happens in either language when a derivative is formed with the help of a prefix. Also, in both languages the meanings of prefixes are comparatively clear and consistent, whereas the meanings of suffixes are relatively vague and less consistent (Liljestrand, 1993: 54; Sager & Ormelius, 1983: 35), rendering suffixes more difficult to grasp for an L2 learner. Moreover, both English and Swedish show consistency in what stems and affixes may be combined. It appears to be the trend that an affix of a certain origin is almost always attached to a stem of the same origin (Liljestrand, 1993: 56; Ljung, 2003: 79). The suffixes -dom and -hood, which are used to create nouns in English, do not, for example, combine with classical words. The opposite applies to the suffixes -ation, -ity and -ive, which are never seen with Germanic stems (Ljung, 2003: 79). Finally, in both languages affixes vary greatly as to how productive they are. The most productive suffix of all in the English language is the adverb-forming suffix -ly. Other suffixes that are almost as productive are -ness (as in kindness) and -er (as in writer). At the other end of the scale, there are suffixes that are never used to create new words, one of which is, as mentioned before, -th, as seen in words like birth and depth (Ljung, 2003: 81). Two examples of highly productive affixes used in Swedish are the suffixes -are as in sångare (singer) and -bar as in användbar (useable) (Liljestrand, 1993: 68, 81). As in English, there are also suffixes in Swedish that are no longer used when coining new words. Examples of this type are -lek (as in tjocklek (thickness)) and -sel (as seen in blygsel (timidity)) (Liljestrand, 1993: 75). In general, it seems that affixes of Nordic/Germanic origin are more productive in Swedish than affixes that are of French, Greek and Latin origin. Based on what has been said so far and the seemingly endless list of new words being formed (e.g. mississippilessly, 3 bagelize,4 mazer5 and unappoint6 ), it might appear that new derivatives may be created somewhat haphazardly. Admittedly, ‘the set of words in a language is never quite fixed’ in that ‘[s]peakers have the capacity to create, and hearers can understand, an almost unlimited number of new words’ (Haspelmath & Sims, 2010: 71). (Also, while some of these neologisms will be heard only a few times – so-called occasionalisms or nonce formations – and then disappear again, others will become part of everyday language and eventually find their way into dictionaries.) Nevertheless, there are indeed restrictions as to how derivatives may be formed, most of which are
Word Formation: With a Focus on Derivational Suffixation 71
language-specific. These can be divided into four main categories. Morphological restrictions, forming the first category, are of great interest to the present study since these restrictions apply to derivational suffixes only. In English quite a few derivational suffixes are very particular as to what stems they are attached to in that a certain suffix only combines with stems belonging to a specific word class. For example, the noun-forming suffixes -ity and -ness can only be added to adjective stems, as in reality and kindness (Ljung, 2003: 2). The noun-forming suffix -er, on the other hand, can only be attached to verbs, as in employer, whereas the adjectiveforming suffix -ish can only be attached to noun stems, forming words like childish (Ljung, 2003: 79). (For a more detailed description of morphological restrictions in English see Ljung, 2003.) Morphological restrictions naturally exist in the Swedish language too. The noun-forming suffix -het, for instance, can only be attached to adjective stems, as in dum – dumhet (stupid – stupidity) (Liljestrand, 1993: 75). (See Liljestrand, 1993, for more examples of morphological restrictions in the Swedish language.) There are also phonological restrictions to adhere to when forming derivatives. These restrictions, which can be either quite structured in character or purely random, also appear to especially affect derivational suffixes. The English suffix -ee, for example, cannot be added to all verbs for reasons that have to do with phonetic processing (i.e. pronunciation and/or perception). Whereas it appears to be acceptable in the word pair pay – payee, it is not so in the word pair accompany – *accompanyee (Haspelmath & Sims, 2010: 118; Raffelsiefen, 1999: 246). A second example of a restriction of a phonological kind involves the suffix -ize, which only attaches to adjectives that start with a stressed syllable followed by an unstressed syllable, and where the suffix itself becomes stressed, thus achieving an alternating rhythm of stressed – unstressed – stressed syllables. This is why the verb in the word pair global – globalize is acceptable but the one in the word pair secure – *securize is unacceptable (Haspelmath & Sims, 2010: 118–119; Plag, 1999; Raffelsiefen, 1996). An example of a purely random restriction involves the suffix -en. It seems that this suffix can only be added to monosyllabic bases and form words such as blacken and tighten. Phonological restrictions are also present in the Swedish language. For example, in Swedish we say that we ser på TV (we watch TV), but the person who is involved in the activity is not called a *TV-seare (TV watcher); that is, the verb se is here not used to form the derivative as it would not be a very successful derived form, phonologically speaking. Instead, Swedish uses the verb titta (look), forming the noun TV-tittare (Liljestrand, 1993: 58). Restrictions as to how derivatives may be formed can also be semantically induced. For example, the English prefix de- can only be combined with verb stems whose action has the potential to be reversed. This is why it is completely acceptable in a word like decolonize but sounds odd in *deincinerate. This specific semantic restriction is general and applies to
72 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
all verb stems to which the prefix de- is added, but there are also other semantically related issues that are more random in character. Here socalled synonymy blocking may serve as an example. It appears that languages prefer not to have too many words denoting the same thing. This means that if a word for an action, object, or concept already exists, this word will block other words for the same action, object, etc. from being formed. For example, English has word pairs like to write – writer, statistics – statistician and bad – badly but does not have the word pairs to type – *typer (typist is the blocking word), gymnastics – *gymnastician (gymnast already exists), or good – *goodly (well is already used). It is interesting to note that the more frequent the blocking word is, the greater is its blocking strength, that is, the less likely it is that a synonym will be created (Haspelmath & Sims, 2010: 119). There are quite a few examples of synonymy blocking in Swedish too. For example, the words vis, klok, förståndig, and intelligent (wise, smart, sensible and intelligent) can all be negated using the prefix o- (un-), rendering the words ovis, oklok, oförståndig and ointelligent. The opposite to dum (stupid) is not *odum, however, blocked as it is by the already existing word klok. Lastly, pragmatic restrictions are also comparatively frequent. In German, for example, the suffix -in denoting female gender is used when forming the words for queen and lioness (Königin and Löwin) but is not used with names of lower animals; words like *Käferin (female beetle) and *Würmin (female worm) would sound inherently odd to a native speaker of German. The reason for this is pragmatic in that there is no use in everyday language for making a distinction between male and female beetles and worms (Haspelmath & Sims, 2010: 120). When forming new words, native speakers subconsciously succeed in adhering to all these intricate rules (Haspelmath & Sims, 2010: 115), and it is therefore interesting to investigate to what extent L2 learners are able to cope with these restrictions. Of special interest here are of course those of a morphological nature. Despite the individual intricacies exemplified above, both English and Swedish make use of affixes to a very similar degree. This is confirmed by Andersson (1987: 26), who categorises Swedish as a slightly less syntactic language than German, which displays a ratio of 1.92 morphemes per word (Haspelmath & Sims, 2010: 6), but slightly more syntactic than English (Andersson, 1987: 21–24), which has a ratio of 1.68 of morphemes per word (Haspelmath & Sims, 2010: 6).7 A further piece of evidence of how close the two languages are from a word formation perspective is that they both have agglutinating as well as fusional characteristics. 8 For example, both languages denote plurality by using inflectional suffixes (an agglutinating characteristic) and both languages make use of vowel change when forming irregular verb forms (e.g. sing – sang – sung (sjunga – sjöng – sjungit)) (which is a fusional characteristic) (Andersson, 1987: 26–27). This similarity in the ability to make use of affixes to form derivatives has
Word Formation: With a Focus on Derivational Suffixation 73
made it possible to construct the two tests used in the present study in almost the exact same way. The parallel tests used for the present chapter
Both tests used in the present chapter (one in English and one in Swedish) are divided into two parts.9 The first part, testing 50 word pairs (one item given; one item sought), is a context-based gap-fill exercise. To construct this part of the L2 test, two booklets comprising exercises on stems and their derivative forms were used, with Cederberg (1993) primarily focusing on upper secondary school level and Sager and Ormelius (1983) mainly focusing on university level, thus ascertaining differences in commonality. A few test items that are more infrequent were selected from an informal list of derivatives that the present author had encountered in authentic text and had, for various reasons, brought up and discussed with previous groups of students. In all cases, all word pairs were picked randomly. The Swedish test items, on the other hand, were primarily picked from Liljestrand (1993) and Möijer (1998), which also offer extensive lists of stems and their derivatives. The few remaining Swedish test items were derivatives that the present author had come across while reading various texts. These items, just like the English ones, were all randomly selected. Furthermore, the context for most of the items on both tests was taken from the books mentioned above, but in a few cases the present author, sometimes with the help of the BNC for the L2 items, constructed a context. In the majority of cases the students were obliged to add a suffix, as in (1)–(3): (1) The drawing has many colours in it.
It is very .
(2) The bus arrived late.
The estimated time of was 10.00 hrs.
(3) Don’t encroach on my rights!
Your is unforgivable.
However, they were in a few cases asked to provide the stem, removing the suffix from the word given, as in (4): (4) I like being comfortable.
I love .
In addition, the students were asked to evaluate the degree to which they believed they knew the words, both those given as well as those that were to be produced. For the given word, they were asked to indicate if they did not know the word, if they thought they knew the word, or if they were sure they knew the word. For the word they were expected to provide, they were asked to make a decision along the same lines. This is exemplified in (5).
74 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(5) The drawing has many colours in it.
It is very .
□ I don’t know the word. □ I think I know the word. □ I am sure I know the word.
□ I don’t know the word to fill in. □ I think I know the word to fill in. □ I am sure I know the word to fill in.
The second part of both tests focuses on the students’ knowledge of word families in a decontextualised form, again testing 50 items (nouns, verbs, and adjectives). While a few of the L2 word families were again picked from Cederberg (1993) and Sager and Ormelius (1983), the vast majority are items selected from the informal list of more infrequent derivatives put together by the present author, as mentioned above. This seemed advisable as the students were given the whole test at once and were thought to be able to benefit from Part A, which they were asked to start with, when doing Part B. The Swedish word families, on the other hand, were primarily taken from Hallström and Östberg (2001), which offers numerous examples of word families and their members. Again, all the items were chosen in a random fashion. As the test items were selected randomly, there is a difference between the L1 and L2 test in the number of word families with two, three, four, or five family members. Table 3.1 offers these differences. Below are three examples of word families included in the English test. As can be seen, the first example, (6), is a word family for which the students were given the verb analyse/analyze and were expected to produce the noun derivative analysis, as well as the two adjective derivatives analytic and analytical. A word family that contains only two members is illustrated in (7), where the students were supposed to give the adjective derivative boyish. The last example, (8), illustrates a word family for which the students were expected to offer two members, the noun appropriator and the verb appropriate. Table 3.1 The distribution of the number of word families with two, three, four, or five family members on the English and Swedish tests respectively No. of word members
English
Two members (one given, one to be formed)
13
2
Three members (one given, two to be formed)
8
8
Four members (one given, three to be formed)
7
8
Five members (one given, four to be formed)
–
2
28
20
Total
Swedish
Word Formation: With a Focus on Derivational Suffixation 75
Noun
Verb
Adjective
(6) (=analys)
analyse/analyze (=analysera)
(=analytisk) or (=analytisk)
(7) boy (=pojke)
–
(=pojkaktig)
(8) appropriation (=anslag, tillägnelse)
(=anslå, tillägna sig)
–
(=den som har tillägnat sig)
Also, on the English test, the students were always given a translation into Swedish of the test items that they were supposed to produce, while on the L1 test, no explanations were given of the meanings of the derivatives. They were, however, clearly shown the word class of the sought derivative. (The native speaker chose to do the test in which Swedish translations were given. However, the explanations probably did not help him a great deal as his knowledge of Swedish was very poor at the time of the test.) The reader is also reminded that the native speaker took all the tests one term before the Swedish students. After the native speaker had done the suffixation test, he complained that the second part was very difficult because he was unsure of what noun, verb, and adjective meant (even though he had just taken a grammar course in which word classes had been discussed). The present author therefore decided to explain these three word classes quite thoroughly to the Swedish students before they took their tests so as to minimise their error score due to lack of comprehension of the form of the word in question. Consequently there is a discrepancy in the instructions given to the Swedish learners as compared to the ones given to the native speaker. Furthermore, as in Part A, the students were asked to evaluate their knowledge of both the given word-family member and the word-family member(s) sought. They could indicate that either they did not know the word given/what word to fill in, that they thought they knew the word given/the word to fill in, or that they were sure that they knew the word given/what word to fill in. At the end of each part of the suffixation tests, the students were also asked to give their opinion as to how difficult the test part in question was along a continuum of ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’.
76 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
In both parts of the test, the items were presented to the students in order of frequency based on the BNC and Språkbanken respectively. Moreover, since, as has been discussed in Section 3.2, there appears to be ample evidence that stems and their derivatives are often stored together in a learner’s mental lexicon, and that their respective frequencies interact when it comes to whether a derivative is decomposed or not, the sum of the frequencies of the members involved in the word pairs/ word families determined the order in which they were presented to the students.10 The scoring system used for the two word formation tests is straightforward. For each correct word produced, the students were awarded 1 point, in both Part A and Part B, so the total score available for each test was 100 points.11 As in Schmitt and Zimmerman (2002: 157), since the focus of the tests was the students’ mastery of derivatives, minor misspellings were ignored as long as the intended derivative could be discerned. Furthermore, in the word family part, any attached inflections, such as the plural -s or -ed signalling past tense, were also ignored. This is again in parity with the approach taken by Schmitt and Zimmerman. From the above description, it can be seen that the two suffixation tests are constructed in a very similar fashion. Nevertheless, in order to be able to draw fair conclusions concerning the students’ results, it is necessary to make a comparison between the total frequencies of the English and Swedish test items. The L2 test items were found to be marginally more common (constituting ≈0.309% of the BNC) than those on the L1 test (making up ≈0.302% of Språkbanken), thus, on the whole, based on frequency, making the English test somewhat easier than the Swedish one. If Part B of the English test is considered separately, though, it can be seen that these test items are less frequently used (≈0.041%) than the Swedish ones (≈0.142%) (hence making the test items in Part A more frequent on the L2 test (≈0.268%) than those on the L1 test (≈0.160%)). However, as the L1 test incorporates more word families consisting of four and five members (see Table 3.1), it may be said to exhaust the students’ knowledge of derivatives to a greater depth, thus, from this perspective, making the L1 test part more or at least equally difficult. The reader is also reminded that while the English corpus contains spoken as well as written language, Språkbanken only involves the latter. Since spoken language usually favours more frequent vocabulary, it is likely that if the English corpus had consisted of written language only there would have been more occurrences of the more uncommon items, making the difference between Part B of the two tests less pronounced. None of the L1–L2 differences in frequency above were confirmed statistically, which means that comparisons may be made between the parallel tests/test parts. Still, the differences will be taken into consideration when discussing the results.
Word Formation: With a Focus on Derivational Suffixation 77
3.4 Results and Discussion
In the following subsections, the students’ results will be approached quantitatively as well as qualitatively. The discussion will start with a presentation of the students’ mean scores and their evaluation of the difficulty level of the two tests. The following two subsections will analyse the students’ results in Parts A (word pairs, context-based and gap-filling) and B (word families, decontextualised) respectively. The final subsection will investigate the students’ knowledge at an individual level. In order to make the following discussion as reader-friendly as possible, the present author will sometimes refer to the words formed as derivatives, although, strictly speaking, some of these are stems that have been formed by truncating the related derivatives. The students’ results
Table 3.2 presents the mean scores for the suffixation tests as a whole as well as Parts A and B separately. Unsurprisingly, the informant group did considerably better on the Swedish test (mean score: 71.13) than on the English test (mean score: 38.80). This difference was confirmed statistically at a significance level of 5%, the validity of which is enhanced even further by the fact that the
Table 3.2 The students’ results on the suffixation tests used in the present study Test part(s)
Students
N
Correctness rate
Mean
SD
Standardised scores
Parts A + B (word pairs + word families)
Native speaker of English
1
35.00% (=35/100)
–
–
–
Swedish students – English test
15
38.80% (=582/1500)
38.80
10.44
Highest: 1.46 Lowest: −1.90
Swedish students – Swedish test
15
71.13% (=1067/1500)
71.13
14.73
Highest: 1.69 Lowest: −2.11
1
60.00% (=30/50)
Part A (word pairs)
Part B (word families)
Native speaker of English
-
-
Swedish students – English test
15
55.20% (=414/750)
27.60
6.29
Highest: 1.34 Lowest: −2.00
Swedish students – Swedish test
15
85.87% (=644/750)
42.93
4.86
Highest: 1.45 Lowest: −2.45
1
10.00% (=5/50)
Native speaker of English
-
-
Swedish students – English test
15
22.40% (=168/750)
11.20
5.20
Highest: 1.69 Lowest: −1.38
Swedish students – Swedish test
15
56.40% (=423/750)
28.20
10.59
Highest: 1.68 Lowest: −1.81
78 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
L2 test was, from a frequency perspective, marginally easier than the L1 test (the test items constituting ≈0.309% of the BNC versus ≈0.302% of Språkbanken). Furthermore, when the two test parts are considered individually, the same trend can be observed; that is, in Part A (Swedish test: 42.93; English test: 27.60) as well as Part B (Swedish test: 28.20; English test: 11.20) the Swedish students did better in their native language than their L2, especially in Part A. Again, both differences are confirmed statistically at a significance level of 5% (as well as the inter-language difference between the two test parts). However, while this is fully valid for Part A, where the L2 test items were more frequent (≈0.268% of the BNC) than those making up Part A of the L1 test (≈0.160% of Språkbanken), the difference on Part B needs to be approached with some caution, as the English test items here were less frequent (≈0.041%) than the Swedish ones (≈0.142%). Nor is the native speaker’s result on the first part very surprising. Even though it is comparatively low (30), he still, as expected, outperformed the Swedish students. The only anomalous result is the one achieved by the native speaker in Part B. Here he offered only five correct derivatives. There may be two reasons for this. As discussed earlier, the native speaker found it difficult to distinguish between the three word classes included in the test and quite agitatedly told the present author so after having taken the test. The Swedish students thus had the advantage of benefitting from this in that they, as a consequence of the native speaker’s protests, received instruction on differences between the word classes before taking the tests. Including tasks that involve metalinguistic knowledge, such as grammatical classification, has been proven to have a negative effect on subjects’ performance in other studies too, as, for instance, in Schmitt (1998, 1999) and Schmitt and Meara (1997), mentioned earlier in the theoretical background section. The same problem is also discussed in Alderson et al. (1997). As the reader will learn in the latter part of the results section, this is not unique to the native speaker as a few of the Swedish learners also achieved very low scores in Part B. The second reason may have to do with the native speaker’s below average aptitude for language in general. This learner achieved poor results on all the linguistically oriented courses he took for the present author. (Through personal communication with the lecturers teaching the literature courses, the present author learned that he was a low achiever in those classes too.) As discussed in the theoretical background (Section 3.2), learners appear to be subconsciously aware of restrictions – morphological, phonological, semantic and pragmatic – that exist when forming derivatives in a language. It is very likely that students characterised as having a gift for language are especially attuned to these restrictions, whereas poor language learners, such as the native speaker included in the present study, are not. The mean scores achieved by the students presented in Table 3.2 are mirrored quite accurately in their self-evaluation. Whereas none of the
Word Formation: With a Focus on Derivational Suffixation 79
Swedish students thought that Part A of the Swedish test was ‘very difficult’ or ‘difficult’, the majority felt that Part B of the same test and both parts of the English test were either ‘difficult’ or ‘very difficult’, Part B of the English test being most difficult, Part A of the English test a little less difficult, and Part B of the Swedish test the least difficult of these three parts. This can be compared to Schmitt and Zimmerman’s (2002) study, in which the testees’ subjective assessment of their achievements also tallied well with what they actually accomplished. This is also confirmed in the correctness rates for items indicated as known as well as those that were only thought to be known, as shown at the bottom of Tables 3.3 (L2 test) and 3.4 (L1 test) for Part A, and at the bottom of Tables 3.6 (L2 test) and 3.7 (L1 test) for Part B. On Part A, there is a correctness rate of 84.1% (known) and 54.0% (thought to be known) on the L2 test, which can be compared to 93.4% (known) and 77.8% (thought to be known) on the L1 test. On Part B, there is a correctness rate of 71.4% (known) and 34.7% (thought to be known) on the L2 test, which can be contrasted with 92.4% (known) and 64.0% (thought to be known) on the L1 test. Results on Part A (word pairs, context-based and gap-filling)
Figures 3.1 (English test) and 3.2 (Swedish test) present the number of correct answers for each of the 50 derivatives tested in Part A. The reader is reminded that the test items were presented to the students in order of frequency so that whereas derivative number 1 is part of the most frequent word pair, derivative number 50 belongs to the most infrequent word pair. Furthermore, the derivatives whose scores are indicated in bold in Figure 3.1 are part of A General Service List of English Words (West, 1953). As mentioned in Chapter 1, no such corresponding word list could be found for the Swedish words. In Figure 3.1, it can be seen that word-pair frequency does have an effect on the students’ results on the English test in that, with a few exceptions, there is a decrease in the number of correct answers from the most frequent word pair to the least frequent one. It is also interesting to note that the majority of the words that are a part of the first two thirds of these 50 words are also found on West’s (1953) list of 2000 head words, thus solidifying the word-pair frequency effect further. (This result can be compared to the native speaker’s achievement, which is not affected by frequency to any great extent. For the first 25 test items (A1–A25) he was able to produce 16 correct derivative forms and for test items A26–A50 he offered 14 correct answers.) On the corresponding Swedish test part, no such clear decrease in relation to word-pair frequency can be observed (Figure 3.2), despite the fact that, from a frequency perspective, the Swedish test items are slightly more difficult than those making up the English test (see Table 3.1). It is not until the very last few word pairs that a slight decrease might be detected.
80 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Total no of correct answers for each word
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
5
10
15
20
25
30
35
40
45
50 Word
Figure 3.1 The total number of correct answers for each of the 50 derivatives that were to be formed in Part A of the English suffixation test, presented in order of frequency, starting with the most frequent derivative
Total no of correct answers for each word
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
5
10
15
20
25
30
35
40
45
50 Word
Figure 3.2 The total number of correct answers for each of the 50 derivatives that were to be formed in Part A of the Swedish suffixation test, presented in order of frequency, starting with the most frequent derivative
Word Formation: With a Focus on Derivational Suffixation 81
Tables 3.3 and 3.4 present the Swedish students’ results on Part A for each word on the English and Swedish test respectively. In both tables, the words formed by the students are presented in order of accuracy so that the word that received the highest score is placed first and the word with the fewest correct answers is placed last. (When more than one Table 3.3 The students’ results on and evaluation of Part A of the English suffixation test, with the test items listed in order of the correct number of answers, starting with the item that received the highest score (KN = the word is known, TK = the word is thought to be known, NK = the word is not known). Figures in parentheses in columns KN(B), TK(B), and NK(B) indicate cases where the students were incorrect in their self-evaluation No.
Word given A
9 advise(d) 8 colour(s) 20 anxious 4 bear (born) 5 enter 3 apply (applied) 10 achieve 17 arrive(d) 6 deep 7 easily 2 early 16 clean(ed) 1 able 24 clever 26 cautiously 11 equal 25 cruel 29 envy (envied) 14 absent 23 courage 30 acquaint(ed) 38 inordinate 32 annoy(ed) 15 comfortable 12 advantage 19 broad 40 intrepid 27 admire(d) 22 complain(ing) 21 bury (buried) 36 contagious 42 abbreviate 35 encroach 43 surly 48 ruminate 28 (did)n’t approve 44 voracious 39 exacerbate(d) 13 eat 34 coward 31 conquer(ed) 33 complacent 50 asinine 46 guileful 37 tempest 18 capacity 47 chastise(d) 49 putrefy 41 arcane 45 parsimonious Total Correctness rate
Word formed B KN(A) (correct answers) advice (15) 13 colourful (15) 13 -ness (8), anxiety (7) (15) 11 birth (14) 13 entrance (11), entry (3) (14) 13 application (14) 12 achievement (13) 13 arrival (13) 13 depth (13) 12 ease (9), easiness (4) (13) 12 earliness (13) 12 cleaner(’s) (13) 11 ability (13) 10 cleverness (12) 12 -ness (2), caution (10) (12) 11 equality (11) 13 cruelty (11) 12 envious (10) 12 absence (10) 11 courageous (10) 11 acquaintance (10) 9 inordinately (10) 2 annoyance (9) 12 comfort (9) 12 advantageous (9) 11 -ness (9), breadth (0) (9) 10 -ness (2), intrepidity (7) (9) admiration (8) 13 complaint(s) (8) 12 burial (8) 11 -ness (7), contagion (0) (7) 10 abbreviation (7) 4 encroachment (7) 3 surliness (7) 1 ruminative (5), ruminant (2) (7) disapproval (6) 13 -ness (4), voracity (2) (6) 1 exacerbation (6) 1 edible (3) 13 cowardice (3) 13 conquest (3) 12 complacence12 (0) (3) 2 asininity (3 ) guile (2) tempestuous (1) capacious (0) 11 chastisement (0) putrefaction (0) arcanum (0) parsimony (0) 416
KN(B) 11 9 9 12 10 (1) 9 12 8 6 5 2 2 7 3 (1) 8 (1) 10 (1) 9 (1) 7 5 6 (1) 7 (1) 1 4 (1) 5 (2) 3 2 (1) 7 (1) 8 (4) 5 (1) 3 (2) 4 2 (1) 1 9 (4) 6 (3) 4 (2) 6 (3) 1 1 (1) 4 (4)
233 (37) 84.1%
TK(A)
TK(B)
1
3 4 4 2 (1) 4 5 (1) 2 (2) 4 (1) 7 (2) 7 12 (2) 9 (1) 5 9 (1) 4 4 (2) 4 (1) 6 (4) 8 (3) 6 (2) 5 (1) 7 (3) 8 (3) 8 (4) 7 (2) 7 (3) 6 (1) 6 (4) 6 (3) 6 (2) 5 (3) 2 (2) 4 (2) 3 (1) 5 (2) 5 (4) 7 (2) 5 (3) 6 (6) 8 (7) 6 (6) 4 (2) 5 (3) 4 (3) 4 (4) 6 (6) 6 (6) 6 (6) 5 (5) 5 (5) 276 (127) 54.0%
3 1 1 2 1 1 1 2 2 3 3 2 2 1 2 1 3 3 4 5 2 2 2 3 4 1 2 2 2 2 4 3 1 4 1 1 1 2 2 2 3 5 3 4 4 3 2 111
NK(A)
NK(B)
1 (1)
2 (1) 2
1 1 1 7
1 10
3 (2) 1 2 (1) 2 (1) 1 1 1 2 (1) 2 6 (5) 2 1 (1) 3 5 (3) 8 (4) 1
1 1 10 9 9 11
3 5 (4) 8 (2) 8 (4) 10 (4) 9 (4)
9 12
7 (1) 9 (4) 2 2 2 8 9 (1) 9 (1) 10 (1) 4 8 8 7 9 183 (46) 74.9%
9 12 11 9 10 10 9 12 165
82 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
derivative received the same score, the word for which the students showed the greatest accuracy in their self-evaluation comes first.) In Table 3.3 (English test), words included on West’s (1953) general service list are, as in Figure 3.1, indicated in bold. In addition, in both tables the students’ evaluation of their knowledge of the given word as well as the word to be formed is incorporated. For example, for Word Pair 3 (apply – application), for which the number indicates where on the test the item occurs and consequently also its frequency in relation to the other word pairs, the table shows that apply was the word given and application the derivative the students were meant to form. (Apply was given in the past tense, as indicated in parentheses after the word’s base form.) Fourteen students provided the correct derivative. Twelve of the students were sure that they knew the meaning of the given word (column KN(A)) and nine were sure that they had offered the right derivative (column KN(B)). Two of the 15 students were less sure of the meaning of the given word in that they indicated that they only thought they knew the meaning of applied (column TK(A)), while five of the 15 students indicated that they only thought they knew what derivative to fill in (column (TK(B)), one of whom was incorrect (as indicated in parentheses). None of the students stated that they did not know the meaning of the word given (column NK(A)) or that they did not know what derivative to provide (column (NK(B)). For this specific word pair, 14 of the 15 students offered the requested information in their self-evaluation. (The sums of the figures in columns KN(A) and TK(A) on the one hand and KN(B) and TK(B) on the other hand both yield 14 answers each.) Finally, those squares that include the majority of the students being correct about the degree to which they knew the items of a specific word pair are coloured grey. This means that, in the example discussed, the majority of the students who answered the requested information knew both the meaning of the word given (12 informants) and what word to form (nine informants). In contrast, for Word Pair 42 (abbreviate – abbreviation) the majority of the students (10) did not know the meaning of the word given, nor did the majority know what word to form. (Of the eight students who indicated that they did not know what derivative to provide, two of them were incorrect and did form the correct derivative. That learners state that they do not know a derivative but then produce the correct form anyway is not unheard of. In Schmitt and Zimmerman (2002), discussed in more detail in the theoretical background section (Section 3.2), for example, this was a comparatively frequent phenomenon.) Also, in a few cases more than one derivative form was considered to be correct. For instance, both intrepidity and intrepidness were accepted as possible noun derivatives of the adjective stem intrepid. The students’ scores for these alternatives are indicated in parentheses immediately after each derivative form. The present author based her acceptance/rejection of such alternatives on the presence/absence of them in the BNC.
Word Formation: With a Focus on Derivational Suffixation 83
Table 3.4 T he students’ results on and evaluation of Part A of the Swedish suffixation test, with the test items listed in order of the correct number of answers, starting with the item that received the highest score (KN = the word is known, TK = the word is thought to be known, NK = the word is not known). Figures in parentheses in columns KN(B), TK(B), and NK(B) indicate cases where the students were incorrect in their self-evaluation No.
Word given A
4 allvarlig 19 cykel 18 glömt (glömma) 24 mätt 30 brutal 43 tolerera(r) 2 vecka 9 organisation(en) 17 arrangera(t) 16 journalist 21 teori(n) 25 skam 29 blyg 33 skryt(a) (-er) 1 förslag 37 vilsen 3 centralt 13 acceptabel(t) 20 fattig(a) 11 erbjud(a) (-er) 42 enfaldig 7 uppenbar(t) 39 förälska(r) sig 34 idiot 31 komfort(en) 35 staty 15 dansa 48 konfirmera sig 5 köra (kört) 46 verifikation 38 teologi 32 tåla 14 medveten 22 tvivel 45 ståupp 27 stipendium 40 fanatism 28 feber(topp) 36 hysterisk 23 sanning(en) 8 aktuell(t) 26 lejon 47 instruera 12 kosta(t) 41 kvantitet(en) 10 (hög)säsong 49 fabrikation 50 gondol 44 ensligt 6 avtal(et) Total Correctness rate
Word formed B KN(A) (correct answers) allvar (15) 14 cyklist (15) 14 glömska13 (12) (15) 14 mätthet (2), mättnad (13) (15) 14 brutalitet (15) 14 tolerant (15) 14 veckovis (15) 14 organsiera (15) 14 arrangemang (15) 14 -ik (11), journalism (4) (15) 14 teoretisk (15) 14 skamfull (1), skamsen (14) (15) 14 -het (14), -sel (1) (15) 14 skrytsam (15) 14 förslagsvis (15) 14 vilsenhet (15) 14 centrum (15) 13 acceptera (15) 13 fattigdom (15) 13 erbjudande (15) 13 enfald (2), enfaldighet (13) (15) 13 uppenbarligen (14) 14 förälskelse (14) 14 idioti (14) 14 komfortabel (14) 14 statyett (14) 14 dansare (10), dansös (3) (13) 14 konfirmand (13) 14 körbar (4), körd (9) (13) 14 verifiera (13) 12 teolog (13) 11 tålig(a) (12) 14 medvetenhet (12) 14 tvivelaktig (12) 14 ståuppare (12) 14 stipendiat (12) 14 fanatiker (12) 13 feberfri (12) 13 hysteri (11) 14 sanningsenlig (11) 14 aktualisera (11) 13 lejoninna (10) 14 -(a)nde (2), instruktiv (8) (10) 14 kostsam (10) 13 kvantitativ (10) 12 säsongbetonat14 (4) (10) 11 fabricera (9) 9 gondoljär (7) 12 ensling (6) 13 avtalsenlig(t) (4) 13 672
KN(B) 14 14 13 13 13 13 12 12 12 11 10 10 10 9 8 8 14 13 13 12 9 13 (1) 12 10 10 7 13 (1) 12 (2) 9 (2) 9 8 (1) 10 (1) 11 (2) 8 (2) 6 6 11 10 (2) 11 (3) 5 (1) 5 (1) 11 (4) 5 10 (2) 8 (1) 2 6 (2) 2 7 (3) 5 (4) 485 (32) 93.4%
TK(A)
1
1
2 1
1
1 1 4 2 1 15
TK(B)
NK(A)
NK(B)
1 1 1 1 2 2 2 3 4 4 4 5 6 6
1 5 1 2 (1) 3 4 (1) 6 1 (1) 2 5 5 (2) 5 (1) 4 (1) 3 (1) 6 (1) 8 (3) 5 (2) 3 (2) 3 3 9 (2) 7 (1) 3 (1) 9 (2) 3 (2) 2 (1) 6 (1) 8 (3) 8 (3) 5 (4) 8 (5) 185 (41) 77.8%
1 1
1
3 (2)
1
1 1
3
4 (2) 4 (3) 4 2 20 (7) 65.0%
Alternative forms also exist in the Swedish material for which Språkbanken was used to decide whether they would be included or not. The students’ results on the English test Part A (Table 3.3) can be divided into three main sections, each section having different characteristics. The derivatives that received between 15 and 10 correct answers
84 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
form one such group, that is, from advise(d) – advice (Word Pair 9) through acquaint(ed) – acquaintance (Word Pair 30). (Inordinate – inordinately (Word Pair 38) is excluded here but will be discussed later in this section.) Firstly, these 21 word pairs can all be found among the 30 most frequent word pairs on the test. As discussed in connection with Figure 3.1, the majority of these derivatives are also found on West’s (1953) general service list. It should be noted that it is not only the majority of the derivatives that are found there, but also the words on which these derivatives are based; that is, the majority of the words in all these word pairs are frequent words that these Swedish learners have probably encountered quite a few times during the 10 years or so that they have been studying English. In the theoretical background, it was discussed whether derivatives are decomposed or regarded as unseparated wholes when encountered by a learner, and that age may be one of many factors determining whether decomposition occurs or not. Marslen-Wilson et al. (1994), for instance, claim that the younger a learner is when first encountering a derived form, the less likely it is that the learner decomposes it when encountering it later in life. Due to their frequency – that is, due to the great number of links having had the time to form for these high- frequency items, these word pairs may thus be tackled as unseparated wholes. The parsing ratios for two of the suffixes involved here appear to point in this direction. The fact that the suffix -ness (as in cleverness (Word Pair 24)) has a higher parsing ratio than -ity (as in equality (Word Pair 11)) does not appear to affect the students’ performance with these derivatives to any great extent, even though, since the former suffix is decomposed more often than the latter, the students should have encountered -ness more often than -ity. (Within this group, however, it should be noted that the derivatives ending in -ness generally received higher scores than the ones ending in -ity.) Furthermore, the students’ ability to produce correct derivatives for these word pairs seems to be independent of the complexity of the suffix in question, a fact that also indicates that the students approach these derivative forms as unseparated wholes. (The reader is reminded that complexity is here based on the frequency, predictability, productivity, and regularity of the suffixes as set up by Bauer and Nation (1993) (see Section 3.2)). For example, the students are not only able to form the derivatives anxiousness, ability/equality and courageous with the suffixes -ness (Affix Level 3, which includes the most frequent and regular affixes), -ity and -ous (Affix Level 4, which also includes comparatively frequent affixes that are still quite regular), but also the derivative forms entrance, arrival, application, and birth/depth with the suffixes -ance, -al, -ion and -th. These latter suffixes, in contrast to -ness and -ful, belong to the more complex Affix Levels 5 (-ance and -al) and 6 (-ion and -th), where Level 5 includes regular but infrequent affixes and Level 6 includes comparatively frequent affixes but whose spelling entails difficulties. If the students approach these derivatives as unseparated units, which seems to be the
Word Formation: With a Focus on Derivational Suffixation 85
case, it may mean that they have not understood what these suffixes do (i.e. their meaning) and consequently not what word classes they are able to form; that is, the students may not perceive these suffixes as psychologically real units. However, incorrect forms of these high-frequency derivatives produced by the students do show signs of awareness of word class affiliation. Table 3.5 gives a less than complete yet representative list of their errors. As can be seen, the suffix of choice by far when an incorrect derivative was produced for a noun-forming suffix was -ness, as in *ableness, Table 3.5 Examples of errors made by the Swedish students in Part A of the English suffixation test No.
Word given
9 8 20 4 5 3 10 17 6 7 2 16 1 24 26 11 25 29 14 23 30 38 32 15 12 19 40 27 22 21 36 42 35 43 48 28 44 39 13 34 31 33 50 46 37 18 47 49 41 45
advise(d) colour(s) anxious bear (born) enter apply (applied) achieve arrive(d) deep easily early clean(ed) able clever cautiously equal cruel envy (envied) absent courage acquaint(ed) inordinate annoy(ed) comfortable advantage broad intrepid admire(d) complain(ing) bury (buried) contagious abbreviate encroach surly ruminate (did)n’t approve voracious exacerbate(d) eat coward conquer(ed) complacent asinine guileful tempest capacity chastise(d) putrefy arcane parsimonious
Word to be formed (no. of correct answers) advice (15) colourful (15) anxiety, anxiousness (15) birth (14) entrance, entry (14) application (14) achievement (13) arrival (13) depth (13) ease, easiness (13) earliness (13) cleaner(’s) (13) ability (13) cleverness (12) caution, cautiousness (12) equality (11) cruelty (11) envious (10) absence (10) courageous (10) acquaintance (10) inordinately (10) annoyance (9) comfort (9) advantageous (9) breadth, broadness (9) intrepidity, intrepidness (9) admiration (8) complaint(s) (8) burial (8) contagion, contagiousness (7) abbreviation (7) encroachment (7) surliness (7) ruminative, ruminant (7) disapproval (6) voraciousness, voracity (6) exacerbation (6) edible (3) cowardice (3) conquest (3) complacence15 (3) asininity (3) guile (2) tempestuous (1) capacious (0) chastisement (0) putrefaction (0) arcanum (0) parsimony (0)
Errors made by the Swedish L2 learners *borness *entress *appliance *achievness *arriviation, *arrivement *easilness *early *cleaning *ableness *clevers *cautioness *equalness *cruelness *envy *absentness *couragable, *couraged, *courageful *acquaint, *acquaintant *inordinated *annoyingness, *annoyment *comfortability *advantageable *broad, *broadity *interpretence *admirance, *admire, *admirement *complainment(s) *buriement *contagious *abbreviatal, *abbreviaty, *abbrevification *encroachability, *encroach(es) *sur, *surlity, *surness, *rumiated, *ruminous *disapprovement *voration, *voracationess, *voracioun, *voraciousion *exacerbateness, *exacerbatity, *exacerbility *eatful *cowardy *conquer, *conqueration, *conquerment *complacation, *complace, *complaceness, *complacinity *asiniation, *asinination, *asininence, *asininess *guilefulness *tempestant, *tempested, *tempesting, *tempestish, *tempestive *capacitable, *capacitive *chastisal, *chastision, *chastisitation *putration, *putrefiance, *putrefyal *arcancy, *arcaness, *arcanity, *arcanous, *arcantage *parsimon, *parsimonia, *parsimonionism, *parsimonity
86 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
*cautioness, and *absentness. In addition to being one of the most frequent and productive suffixes in the English language, this suffix is, since it is placed on a low difficulty level by Bauer and Nation (1993), also highly regular and very predictable in nature. Furthermore, as pointed out above, the suffix -ness carries great memory strength due to its relatively high parsing ratio; that is, it is most often seen in derivatives that are decomposed. It may be that the students are intuitively aware of the usability of this suffix. That is, put differently, it may mean that the students use -ness as a default suffix when uncertain of the form of a noun derivative. However, in using -ness as a default suffix the students have disregarded the fact that there are morphological restrictions on its use, namely that -ness can only be attached to adjective stems (Ljung, 2003: 92). Thus errors such as *borness (instead of birth), *entress (instead of entrance), and *achievness (instead of achievement), where -ness has been added to verb stems, are all examples of overgeneralisations of this suffix. For the highfrequency word pairs discussed in the present paragraph, this error type appears to be most pervasive with -ness, but it can also be seen in connection with Word Pair 23, for which the derivative *couragable, in which the suffix -able has been added to a noun stem even though it may only be attached to verb stems, was offered by one student (Ljung, 2003: 102). Although the learners’ understanding of word class affiliation is at its clearest with the suffix -ness, there are more errors that signal such awareness. For example, the incorrect forms *arriviation and *arrivement, containing the noun-forming suffixes -ion and -ment, were offered instead of the noun derivative arrival, and *couragable and *courageful, with the adjective-forming suffixes -able and -ful, were produced instead of courageous. In fact, in this group few errors occurred for which the students showed ignorance of the word class affiliation of the derivative forms. Instead the main problem seems to have to do with what type of stem the affix should be added to. Finally, the students’ stellar results on these high-frequency word pairs are also reflected in their own evaluation of the degree of knowledge they possess of these derivative forms (see Table 3.3). Not only are the majority of the students in most of these cases sure of the meanings of the words given (column KN(A)), but in most of these cases the majority also indicate that they are sure of what derivatives to form (column KN(B)). This applies to all but four derivatives (ease/easiness, earliness, cleaner(‘s) and cleverness).16 Among these high-frequency items, Word Pair 38 (inordinate – inordinately) is an interesting anomaly since, even though 10 students were able to produce the correct derivative form, the majority claimed that they did not know the word given and that they were not certain of the form of the derivative. It may be that the highly frequent and productive nature of the suffix -ly played a greater psychological role than the infrequency of the stem, thus helping the students produce the correct derivative.
Word Formation: With a Focus on Derivational Suffixation 87
The description offered of the 21 high-frequency word pairs above stands in stark contrast to the word pairs seen in the bottom section of Table 3.3, ranging from contagious – contagion/contagiousness (Word Pair 36) through parsimonious – parsimony (Word Pair 45). First of all, the large majority of these derivatives, with scores between 0 and 7, are not on West’s (1953) general service list. Instead the majority of these 20 word pairs are found among the last word pairs presented to the students; that is, the majority of these words are highly infrequent even in the large BNC corpus. That the students were able to produce correct derivatives using, for example, the suffixes -ness and -ity in connection with the highfrequency items discussed above, but not with the highly infrequent items discussed in the present section (as in surliness, voracity and asininity), clearly shows the effect that frequency has on the students’ performance. The same can be seen with the suffix -ous, with which the majority of the students were able to form the derivative courageous, forming part of a high-frequency word pair, but not the derivative tempestuous, which is part of a low-frequency word pair. Furthermore, due to the infrequencies of the majority of these words, the relative simplicity (according to Bauer and Nation’s [1993] Affix Level Continuum) of some of the intended suffixes seems irrelevant. Again the suffix -ness (Affix Level 3) can serve as an example in connection with the highly infrequent stem surly, for which the majority of the students were not able to produce the correct derivative surliness (Word Pair 43). The suffix -ful, to be removed from the word guileful to form guile (Word Pair 46) and belonging to Affix Level 4, may serve as a second example. Put simply, it seems that the infrequency of the word pairs takes its toll on the subconscious knowledge these learners may have of suffixation rules in English. The contrast between the high- and low-frequency word pairs is also clearly visible in the students’ own evaluation of their knowledge of these word pairs. Whereas there is, as discussed above, a concentration of great certainty about the meanings of the words given and the words to be formed for the test items with scores from 15 through 10, for the majority of the low-frequency word pairs most of the students indicate their total lack of knowledge of both the words given (column NK(A)) and the words to be formed (column NK(B)) (see Table 3.3). However, similar to the high-frequency word pairs, there are some exceptions among the lowfrequency word pairs. One of the six exceptions is capacity – capacious (Word Pair 18). The reason this ended up among the least known suffixes may be explained by the fact that it is the first word given in the word pair, capacity, that almost entirely contributes to the relatively high frequency of the word pair; capacious is a derivative that the students probably have not encountered many times, if at all. A typical trait of the errors produced in connection with these littleknown word pairs is that the students appear to try a wider variety of suffixes (see Table 3.5 again). For instance, instead of the correct
88 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
derivative form tempestuous (belonging to Word Pair 37) students offered the incorrect forms *tempestish, *tempestive, *tempestant, *tempested and *tempesting, and instead of the correct noun derivative parsimony (Word Pair 45) the incorrect forms *parsimon, *parsimonia, *parsimonionism and *parsimonity were given. It appears that the more infrequent the stems are, the more uncertain the students are about what suffixes to use. Interestingly enough, in general the students still seem to experience few problems with word class affiliation here in that when, for example, an adjective is sought, as in the examples above, the students offer stems to which another adjective-forming suffix has been added, though the derivatives are incorrect. Again, it is rather the semantic restrictions placed on the stems that cause the most problems. *Exacerbateness, *capacitable and *tempestive are all examples of errors where the students have added the right type of suffix to the wrong word class stem. *Exacerbateness and *capacitable belong to the error types already exemplified and discussed above. In *tempestive the adjective-forming suffix -ive has been added to a noun stem when it really can only be added to verb stems (Ljung, 2003: 102). The mid-section of Table 3.3, from annoy(ed) – annoyance (Word Pair 32) through bury (buried) – burial (Word Pair 21), forms the third group. For the majority of these word pairs, most of the students seem to be sure about the meaning of the word given (column KN(A)), but less sure about the form of the derivative, indicating in the majority of cases that they only think they know the form of the sought word (column TK(B)), with the solidifying of the students’ knowledge of these derivatives thus depending on how many (additional) times they will encounter these in the future. (There are only three exceptions to this: intrepid – intrepidness/intrepidity, admire(d) – admiration and complain(ing) – complaint(s).) The uncertainty the students feel when dealing with this group of word pairs appears to have many causes. In some cases, it may be the relatively high difficulty level of the suffix alone that may be troublesome to the students, as is the case with the suffix -ance (belonging to Affix Level 5) used to form the derivative annoyance (Word Pair 32), the stem annoy being a comparatively frequent item. In other cases, it may be the relative infrequencies of the stem and derivative in combination with the comparatively high difficulty level of the affix in question, such as with the derivative burial (the suffix -al also belonging to Affix Level 5) seen in Word Pair 21. In yet other cases, it may be the fact that the students were instructed to remove the suffix rather than add one that made it difficult, as in Word Pair 15, where the students were supposed to truncate the adjective comfortable and produce the noun comfort. The students here appear to be in a transitional stage in which they have gained knowledge of the meaning of the words given but a number of factors seem to hinder production of the correct derivatives. No clear trend could be detected for the errors made by the students, but here too they seem to be intuitively
Word Formation: With a Focus on Derivational Suffixation 89
aware of word class affiliation. Errors in which suffixes have been attached to the incorrect word class stem can be seen, however. Lastly, the students appear to be subconsciously aware of what combinations are acceptable in terms of the origins of stems and affixes, since no such errors could be detected for any of the word pairs on the English test. As pointed out in the theoretical background (Section 3.2), studies show that for many L2 learners it is very often the nature of the stem rather than the affix that determines whether a learner is able to produce a derivative form or not (see also Hancin-Bhatt & Nagy, 1994). The obvious problem with Bauer and Nation’s framework is therefore that it totally ignores the role played by the stem. However, considering the nature of the stem alone also seems to be too simple an approach. Rather, the success or non-success with which a learner is able to produce a correct derivative form appears, based especially on the students' results on the words pairs in Part A, to depend on an intricate interplay of three dimensions: (a) the independent frequencies of the stem and derivative, (b) the complexity of the suffix (in terms of its frequency, predictability, productivity, and regularity), and (c) the relative frequencies of the stem and the suffix (in which the parsing ratio of the suffix is a useful tool). Thus it seems that if these three factors could be merged into a continuum, this could help predict the ease or difficulty with which a learner will be able to produce a correct derivative form and could, by extension, help teachers when selecting items to be discussed in the L2 classroom. Whereas the English test results on Part A formed three main groups with discernible differences, the corresponding Swedish test part can only be divided into two sections (see Table 3.4 and Appendix 1 for a translation of these Swedish word pairs). (The reader is reminded that no affix level continuum equivalent to the one set up by Bauer and Nation (1993) could be found for Swedish affixes, nor could information about parsing ratios be gained. The discussion can therefore not be as detailed as with the students’ results on the English test.) The first section includes 39 word pairs, starting with Word Pair 4 (allvarlig – allvar (serious – seriousness)), for which all 15 students gave the correct derivative, down to, and including, Word Pair 36 (hysterisk – hysteri (hysteric/al – hysteria)), for which 11 students were able to produce the correct derivative form. Among these, high-frequency word pairs as well as very infrequent word pairs can be detected. It is of course due to their greater L1 vocabulary knowledge that the students are less susceptible to the frequencies of these word pairs than those tested in their L2. Here the students are also less sensitive to the degree of productivity in that they show mastery not only of highly productive suffixes, such as the verb-forming suffix -era (as in verifiera (=verify) (Word Pair 46)), but also of suffixes that only show some productivity, such as the noun-forming suffix -het (as in vilsenhet (=forlornness) (Word Pair 37)), as well as those suffixes that are not productive at all, such as the noun-forming suffix -dom (as in fattigdom (=poverty) (Word
90 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Pair 20)). Furthermore, for all of these items, the majority of the students indicated that they were sure not only about the meaning of the word given (column KN(A)) but also about the derived form sought (column KN(B)). Thus this group clearly corresponds to the high-frequency word pairs ranging from advise(d) – advice through acquaint(ed) – acquaintance seen in the English material. Based on the few errors in connection with this group of derivatives, it is impossible to decide whether the students actually perceive the suffixes as psychologically real units or whether the derivatives have been acquired as unseparated wholes. The students do not, however, appear to have used any suffix as a default suffix, which seemed to be the case with -ness on the English test, especially with the high-frequency word pairs. For the majority of the remaining word pairs, having received between 4 and 11 correct answers, the students were still sure of the meaning of the word given but less sure of the word to be formed, thus being in a transitional stage similar to the one seen in connection with the word pairs forming the mid-section of the Table 3.3 representing the English data. This group, just like the corresponding L2 group, incorporates both highand low-frequency word pairs for which the students experience difficulties in producing the correct derivative for a variety of reasons. Consequently, it is with this miscellaneous group that the complex interplay observed in connection with the English material (i.e. (a) the independent frequencies of the stem and derivative, (b) the complexity of the suffix and (c) the relative frequencies of the stem and the suffix) is most pronounced on the Swedish suffixation test. On only a few occasions in the Swedish material could errors be seen in which the students had attached a suffix to the wrong word class stem. Also, as in the English material, no errors showing signs of incongruity in terms of the origin of the stem and suffix could be detected. Results on Part B (word families, decontextualised)
Tables 3.6 and 3.7 present the students’ results on Part B of the English and Swedish tests respectively.17 In both tables the word families are presented according to the number of members, starting with the families consisting of two members only. Within these subgroups, the word families are given in order of accuracy so that the family for which the students were able to produce most correct answers comes first. Furthermore, the number given to each word family does not only signal where on the test the family appeared, but also its frequency in relation to the others. (As in Part A, if two or more families received the same score, they are listed in accordance with the students’ self-evaluation so that the word family for which the students were most accurate is placed first.) Frequency within each word family is colour-coded so that the most frequent word-family member remains white, the second most frequent is
Word Formation: With a Focus on Derivational Suffixation 91
Table 3.6 The students’ results on and evaluation of Part B of the English suffixation test (WF = word family, KN = the word is known, TK = the word is thought to be known, NK = the word is not known) (□ = most frequent in the WF, = second most frequent, = third most frequent, = fourth most frequent). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation WF 2 9 20 18 11 7 14 23 13 24 19 17 3
Words given (correct answer/total) boy (14/15) deflection (13/15) sultriness (6/15) adorn (4/15) acrimony (4/15) arduous (4/15) judicious (3/15) assiduity (3/15) opulent (2/15) portent (2/15) abjection (2/15) affront (verb) (1/15) brief (0/15)
KN 13 3 2 1 1 1
12
TK 2 2 5 3 2 2 1 1 2 2 3 1
9 11 6 9 10 10 14 11 13 10 11 1
Words to be formed (errors) boyish (1) deflect (2) sultry (9) adornment (11) acrimonious (11) arduousness (11) judiciousness (12) assiduous (12) opulence (13) portentous (13) abject (13) affront (noun) (14) brevity (15)
2 (2)
4 (1) 4 5 (3) 5 (2) 4 (3) 3 (2) 2 (2) 5 (4) 5 (3) 3 (2) 3 (2) 3 (2) 8 (8)
2 1
5 (2) 4 (2)
6 (4) 8 (3)
2 1
5 4 (3)
5 (2) 7 (1)
2 1 (1)
10 (7) 11
NK
KN
TK
NK
7 3 (1)
2 (2) 7 (7) 8 (3) 7 (1) 9 (1) 8 (4) 10 (2) 10 (2) 8 12 (1) 9 11 4
1 (1) 1
27
procrastination (15/30)
6
2
5
procrastinate (6) procrastinator (9)
26
appropriation (13/30)
5
6
1
appropriate (5) appropriator (12)
28
expiate (11/30)
1
11
expiation (4) expiatory (15)
8
abrasion (5/30)
2
10
abrasive (noun) (12) abrasive (adj.) (13)
22
nascent (3/30)
1
6
nascence (13) nascency (14)
1 (1) 1 (1)
6 (1) 6
10
acquiescence (2/30)
3
10
acquiesce (13) acquiescent (15)
3 (3) 1 (1)
10 (2) 12
25
purvey (1/30)
2
9
purveyor (14) purveyance (15)
2 (2) 2 (2)
9 (1) 9
16
zealous (0/30)
zealot (15)
1 (1)
10
zealotry (15)
4 (4)
7
3 5 (3) 3 (1)
3
3 (1) 2 (2) 3 (3)
8 (7) 9 (6) 8
4 (4) 5 (3) 8 (3)
7 (2) 8 (1) 1
1
analys/ze (24/45)
21
adulation (16/45)
4
diagnostic (14/45)
2
1
4
6
analytic (6) analysis (7) analytical (8)
8
3
6
8
3 (2) 3 (3)
2 3 (1) 3 (1)
adulate (5) adulator (9) adulatory (15)
2
7
diagnostician (7) diagnostics (11) diagnosis (13)
1 2 (2)
4 (2) 3 5
4 (4) 2 (2) 3 (3)
8 10 8
1 (1) 1 (1) 1 (1)
10 (1) 10 10
147 (96) 34.7%
357 (68) 81.0%
6
derision (3/45)
2
3
derisive (12) deride (15) derisory (15)
5
sagacious (2/45)
1
11
sage (noun) (14) sagacity (14) sage (adj.) (15) serfdom (14) serfage (15) serfhood (15)
15
serf (1/45)
3
1
7
12
recalcitrant (adj.) (0/45)
-
-
-
64
61
202
2 (2)
11 (2) 11 (2)
4 (1)
1 (1)
recalcitrant (noun) (15) recalcitrance (15) recalcitration (15)
Total Correctness rate
35 (10) 71.4%
92 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 3.7 The students’ results on and evaluation of Part B of the Swedish suffixation test (WF = word family, KN = the word is known, TK = the word is thought to be known, NK = the word is not known) (□ = most frequent in the WF, = second most frequent, = third most frequent, = fourth most frequent, = fifth most frequent). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation WF
Words given (correct answer/total)
KN
11 17
umgås (12/15) generös (9/15)
12 12
TK
NK
KN
TK
NK
umgänge (3) generositet (6)
7 5
4 (1) 5 (2)
1 2
avgöra/s (1) avgörande (adj.) (3)
9
4 (1) 12 (2)
1 (1)
4
avgörande (noun) (26/30)
11
13
söt (20/30)
12
söta/s (2) sötande (1), sötma (3), söthet (3) (8)
4 1
6 (1) 7 (3)
2 (1) 4 (1)
9
mörk (mörkande, mörka, mörknande) (20/30)
11
mörkna (4) mörker (6)
2 2
7 (1) 7 (3)
2 2 (1)
3
resultera (resulterande) (19/30)
11
2
resultat (0) r-inriktad (0), -givande (2), resultativ (2) (11)
8
5 4 (2)
9
18
vidgande (adj.) (17/30)
10
1
vidga/s (2) vidd (1), vidgande (noun) (1), vidgning (2) (11)
6 2 (1)
5 5 (3)
4
12
tränga/s (sig) (trängande) (17/30)
10
1
trång (4), trängd (5) (6) trängning (3), trängsel (5) (7)
2 (1) 1
8 (2) 7 (3)
1 3 (1)
8
tömma/s (tömmande) (13/30)
11
1
tomhet (3), tömning (5) (7) tom (10)
2 (1) 1 (1)
7 (2) 7 (4)
3 (1) 4
7
jämföra/s (sig) (jämförande) (11/30)
9
1
jämförelse (6) jämförbar (13)
5 1 (1)
5 (4) 6 (4)
3
7
1 (1) 4 (1)
6
1
salt (adj.) (34/45)
givare (givande) (28/45)
2
Words to be formed (errors)
2 (1)
3 9 (1) 4
9
ge, giva/s (0) gåva (7) givmild (10)
6 1 1
3 7 (2) 2 (1)
1 6 (1)
8 5 2
3 (1) 4 (1) 8 (5)
2 1
9
salta/s (1) salt (9), saltande (1), saltning (1) (3) sälta (7)
1
16
ljuga (ljugande) (27/45)
11
lögnare (3) lögn (5) lögnaktig (10)
2
glad (27/45)
10
glädje (3) glädja/s (4) glädjande (11)
2 1
7 (1) 9 (3) 8 (6)
14
skämmas (24/45)
10
skam (9), skamsenhet (1) (5) skamsen (7) skamfull (3), skamfylld (1), skamlig (2) (9)
2 2 1
8 (1) 7 (2) 7 (4)
1 2 (1)
2
8 (2) 5 (2) 6 (4)
1 2 (1) 3 (1)
1 5 (4) 3 (3)
5 (2) 8
2 2 3 (2)
4 (1) 4 (1) 3
4 3 6 (5) 6 (3)
1
1
5
tvinga/s (21/45)
9
tvingande (6) tvång (7) tvångsmässig (11)
20
skvaller (skvallrande) (18/45)
12
skvallra/s (2) skvallrig (11) skvalleraktig (14)
19
blyg (15/45)
6
blyghet (7), blygsel (1) (8) blygsamhet (10) blygas (12)
10
sköta/s (sig) (35/60)
10
skötare (1) sköterska (2) skötsel (11) skötsam (11) svikare (1)
15
svika (30/60)
svek (7)
7
sviken (10) Total Correctness rate
202
9
0
svekfull (12)
10 1
5 7 2 (1) 3 (2)
1 2
2 1
7 3
3 (1)
4 (1)
2 (1)
1
2 (1) 144 (11) 92.4%
2 (1) 261 (94) 64.0%
3 (1) 101 (17) 83.2%
given in black and the least frequent word-family members are indicated in different shades of grey (see Tables 3.6 and 3.7 for specifics), with word families of five members only existing in Swedish (see Table 3.1). Finally, the squares colour-coded in grey in columns KN, TK and NK indicate in which cases the students were most accurate in their self- evaluation. For example, among word families consisting of four members,
Word Formation: With a Focus on Derivational Suffixation 93
number 21 is the one the students knew second best. The verb adulate was here the easiest member to produce, with only five of the 15 informants (indicated in parentheses) being unable to offer this derivative. The verb derivative is followed by the noun derivative adulator, with as many as nine of the 15 students not being able to come up with this word. Interestingly, even though the majority of the students wrote that they did not know the noun, six of them actually produced the correct form (column NK). Finally, the adjective derivative adulatory proved to be the most difficult one. None of the 15 students gave the correct derivative form; even three students who indicated that they thought they knew this word failed to provide it (column TK). Table 3.7, showing the students’ results on the Swedish test, should be deciphered in the exact same way as the results of the English test. (For a translation of the Swedish word family members, see Appendix 1.) In the L2 data (Table 3.6), if the most common word families are contrasted with the least common ones in relation to the number of members, there is a clear frequency effect for families consisting of two members (which was also observed in Part A) as well as for those consisting of four members; that is, frequency plays a part here in whether a word family is well known or not. However, the same effect cannot be observed with three-member word families, though, this discrepancy may be entirely attributed to some of the test words, as two of the low-frequency items simply involved truncation of a noun (procrastination → procrastinate and appropriation → appropriate), yielding many correct answers. The conclusion must therefore be drawn that there is a frequency effect between word families. This effect is even stronger in the L1 material. The reader is here reminded that the total frequency of all the word families in the Swedish data is higher (constituting ≈0.142% of Språkbanken) than the total frequency of the English ones (making up ≈0.041% of the BNC). This higher frequency in the L1 material has therefore given room for a greater frequency span and may hence explain the L1–L2 difference. Interestingly, there also appears to be a frequency dependency within word families in the English material. In nine of the 14 word families for which the students were asked to produce two or three derivatives, they did best on the most frequent family member (six out of eight for two-member families; three out of six for three-member families). The same can be seen in the Swedish data. For 10 of the 18 families, the students offered most correct answers for the most frequent family member (five out of eight for word families where two and three members were required respectively). Errors made in connection with these word families exist in abundance. However, as a separate section was dedicated to this in connection with Part A of the test, additional examples will not be provided here. Another aspect considered in connection with students’ knowledge of word families is their mastery of different word classes. Table 3.8 offers the students’ results on the different word classes included in the present investigation: nouns, verbs and adjectives.
94 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 3.8 Correctness in relation to word class (noun, verb and adjective) in Part B of the English and Swedish tests Part B
Nouns
Verbs
Adjectives
English test
19.1% (29 types)
48.9% (6 types)
18.1% (15 types)
Swedish test
62.8% (24 types)
79.3% (9 types)
35.3% (17 types)
As in the Schmitt and Zimmerman (2002) study discussed in the theoretical background, the students in the present study were, in both languages, able to produce verb derivatives more accurately (English: 48.9%; Swedish: 79.3%) than noun derivatives (English: 19.1%; Swedish: 62.8%) and adjective derivatives (English: 18.1%; Swedish: 35.3%). (The differences on the Swedish test could be confirmed statistically at a significance level of 5% between all three word classes, while the same was only possible for the difference between verbs and nouns on the one hand and verbs and adjectives on the other hand on the English test.) The reason for this result, according to Schmitt and Zimmerman, is that verbs are very often the stem of a word family on which all the other members are based, and the verb is thus probably the member most often encountered by learners. Furthermore, Schmitt and Zimmerman, as well as Nagy et al. (1989), again discussed in the theoretical background (Section 3.2), noticed that adjectives form one of the more troublesome word classes. This also tallies with the present investigation, although it was not statistically confirmed on the English test. However, it needs to be pointed out that the six verb forms that the students were meant to form, the majority of which are high-scoring items, could be produced by simply truncating the nouns given. This was fairly straightforward in four of the six cases (deflect, appropriate, adulate and procrastinate from the nouns deflection, appropriation, adulation and procrastination). The same thing can be observed in the Swedish material, in which eight of the nine verb derivatives, all of which are highscoring items, could also be formed by truncating the given nouns. Even though these word families were not presented consecutively, the students may have become aware of the recurring theme in the forming of these verbs, something which, in addition to there being only six(English)/nine (Swedish) verb derivatives on the tests, may have favoured the students’ result with this word class. On the other hand, the majority of verbs are, as mentioned above, the stems on which other derivatives are based, so it may simply be that the students are intuitively aware of this. Whichever the case may be, it was not enough for students to form the verb acquiesce when they were faced with the noun acquiescence. It seems that the students here found it difficult to see what part of the given word made up the stem (possibly due to its spelling), so many of them added letters to the word rather than truncating it. It is therefore questionable whether they have really understood that the suffix -ence forms a semantic unit.
Word Formation: With a Focus on Derivational Suffixation 95
Also, whereas there were a greater number of different suffixes tested and Affix Levels included in connection with the noun derivatives, there was a higher concentration on a few suffixes in connection with the adjective derivatives. Generally, there are, of course, more noun- than adjective-forming suffixes, thus making the test representative of what students encounter in real-life situations. However, this difference may have skewed the results in favour of adjectives. (The same can also be observed in the Swedish material.) Individual L1 and L2 affixation knowledge and the correlation between affixation knowledge and vocabulary size
When the informants’ individual results are considered, there are L1– L2 correlations for six of the 15 students. Three of these consistently obtained high scores on both test parts in both languages and consequently achieved high scores on the whole tests (Students 2, 5 and 11). In fact, these three learners were consistently among those who were able to achieve the top five results. Another three subjects were low achievers in both languages, producing among the five lowest scores on both test parts and consequently on the tests as a whole (Students 1, 10 and 13). Transfer of affixation skills from a learner’s L1 to their L2, which may be in play for these students, has been observed in other studies too (e.g. Piasecka, 2006: 247). There is a correlation not only between the students’ ability/inability to form derivatives in the two languages but also between their affixation skills and their vocabulary breadth. (The Schmitt and Meara (1997) and Mochizuki and Aizawa (2000) studies presented in Section 3.2 also observed the existence of such a correlation.) In Chapter 2, the results of three of the parallel L1–L2 tests were discussed. The first test involved 100 items of vocabulary usually taught at upper secondary school level (i.e. comparatively frequent items), 50 of which were tested receptively and the rest productively. The second test included specialised uses of vocabulary in the areas of school/work, business, medicine, and technology, ranging from comparatively frequent to very infrequent items, all of which were tested receptively. The last of the three tests implemented, also constructed to test the students’ receptive knowledge, involved 100 low-frequency items, many of which proved to be difficult even for the native speaker. The results on these three tests can consequently be assumed to give a good indication of the students’ vocabulary size and can be compared to what they achieved on the word formation tests. This correlation is at its clearest for three informants whose performance was either consistently among the top scores on all the tests in both English and Swedish (Student 11) or consistently among the bottom scores (Students 4 and 10). For another seven informants, there were correlations on the majority of the tests in both languages, four learners being high
96 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
achievers (Students 2, 5, 9 and 14) and three being low achievers (Students 1, 6 and 13). It therefore seems legitimate to claim that the number of words known can help determine what a student may accomplish on a word formation test. There are also a few students for whom correlations can only be detected in one of the two languages. Whereas two of the 15 learners were among the high achievers on the majority of the English tests (Students 3 and 15), one achieved high scores on the Swedish tests only (Student 12). In the latter case, being the oldest informant included in the study, this 49-year-old male was probably able to excel in his native language due to the experience that comes with age. His consistently low scores on the English test can be explained by the fact that he had not used English actively for a great many years. (For a more detailed description of the informants, the reader is referred to Section 1.2.) Other studies show that there is a correlation not only between vocabulary size and morphological knowledge but also between general language skills and the mastery of word formation (e.g. for reading, see Zhang, 2017; for writing, see Leontjev et al., 2016). It is here interesting to note that the correlation observed with reading appears to be of a reciprocal nature in that morphological knowledge is not only a by-product of skilled reading but also contributes to superior reading (Carlisle, 2000; Mahony et al., 2000; Nagy et al., 1993; Singson et al., 2000). This may pave the way for those students showing great mastery of suffixation in the present study. 3.5 Conclusions and Pedagogical Implications
During informal discussions with the students after they had taken the two suffixation tests, it came to the fore that hardly any of them had ever worked with word formation in school before in either language, at least not in a structured way. On a more general note, since the construct of a word family is not only a linguistic term used for a set of words whose meanings have close affinities, but, as many studies have shown, a psychologically justified concept, students should, when taught a specific word, also be confronted with its derivative forms to a greater extent than what is apparently the case, and in connection with these forms should be taught what the affixes in question mean and do and to what stems they can be attached. This is important as a growing body of research indicates that explicit training helps learners (L1 as well as L2) to transfer their receptive affixation knowledge into productive use (L1 children: Cunningham, 1998; Stahl & Shiel, 1992; L2 children: Carlo et al., 2004; L2 adults: Mirhassani & Toosi, 2000). In addition, the results of the present study show that for derivatives that have not yet been ingrained into long-term memory, a continuum that considers (a) the independent frequencies of the stem and derivative, (b) the complexity of the suffix (in
Word Formation: With a Focus on Derivational Suffixation 97
terms of its frequency, predictability, productivity, and regularity) and (c) the relative frequencies of the stem and the suffix (in which the parsing ratio of the suffix is a useful tool) could help decide what suffixes should be selected to be made part of the teaching syllabus at different educational levels. Furthermore, the present investigation shows that for quite a few students there is a correlation between what they know and can do in terms of affixation in their L1 and their L2. It therefore seems that word formation knowledge gained in one’s native language may be transferred to a second language (and perhaps vice versa), thus automatically increasing the student’s vocabulary knowledge in two or more languages at the same time. This, if anything, should be an incentive to introducing discussions and exercises on word formation in class. Teachers should, however, be aware that ‘word family’ is not an equally real concept to all learners. The validity of this construct appears to depend primarily on the maturity level of the learner in question (Gardner, 2007), and discussions and instruction should be shaped accordingly. More systematic truncation of derivatives of various frequencies could perhaps, for instance, help students perceive existing patterns. Additionally, as quite a few spelling errors could be seen with the more irregular forms in the present study, with the students breaking phonological restrictions of which native speakers are mostly intuitively aware, teachers would gain the most from starting with derivatives that are regular in character (i.e. involve few changes in pronunciation, spelling, etc.) and not moving on to more irregular ones until these have been mastered. 3.6 The Chapter That Follows
In the following chapter, the 15 university students’ L1 and L2 knowledge of idiomatic expressions and sayings will be explored. As before, quantitative and qualitative aspects will be investigated thoroughly. While the more common and/or transparent of these items may form part of the learners’ vocabulary breadth, the more infrequent and opaque ones will, due to their double semanticity and the additional processing this entails, most likely offer valuable insights into the students’ vocabulary depth. Notes (1) The clip is available at https://www.youtube.com/watch?v=8gWt87sFTwk&ab_cha nnel=IntroducingCommunication. (2) Some of the affixes of Nordic origin used in Swedish have direct counterparts in English. The prefix un- (as in unpleasant), for example, corresponds directly to the Swedish prefix o- (as in otrevlig) (Liljestrand, 1993: 54). (3) See the introduction to this chapter (Section 3.1).
98 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(4) This word was invented to illustrate a fictitious situation in which bagel shops in Europe are increasing in number, and people might therefore say that Europe is being bagelized (Haspelmath & Sims, 2010: 71). (5) This was heard by the present author on an American reality show in which a man training to become a police officer was telling his siblings that as part of his training he would have to endure being mazed by his colleagues, with the man’s brother responding in a jovial manner that he would like to be the mazer. (6) This derivation was heard in an episode of the American TV series CSI in which a criminal displeased with his appointed lawyer tells the court that he wants to unappoint him. (7) A syntactic language is a language in which the majority of the words consist of a small number of morphemes (Andersson, 1987: 21–24). (8) Languages that have agglutinating characteristics add morphemes that are easily segmented to the base forms of words. In languages with fusional characteristics, morphemes that are not easily segmented are attached to the base forms of words (Andersson, 1987: 24, 25). (9) In only one case on the English test were the students asked to provide a prefix, which was in addition to supplying a suffix to the same stem (didn’t approve → disapproval). (10) It is the frequency of the non-inflected form of the word given that has been converged with the frequency of the word to be formed by the students; i.e. the frequencies of inflectional forms, such as the plural -s and the past tense form on regular verbs, have not been considered. (11) On a few occasions on both tests students gave more than one answer. If one of these was incorrect, no point was awarded. (12) Complacency (1) and complacement (2) were also considered to be correct answers. They are listed here for spatial reasons. (13) Glömskhet (3) was also accepted and is listed here for spatial reasons. (14) Other forms judged to be correct are säsongartat (1), säsongsaktigt (1), säsongsberoende (3), and säsongsmässigt (1) (=seasonal), presented here for spatial reasons. (15) Complacency and complacement are also possible derivatives. For spatial reasons these options are listed here. (16) It is interesting to note that for three of these four exceptions, the suffix -ness is involved. It may mean, as discussed in the text, that -ness is used as a default suffix when the students are unsure of what noun derivative to use. (17) In quite a few cases on the Swedish test, more than one derivative was considered to be correct. These are all listed in Table 3.7.
4 Idioms and Proverbs
4.1 Introduction
While students usually appear to appreciate the occasional treatment of L2 idioms, they also seem relatively reluctant to approach them in a more comprehensive and structured way, the main reason offered being that they have very rarely heard these expressions used by native speakers. Admittedly, idioms may not be as frequently used as other items of vocabulary. In Moon (1997), for example, studying the frequency of 6700 fixed phrases in the Oxford Hector Pilot Corpus, a corpus of British English consisting of 18 million words, it is shown that the number of occurrences of idioms and proverbs is indeed quite low. In fact, only slightly more than 11% of the idioms included in the study (1657 items considered) occurred 1–5 times per 1 million words, 8% 1–2 times, 3% 2–5 times and about 1% 5–50 times. Proverbs (276 items investigated) were found to be even rarer, with the vast majority occurring less than one time per 4 million words. These findings are also substantiated by Biber et al. (1999) in their analysis of the Longman Spoken and Written English Corpus, which consists of 40 million words. In their study, idioms occurred on average fewer than five times per one million words. At the same time, some researchers claim that idioms are omnipresent elements that need to be approached in the L2 classroom in systematic ways (Liu, 2008). The explanation for these two diametrically different perceptions may be due to the approach chosen when looking at the frequency of idioms in text (Liu, 2008). If, instead of investigating how frequent specific individual idioms are in a certain corpus, the number of idioms used per minute or per number of words is considered, a different picture will emerge. Cooper (1998), for instance, who studied idioms used on soap operas, detected approximately three idioms per minute. Pollio et al. (1977), studying different kinds of speeches, political debates, and psychotherapy sessions, detected an even higher rate of as many as 4.08 idioms per minute. Another reason for these diametrically different perceptions is that idiom use is highly register-sensitive. Generally speaking, it seems that idioms are more common in informal discourse, but this even varies according to the type of informal situation. Some subgenres of written language, usually emotive in character, also appear to produce more idioms than others. Moon (1997) mentions horoscopes (see also 99
100 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
McCarthy, 1992 and McCarthy & Carter, 1994) and sports reports as examples of such subgenres. In order to get a more hands-on perspective, the present author decided to write down all the idiomatic expressions encountered when reading and watching TV for a few weeks.1 The lasting impression of that mini investigation is that these items are prolific, but that, especially in spoken language, in order to notice them one needs to already know them. That is, due to the double semanticity of many of these items and the additional processing this entails, idioms often fly under students’ radar, giving the impression that they are comparatively uncommon. This interpretation is strengthened by the fact that many expressions are, to varying degrees, manipulated. Whereas some of them may be described as mere dents in a recently purchased aluminium can (systemic manipulations), others can be likened to a can discarded and processed beyond recognition, looking more like scrap metal (creative manipulations), with the latter type, if noticed at all, causing major comprehension problems (Karlsson, 2019). (For a more detailed discussion on manipulated idioms and L2 students’ understanding of them, the reader is referred to Karlsson, 2019, especially chapter 5.) This implies that idiomatic expressions do need to be addressed in the L2 classroom. 4.2 Theoretical Background and Previous Research
The present section will begin by offering the reader a definition of idioms and proverbs. It will continue by focusing on theories on L1 idiom processing and comprehension, as well as making a comparison between a first and second language in these respects. The third subsection will discuss the effects of contextual support, transparency, frequency and (cultural) sameness on idiom comprehension, while the final subsection will present a technique that may be beneficial when teaching and learning idioms. Idioms and proverbs defined
The main problem with the term ‘idiom’ is that it is very often used as an umbrella term for ‘all fixed phrases, clichés, formulaic speeches, proverbs, slang expressions, and, at the extreme, even single polysemic words’ (Liu, 2008: 3), making it ‘an ambiguous term, used in conflicting ways’ (Moon, 1998: 3). The present study, however, adheres to two definitions. In The Concise Oxford Dictionary (1985: 495), an idiom is defined as a form of expression peculiar to a language or person, peculiarity of phraseology approved by usage though having meaning not deducible from those of the separate words.
And in Moon (1997: 46), restricting the Oxford English Dictionary (OED) definition even further, idioms are regarded as
Idioms and Proverbs 101
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 gives several examples of what she, in accordance with her definition, considers to be classical idioms – for example, spill the beans, have an axe to grind and kick the bucket – some of which are comparatively straightforward to decode, such as bite off more than I can chew, while others are totally opaque, such as rain cats and dogs. The two definitions offered above together consider idioms (and all other multi-word items) along three separate continuums. The extents to which idioms adhere to these continuums are not absolute but vary in accordance with each and every item. The first continuum deals with noncompositionality, which has to do with the degree to which a literal reading of an idiom can help in interpreting its figurative meaning. The second continuum is institutionalisation, which focuses on the degree to which an idiom is conventionalised in a language based on its frequency. Thirdly, there is the continuum of fixedness, which describes the degree to which an idiom is frozen in a specific form. Idioms, which are considered to be lexically fossilised items since they cannot normally be interpreted on a word-by-word basis, are seen to be highly non-compositional items having a unitary semantic meaning. Furthermore, even though idiomatic expressions are comparatively infrequent items in certain registers, as described in Section 4.1, there is usually a consensus in a language community about what phrases should be interpreted as whole units of meaning, so idioms rate highly on the institutionalisation continuum too. Finally, idioms have long been regarded as relatively frozen units of meaning. This means that they are usually also rated high on the fixedness continuum. However, more recent research has shown that these kinds of expressions are perhaps not as frozen as previously suggested (see also Moon, 1997, 1998). In fact, ‘[M]uch due to corpus-based research, the view on idioms is shifting. Once regarded as frozen phrases with monolithic meaning, idioms are today recognised as surprisingly flexible phrases whose meanings may be unexpectedly complex’ (Gustawsson, 2006: 7). In Liu (2008: 36), for instance, quite a few examples are given of different types of idiom variation, such as set/start the ball rolling (verb variation), a skeleton in the closet/cupboard (noun variation), and a bad/rotten apple (adjective variation). See also Gustawsson (2006) and Moon (1998) for more examples. Proverbs, just like idioms, rate comparatively high on the institutionalisation and fixedness continuums (though they are quite semantically compositional) (Moon, 1998: 21), which makes it clear that the boundary between idioms and proverbs is not a black and white one. According to Fernando and Flavell (1981: 30), ‘[T]he relationship … is best pictured as 2 circles that overlap’, in which a proverb may be described as
102 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
[a] short pithy saying in common and recognized use; a concise sentence, often metaphorical or alliterative in form, which is held to express some truth ascertained by experience or observation and familiar to all. (OED)
An even more detailed definition, but one that still agrees with the one given above, is offered by Cignoni and Coffey (2000: 549): Proverbs are characterised by certain inherent characteristics. These are: (i) they constitute complete units of meaning, syntax and intonation, rather than being lexical units or clauses, (ii) within a given culture, past or present, they are either metaphysical statements about the human condition, or statements about the physical world around us or statements about how people live together, or recommendations about how one should live, either within oneself or in relation to others, (iii) they are associated with the acquired wisdom of society in general rather than being the thoughts or opinions of a particular person. Many proverbs display a number of other characteristics, notably metaphoric features, euphonic features such as alliteration or rhyme, and structural and semantic parallelism. These, however, are tendencies rather than defining features. The processing and comprehension of L1 and L2 idioms
Over the last few decades there have been several approaches to the processing and comprehension of idiomatic expressions. The present subsection aims to describe the development of the theoretical thinking behind some of the major approaches. Two of the first hypotheses that were proposed – the literal first hypothesis (Bobrow & Bell, 1973) and the figurative first hypothesis (Gibbs, 1980) – both held that the meaning of an idiom was stored in a separate mental idiom list. Bobrow and Bell based such conclusions on an experiment in which subjects were presented with either a number of sentences whose meanings were literal but ambiguous or a number of sentences containing idiomatic expressions that could be interpreted either literally or figuratively. An example of the former type is Mary fed her dog biscuits, which can either mean that Mary gave biscuits to her dog or Mary gave a female person dog biscuits; an example of the latter type is John and Mary buried the hatchet, which can either mean that John and Mary had a hatchet that they for some reason decided to bury or that John and Mary stopped fighting. The same two groups of subjects, irrespective of which sentence type they had been confronted with, were then presented with yet another sentence that could be interpreted both literally and figuratively. It turned out that those subjects who had been faced with a set of sentences that allowed an idiomatic interpretation offered an idiomatic answer, while those informants who had been faced with a set of sentences that allowed literal interpretations offered a literal answer. Bobrow and Bell (1973) concluded that the literal meaning is always
Idioms and Proverbs 103
accessed first, and that it is not until the literal interpretation has been rejected that learners start to think in a figurative way, which in turn means that literal meanings are always understood more quickly than idiomatic meanings. This was, however, contradicted in later studies. Although he maintained that idiomatic expressions were stored in a separate idiom list, Gibbs (1980) was able to show that literal meanings are not retrieved faster than figurative ones. His subjects were presented with sentences such as He is singing a different tune, which could be interpreted either literally (He is singing a different song) or figuratively (He has now changed his mind). The sentences were either given in context, which induced either a literal or figurative reading, or without context, so that the expressions themselves were either interpreted to be literal or figurative. After having encountered these types, the subjects were presented with a paraphrase that was either literal or figurative and asked to decide whether the paraphrase was correct or incorrect. According to Gibbs, the subjects’ response time showed that it took less time for them to understand the idiomatic uses of the expressions than their literal ones. Based on his results, putting forth the figurative first hypothesis, Gibbs thus claimed that native speakers rarely attempt literal readings of idiomatic expressions, for the simple reason that they recognise an idiom when they see it and so they can bypass the literal route and access its figurative meaning directly. Other studies at this time also pointed in the direction that idiomatic expressions were processed faster than literal ones (e.g. Ortony et al., 1978), or that they at least took an equal amount of time (e.g. McElree & Nordie, 1999). In contrast to the literal first hypothesis and the figurative first hypothesis discussed above, the simultaneous processing hypothesis, proposed by Swinney and Cutler (1979), suggests that literal and figurative expressions are accessed simultaneously, and that it is not until the context disambiguates their meaning that the learner lets go of one of the interpretations. Swinney and Cutler based their conclusions on an experiment in which subjects had to decide whether strings of words presented to them were meaningful English expressions or not. Of the 152 strings tested, 23 were of the break the ice type, that is, meaningful and grammatically correct strings of words that could be interpreted figuratively as well as literally. Another 23 were of the break the cup type, that is, grammatically correct literal expressions that had been formed based on idioms in which the key word had been changed. Yet another 30 strings were grammatically correct but non-idiomatic phrases, while the remaining 76 were ungrammatical strings of words. It turned out that the subjects in Swinney and Cutler’s study all recognised the idiomatic expressions faster than the non-idiomatic ones. According to Swinney and Cutler (1979), this happened because, in contrast to idiomatic expressions that are seen as units and therefore do not warrant further analysis, ordinary vocabulary items are, in addition to being analysed lexically, also analysed syntactically and semantically.
104 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Thus idiomatic expressions have an advantage in terms of processing speed since the whole phrase is treated as a single unit. Other studies, however, have shown that literal readings of idiomatic expressions appear to be activated and stay activated during the whole interpretation process. This is what is suggested by, for example, the idiom key hypothesis, in which the strict separation between literal and figurative meaning of idioms, as seen in the three models described above, is absent (Cacciari & Tabossi, 1988; Flores d’Arcais, 1993; Tabossi & Zardon, 1993, 1995). In accordance with this hypothesis, the processing of an idiomatic string of words begins literally until the string, at one specific point (i.e. the key point), is recognised as an idiomatic expression. The concept is an important one since there are fundamental differences between the key point of an idiom and the corresponding points of recognition of individual words and sentences (Tabossi & Zardon, 1993). In English, for example, several words start with the syllable pow, such as powder, power and pow-wow. For these words, the point of recognition comes directly after the first syllable; that is, having seen d, e or w, the learner knows which of the words is intended. In idiomatic processing, even when the key point has been spotted, the words that follow are not fixed to the same extent. For example, the phrase to be in seventh heaven does not have to end with heaven, even when the key point seventh has been detected. The intended phrase could equally well be to be in seventh place, for instance. Also, whereas individual words can sometimes be recognised before their points of recognition due to contextual information, idioms usually cannot since they are not context-dependent to the same degree (Cacciari & Tabossi, 1988). As mentioned above, the key word of an idiom and the point of recognition of a sentence also differ from each other. In Flores d’Arcais (1993), two main differences are discussed. Firstly, while the two interpretations of an ambiguous sentence usually entail two different grammatical analyses (e.g. Flying planes can be dangerous is analysed differently syntactically in To fly planes is dangerous as compared to Planes that fly are dangerous), the interpretations of an idiom (literal and figurative) do not. Also, in contrast to the meanings of an ambiguous sentence, the two meanings of an idiom are often related, the idiomatic meaning being an extension of the literal one (e.g. being in hot water). Finally, the key word of an idiom appears to be the most important part, which, in contrast to other parts, is not subjected to variation; that is, this is the truly fixed part of an idiom (Liu, 2008: 62). In contrast to the four processing theories discussed so far, a more recent theory referred to as the idiom decomposition hypothesis claims that the processing and comprehension of idiomatic expressions do not only involve normal language processing but that this processing occurs together with a pragmatic interpretation of the use of the idiom in its discourse context. This model was first proposed by Gibbs et al. in 1989, based on an experiment in which subjects were faced with idiomatic
Idioms and Proverbs 105
strings of words that were either semantically decomposable or totally non-decomposable. While semantically decomposable idioms are items in which the individual key words have meanings that contribute to the interpretation of the idiom as a whole, as in get the picture and miss the boat, non-decomposable idioms are items in which the meanings of the key words do not contribute, as exemplified by shoot the breeze. The results of the Gibbs et al. study were quite straightforward, showing that the subjects needed significantly less time to process decomposable expressions than those that could not be decomposed. This clearly suggests that learners first try to make sense of the individual parts, and not until they fail at doing so do they consider the expression an unanalysable unit. In an experiment by Titone and Connine (1999), who made use of the insights gained in the work of Gibbs et al., it was shown that it took longer to integrate the correct meaning into the idiomatic context for idioms that were non-decomposable than for those whose individual components contributed to the interpretation of the idiom as a whole, thus substantiating the findings of Gibbs et al. Other studies on idiomatic expressions (e.g. Cacciari & Levorato, 1989; Flores d’Arcais, 1993) and proverbs (Honeck & Temple, 1994) point in the same direction. The models described so far are all limited to the native mental lexicon. A theory that deals not only with the processing and comprehension of L1 idioms but also with idioms in a second language is the model of dual idiom representation. It has its basis in the work of Titone and Connine (1994) and was developed further by Abel (2003). The model is based on four assumptions, the first two dealing with lexical representation only. Firstly, the model puts forth that it is the degree of decomposability of an idiom that decides how the idiom is represented in a learner’s mental lexicon. Whereas an idiom that is non-decomposable requires a special idiom entry, an idiom that can be decomposed is represented via the entries of the constituents of which the idiom is made up. The evidence on which this assumption is based comes from research on compound words. Just like idiomatic expressions, these can be divided into those that are decomposable and those that are not. Studies focusing on two-part compounds (e.g. Sandra, 1990) show that for semantically transparent ones both constituents are accessed, while for semantically opaque ones only the first constituent is accessed, which indicates that the latter but not the former type must have a separate lexical representation. The same result was seen in a study by Zwitserlood (1994). Secondly, the model also suggests that it is the frequency of an idiom that decides whether an idiom entry is developed or not, so that the more frequent an idiom is, the more likely it is that an idiom entry will be formed. In addition, the model holds that this does not only occur with non-decomposable idioms but also with decomposable ones. Evidence of the existence of the frequency effect comes from research on morphology. Caramazza et al. (1988), for instance, detected a major difference between
106 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
frequent derivatives and inflected words on the one hand and those that were highly infrequent on the other hand. Whereas the frequent forms were approached as unseparated wholes, less frequent items were decomposed into stem and affix. In another study by Frauenfelder and Schreuder (1992) it was found that morphologically complex words were represented via both their constituent parts and their full entries, with the frequency (and transparency) of the word in question determining which was made use of first. If an item was comparatively infrequent, the stem and affix entries were quite consistently shown to be accessed first. If, on the other hand, an item was comparatively frequent, the full entry was accessed first, bypassing the stem and affix entries. (See also Chapter 3, where this phenomenon is discussed in more detail.) This is why idioms that are comparatively frequent will develop their own idiom entries, while infrequent ones will not develop idiom entries at all or will take a longer time to do so. Thirdly, the model of dual idiom representation assumes that if there are no lexical entries at the lexical level, which is very often the case for non-native speakers in the early stages of L2 learning, the processing and comprehension of idioms instead occur at a conceptual level. These conceptual representations are non-linguistic in character and instead involve pragmatic knowledge; that is, they occur at a more general cognitive level. This means that if no lexical information is available at the time of processing, learners will resort to conceptual metaphors, such as ‘anger is fire’, which would help learners understand idioms like smoke was coming out of his ears, she was spitting fire and he was fuming (Lakoff & Johnson, 1980). Finally, the fourth assumption deals with differences between L1/L2 processing and comprehension, and it posits that because L2 learners have not been exposed to L2 idioms to the same extent as native speakers, there are not as many idiom entries in their mental lexicon. L2 learners therefore have to rely on literal readings of decomposable as well as nondecomposable idioms to a higher degree than native speakers; that is, put differently, non-native speakers have to rely to a greater extent than L1 speakers on constituent entries and, if there are none of these, they have to resort to conceptual representations instead. This conclusion is based on two studies. In the Titone and Connine (1994) study mentioned above, native speakers of English were asked to judge whether a number of idioms were decomposable or not. The same subjects were also asked to rate the familiarity of the idioms on a seven-point scale on which one was the lowest. It was found that the native speakers judged 41.9% of the idioms to be decomposable in character and 58.1% to be non- decomposable. It was also found that the mean value for the familiarity of the decomposable idioms was 5.92, while the corresponding value for the non-decomposable idioms was 5.76; that is, decomposable expressions were rated as being more familiar than non-decomposable ones. The Abel
Idioms and Proverbs 107
study (2003), also mentioned above, includes two experiments performed in almost exactly the same way, only her subjects were L2 students whose first language was German. In Abel’s first experiment, the testees judged 56.6% of the idioms to be decomposable and 43.5% to be non- decomposable. In her second experiment, the percentages were 55.2% and 44.8% respectively; thus the results of the second experiment tally well with the results of the first one. As for familiarity, the mean value for the decomposable idioms was 4.90 and for the non-decomposable idioms was 2.99; that is, just like the native speakers, the non-native speakers rated decomposable idioms as more familiar than non-decomposable ones. When a comparison is made between the Titone and Connine study and the Abel study, it can be seen that the native speakers judged fewer idioms to be decomposable than did the non-native speakers – that is, it appears that many more idioms had received entries in the native speakers’ mental lexicon than what had occurred in the mental lexicon of the German subjects. This indicates that non-native speakers have a greater tendency to try to decompose idioms than do native speakers, who simply activate their many idiom entries. Moreover, in the Abel study, contextual and conceptual factors were found to play especially important roles when L2 decomposition occurred. In addition, when the mean values for familiarity are compared, it can be seen that, for native speakers as well as nonnative speakers, the more familiar an idiom is, the more likely it is that it is decomposed; familiarity appears to go hand in hand with decomposability whether one is a native speaker or not. Liu (2008: 74), who makes an attempt at summing up the differences between L1 and L2 idiom processing and comprehension, writes that L2 idiom comprehension appears to be a slower and much more complex process than that for L1. It involves the use of more strategies in terms of both type and quantity. The process does not seem to conform to any of the major L1 idiom comprehension models. Instead, it takes the form of a heuristic approach, a process in which L2 learners approach an unknown idiom as a problem and try to solve it on a trial and error basis by using a variety of strategies.
Several of these strategies are discussed in great detail in Liu. The most frequent one used by L2 learners when accessing an idiom appears to be to make use of the context in which the idiom is placed. This seems to be not only the most frequently used strategy but also the most effective one. In Cooper (1999), for example, making use of contextual information made up 28% of all the strategies considered (followed by ‘discussing and analysing’ (24%) and ‘using literal meaning’ (19%), accounting for 57% of all the correctly guessed idioms). (In the next subsection, the reader will gain even more information about the facilitating effect of contextual support.) A second strategy employed by L2 learners is to make use of their native language. This is naturally especially beneficial if the learner’s L1
108 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
has a counterpart that is highly similar to the L2 idiom. Several studies report such results. In Irujo (1986), for example, it was found that the performance (comprehension, recall and production) of Venezuelan learners of English was considerably better with L2 idioms that had direct counterparts in their L1 than those that did not. In Abdullah and Jackson (1998), investigating Syrian college students’ comprehension and translation of English idioms, similar results were seen. In addition, they were also able to show that the students achieved very low scores with so-called false cognates, that is, idioms that are identical in form but differ as to meaning. Finally, Abdullah and Jackson also noticed that whereas the students were comparatively good at translating cognate L2 idioms into their native language, they were more hesitant about translating the same idioms, when offered in their L1, into English. This led the researchers to draw the conclusion that learners appear to believe that idioms are culture-specific and therefore often avoid translating identical idioms into the L2, even in those cases where it is fully possible. This means that making use of one’s first language does not only achieve positive results but may also induce interference. Moreover, making use of L1 knowledge does not only mean making use of linguistic knowledge but could also mean making use of more general L1 cultural knowledge. Research has shown that if a learner’s native language tends to make use of idioms in a certain area of life, say food or animals, and the same area is prolific in the language the student is trying to learn, this helps in the comprehension and production of such idioms (Boers et al., 2004). The reverse is also true: idioms that are culture-specific, such as carry coals to Newcastle and kiss the Blarney stone, are generally difficult for learners to understand and remember. (The reader will learn more about the effects of an L1 in the subsection that follows.) Still another strategy often employed by L2 learners is to make use of even more general knowledge, that is, pragmatic knowledge or knowledge of the world. For instance, in the Abdullah and Jackson (1998) study already discussed above, it was shown that their subjects could draw conclusions about the meaning of the idiom to give someone the cold shoulder as they interpreted cold to mean unwelcome since cold is the opposite of warm (a warm welcome). Similarly, the same informants used their knowledge of the world when they interpreted the idiom to skate on thin ice to mean that the person doing this was doing something very risky. Research in this area has shown that world knowledge is a more useful tool when interpreting decomposable idioms than when dealing with idioms that cannot be analysed on a word-by-word basis (Abdullah & Jackson, 1998). Based on what has been said above, L2 learners have a well of different strategies at their disposal when encountering unknown L2 idiomatic expressions. Yet, as Liu (2008: 74) states,
Idioms and Proverbs 109
this heuristic approach does not seem to apply …, for when a person, be it a native speaker or L2 speaker, encounters a known idiom, a normal linguistic analysis may not be activated or may soon lead to a direct memory retrieval. Thus, a complete L2 idiom comprehension model needs to be a dual-process one, with the heuristic approach in charge of unknown idioms and direct memory retrieval being used for known idioms in most cases.
This makes L2 idiom processing and comprehension, as realised by the model of dual idiom representation, a much more complex phenomenon than the processing and interpretation of idioms in a native language. Idiom comprehension: The effects of contextual support, transparency, frequency and (cultural) sameness
With the help of four investigations, the present subsection will discuss the effects of contextual support, transparency, frequency and (cultural) sameness. In the first study, Karlsson (2019), 71 students between the ages of 13 and 18 were tested, and the simultaneous effects of contextual support, transparency, and frequency were explored. Two parallel tests, one in Swedish (native language) and one in English (second language) were constructed, each incorporating 27 test items along three continuums. Nine of the items were considered from a contextual perspective, with three offering no support, another three providing some support, and the last three giving full support. Another nine of the items were considered from a transparency perspective, where the learners’ L1 was incorporated as an important factor in the categorisation process. Three of these items were opaque in character, another three were categorised as semitransparent, and the remaining three were seen as fully transparent. Lastly, nine test items were considered from a frequency perspective, three of which were infrequent, another three of which were frequent, and the last three of which were very frequent. Accordingly, the 29 idioms tested displayed different degrees of all three aspects. The findings show clearly that contextual support is the main facilitator, and that transparency, albeit to a lesser degree, also has a role to play in learners’ L1 and L2 disambiguation processes irrespective of age group. The impact of frequency on the informants’ comprehension, on the other hand, was marginal, again applying to both languages. In Cieślicka (2006), studying native speakers of Polish, the English idioms included in the investigation were divided into three groups according to their degree of translatability (as categorised in Liontas (2002)). One group consisted of idiomatic expressions for which there were exact translation equivalents in the students’ L1 (referred to as ‘lexical level idioms’). Another group included idioms that could only partially be translated directly into the students’ L1 (‘semi-lexical level idioms’), and a third group was made up of idiomatic expressions for which no
110 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
corresponding idioms in the learners’ L1 existed (‘post-lexical level idioms’). In addition to investigating the L1 effect on the understanding and production of these L2 idioms, wanting to study the transfer effects of the L2 idioms and the strategies used by the learners when faced with the task of interpreting and explaining the L2 idioms in their L1, Cieślicka (2006) incorporated two additional aims. Hypotheses were then put forth. Firstly, it was assumed that the learners’ first language would have considerable effects on the students’ comprehension and production skills, and that ‘lexical level idioms’ would be the easiest ones to process and ‘postlexical level idioms’ the most difficult. Secondly, it was also assumed that the degree of agreement between the two languages would affect the students’ ability to translate L2 idioms into their native language. It was here suggested that ‘semi-lexical level idioms’, due to their partial similarity to idioms in the students’ L1, would cause the most interference, whereas hardly any interference would occur with ‘post-lexical level idioms’. Finally, it was assumed that, irrespective of idiom type, the literal meanings of the key words would affect the explanations of the meanings of the L2 idioms. Thirty-six second-year university students, divided into three groups, took part in the experiment. The first of these three groups was tested on their comprehension skills as well as their productions skills. The second group, on the other hand, was only subjected to a production test, whereas the third was only tested on their comprehension skills. The comprehension exercise, thus taken by Groups 1 and 3, tested 21 idiomatic expressions presented randomly to each student so that no two students would encounter the idioms in exactly the same order. When faced with an idiom, the students were asked to do three things. Firstly, they had to give the meaning of the idiomatic expression at hand. Secondly, they were asked to write down everything they could about their thoughts, reading strategies, and/or the mental images that popped up while they were processing the idiom. Lastly, the students were requested to offer the best Polish counterpart of the English idiom. The production exercise, which was taken by Groups 1 and 2 only, involved two different tasks. The first of these was a context-based gap-filling completion task, while the other one was a context-based translation task. The items produced by the learners were judged by a Polish–English bilingual to be ‘correct’, ‘incorrect’, or ‘incorrect involving transfer’. The results show that ‘lexical level idioms’ were the easiest ones, in terms of not only comprehension but also production, in that they elicited the lowest number of incorrect responses on both test types. This thus agrees with the research hypotheses put forth by Cieślicka. Furthermore, ‘semi-lexical level idioms’ were noticed to be particularly prone to transfer. In fact, this type of idiom elicited the highest percentage of transfer-based errors on the comprehension test as well as the production test. This also tallies with the hypotheses put forth by the researcher. However, even
Idioms and Proverbs 111
though ‘post-lexical level idioms’ were most difficult to interpret on the comprehension task, these posed fewer problems than ‘semi-lexical level idioms’ on the production tests, which contradicts the hypothesis. Cieślicka (2006) interprets this to mean that the Polish students did not refrain from translating opaque idioms literally even though the result may have sounded odd. Finally, whereas a number of different strategies were employed when processing the idioms, the most frequent ones all involved some kind of literal reading of the idiom, hence putting the key word in focus. Cieślicka concludes by saying that, as there were so many incorrect answers that were transfer-induced, learners’ native language appears to play a dominant role when processing idiomatic expressions in a second language, even for those expressions that are more opaque in nature. In Piasecka (2006), the main focus was also to find out to what extent an L1 (Polish) affects the comprehension and production of idiomatic expressions in an L2 (English). Ninety university students were presented with 23 Polish idioms, for which they were asked to provide L2 equivalents. They did so in two different types of tasks. One of the tasks was a contextualised gap-fill translation exercise, testing 13 Polish idioms all involving parts of the body. That is, the test focused on a source domain that both English and Polish make extensive use of. (The concept of source domains will be more thoroughly discussed in the next subsection.) The other task was a decontextualised discrete items translation exercise, testing 10 L1 idioms of various source domains. With the help of this setup, the researcher wanted to explore how advanced learners render L1 idioms in a second language, and whether such learners are likely to offer literal translations of L1 expressions in the L2. Moreover, the researcher also wanted to investigate whether there would be students who avoided translation of L1 expressions, and whether any of the students would avoid using L2 idioms altogether. The subjects’ answers formed five different categories. In some cases, an exact translation of the L1 idiom was offered. In other cases, only parts of the L1 idiom were translated, thus not achieving the correct target idiom. In still other cases, a totally different L2 idiom was used or an explanation was offered rather than a translation. Finally, in some cases no answer at all was produced. On the contextualised task, disregarding the ‘no answer’ category (26.81%), the majority of the students gave an exact translation of the L1 idiom (26.07%), which was followed by a non-idiomatic response; that is, an explanation was offered but without a corresponding L2 idiom (23.90%). A somewhat different pattern was observed on the decontextualised task. Again disregarding the ‘no answer’ category (47.55%), most students offered a partial translation of the L1 idiom (23.33%), followed by an exact translation (20.87%). It is interesting to note here that offering exact translations was a frequently used strategy on a task that involved a source domain that is prolifically used in both languages. It is also
112 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
interesting to note that the idioms for which the students gave the most correct responses were idioms that were most familiar and probably most frequently used in their L1 as well as their L2. Finally, in Trulsson (2007), dealing with Swedish learners’ knowledge of L1 idioms, 139 subjects were incorporated. These were divided into two groups, one consisting of 72 learners between the ages of 13 and 16 and the other of 67 people above the age of 50 (most between 50 and 65). Both groups were tested on the same 20 idioms, all of which were decontextualised and from the source domain of animals. For each item tested, four alternatives were offered, three of which were distracters. The results of the study show that, on average, 70% offered the correct expression and 27% the incorrect alternative, while 3% did not offer an answer at all. When the results were considered according to age, it was observed that the older subjects did considerably better than the younger ones. Whereas an average of 87% of the former group supplied the correct response (while 10% were incorrect and 3% gave no answer), only an average of 54% of the latter group did the same (while 43% were incorrect and 3% gave no answer). Age was therefore a determining factor as to whether the meaning of an idiom was known or not. This was also observed in Karlsson (2019), discussed earlier. Lastly, it should be pointed out that quite a few idioms in use in Swedish today have been borrowed from the English language. Examples of such English idioms are the ball is in your court, back to square one, put somebody/something on the map, be caught with your trousers/pants down and get/have cold feet (Moberg, 1996: 217). This borrowing has, according to Moberg (1996: 216), been made possible primarily due to the shared outlook on life in general. Pedagogical approaches to idioms
As has been discussed in the previous subsections, it is part of the very nature of many idioms that their meanings, based on their constituent parts, are comparatively idiosyncratic, at least superficially so. This means that the main task when approaching idiomatic expressions from a pedagogical point of view is, in order to make the meaning of these relatively fixed phrases more easily accessible to learners, to develop teaching and learning techniques that are more systematic in character. In research on idiom processing, it has been noticed that both identifying the source domain of an expression and offering its etymological background help students’ comprehension, retention, and production of these types of expressions. This finding has its basis in cognitive-semantic theory (e.g. Gibbs, 1994; Johnson, 1987), levels of processing theory (e.g. Cermak & Craik, 1979; Craik & Lockhart, 1972), and dual coding theory (e.g. Clark & Paivio, 1991; Paivio, 1986). According to cognitive-semantic theory, the meanings of a great many idioms are not arbitrary, but are, to varying
Idioms and Proverbs 113
degrees, quite motivated in that they can be traced back to a relatively small number of so-called source domains. For metaphoric expressions, these source domains can be divided into two main types. Firstly, there are those domains that are based on relatively straightforward spatial relations, such as up–down and in–out (Lakoff, 1990). These can be extended to more abstract source domains through conceptual metaphors. The source domains of the metaphors an above average IQ and he served under the president, for example, can thus be interpreted along the up– down source domain, the abstract source domain for the former being interpreted as ‘more is up’ (‘more’ equalling ‘above’) and the abstract source domain for the latter being interpreted as ‘power is up’. These source domains appear to be universal in nature and are therefore unlikely to cause problems for L2 learners. Secondly, there are those source domains that are culture dependent (Lakoff, 1987) and hence more likely to create problems for non-native speakers. For instance, idioms referring to boats and sailing, such as rock the boat, learn the ropes and show your true colours, seem to be especially prolific in British English due to Britain’s nautical history. An L2 learner of English whose L1 does not make use of this source domain to the same extent may therefore have a hard time dealing with these types of idioms. In fact, research shows that learners indeed have more problems with culture-dependent source domains than those that are more universal. In Boers and Demecheleer (2001), for example, it was observed that French-speaking learners of English found it more difficult to interpret the meanings of idioms from the source domain of boats and sailing than those found within the source domain of eating, simply because the former is not a domain seen in French to the same extent as the latter. It is here that levels of processing theory comes into play. If learners are trained in a more systematic way in identifying the source domains of idioms (culture-dependent or not), it is believed that their retention of these idioms will increase as the identifythe-source procedure is an operation that probably occurs at a deeper level than rote memorisation. Furthermore, if learners can be prompted to create mental pictures of idioms, which may be the result when explanations of their origin are offered, an extra route for recall will have been formed since the information is being encoded (and possibly stored) in a dual fashion. This will, according to dual coding theory, strengthen learners’ comprehension and retention of these idiomatic expressions. In two experiments, Boers et al. (2004) tested these hypotheses (see also Beréndi et al., 2008; Skoufaki, 2008; Zimmerman, 2006) by focusing on finding out if etymological elaboration is as effective for the memorisation of etymologically opaque idioms as for the memorisation of etymologically transparent ones (2004: 61). On the one hand, Boers et al. argue that, while etymologically transparent idioms are comparatively easy to visualise and thus more likely to induce dual coding, etymologically opaque idioms cannot be visualised as easily, and dual coding can therefore not
114 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
occur. This would indicate that etymological elaboration may not be as efficient for opaque idioms as it may be for idioms that are transparent in nature. On the other hand, Boers et al. (2004) also argue, since images can easily be created for etymologically transparent idioms, identifying their source domain does not involve the same amount of effort as when identifying the source domain of transparently opaque idioms. This reasoning indicates that etymological elaboration may be beneficial to the learning not only of transparent idioms but of opaque ones too. Boers et al. believed that the answer to this may lie hidden in the convoluted interplay of (at least) three dimensions: the likelihood of dual coding occurring, the degree of cognitive effort involved, and the degree of eagerness to learn by the individual student. The first experiment tested, on a voluntary self-study basis, Flemish students of English studying at various levels at a tertiary education college. The students were divided into one control group and one experimental group, both of which were given computerised multiple-choice exercises testing the students’ receptive skills according to the students’ educational level. The control group was given three potential meanings of a specific idiom and then asked to pick the correct one. The experimental group was also given multiple-choice tasks, but they were instead asked to identify the source domain of the idiom in question, after which they received feedback about the origin of the idiom. One week after having taken these multiple-choice tests, both groups did the same gap-fill exercise intended to test more active knowledge of the same idiomatic expressions. The conclusion, according to Boers et al., would be that if the students in the experimental group experienced more difficulties than the students in the control group when asked to produce the last word of an earlier tested idiomatic expression for which they knew the source domain, and thus had been able to create a mental image, etymological elaboration would not be a useful tool when teaching students idiomatic expressions. It turned out that whereas each idiom was remembered by 39% of the students in the experimental group, only 28% did so in the control group. This meant that for 65% of the idioms tested the score of the experimental group was higher than the corresponding score achieved by the control group; that is, etymological elaboration did seem to have an effect on the retention of idiomatic expressions. However, when the results for opaque idioms only were investigated, the same positive result could not be detected, which may indicate that there is no point in offering students etymological explanations for idioms whose origins are so opaque that it is impossible for the students to retrace them themselves. Boers et al. (2004) felt that the design of the experiment may have been responsible for this result in that it did not create enough motivation for the students to find out about the historical origins of the idioms tested. In order to gain a better understanding of why the source domains of certain idiomatic expressions were more difficult to identify than others,
Idioms and Proverbs 115
the researchers looked more thoroughly at the idioms with high error scores. According to Boers et al., four factors may have prevented the students from being able to process the idioms at a deep level and thus played a significant role in the students’ inability to retain their meanings. Firstly, there were expressions in which the key word was highly likely to be misinterpreted and thus assigned the wrong source domain. This happened, for instance, with the expression to follow suit, belonging to the source domain of card games. Students may here have wrongly assumed that the word suit had its most common meaning, belonging to the source domain of clothing. However, after having been given etymological information about these types of idioms, the students in the experimental group did better on the productive task than the students in the control group. Secondly, there were also idioms that contained a key word whose meaning was totally unknown. This was often the scenario with key words of low frequency, such as loggerheads and tether used in to be at loggerheads and to be at the end of one’s tether. When encountering such idioms, the students probably resorted to guesswork and thus little cognitive effort was involved. Yet, as with the idiom type discussed above, when performing the gap-fill task, the students in the experimental group did better on these idioms than the students in the control group. According to Boers et al., it may be that the experimental students were intrigued by these unknown words and therefore, when given etymological information, created mental images. Thirdly, there were idioms containing key words whose literal use is obsolete. For example, the use of the words doldrums and shrift seem to be limited to the expressions to be in the doldrums and to get short shrift. The difficulty involved in such cases is substantiated by the fact that the students in the experimental group did not do better than the students belonging to the control group even though the former students received feedback about the source domains. Finally, there were idioms for which the students had to possess cultural knowledge in order to be able to understand their meaning. To hear something on the grapevine and to give someone the cold shoulder both belong to this type. In order to find out the extent of the negative effects of these four factors, Boers et al. set up a new experiment in which, in contrast to the previous one, the tests were given in class, involved more testees and more test items, and informed the participants which idioms they had given the incorrect answers for. This new set-up yielded some interesting results. For example, in the majority of cases (66.6%), students who indicated the correct meaning of an idiom on the comprehension task were also able to identify the idiom’s source domain. However, as pointed out by Boers et al., being able to identify the source domain of an expression does not entail that its meaning has been fully understood. For instance, even if the source domain of being in the saddle is identified as horse riding, it does not automatically lead to its exact meaning, namely ‘being in control or having power’. Interesting results were also achieved on the gap-fill
116 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
exercise that which tested the students’ recall. Unsurprisingly, students scored the highest on the production task with idioms for which they had been able to both pick the right meaning on the comprehension task and indicate the right source domain. As soon as the source domain was unidentified but comprehension was still accurate, the score dropped. The third highest score was achieved for idioms for which the students failed to offer the right meaning but succeeded in finding the right source domain. Finally, the students scored the lowest on idioms for which they failed on both receptive tasks. The most interesting thing with these results, however, is that, in contrast to the first experiment, etymological elaboration not only appeared to have a great impact on transparent idiomatic expressions (i.e. expressions whose source domains are easy to identify) (68.68% correct answers) but also had a similarly strong effect on the retention of idiomatic expressions that are opaque in nature (i.e. idioms whose origins are hard to retrace) (67.96% correct answers). To sum up, whereas etymological elaboration may not be able to help students with idioms categorised as opaque in a learner autonomy situation (as in the first experiment), in contexts of explicit instruction (as in the second experiment) elaboration in the form of etymological origin and source domain appears to be highly beneficial to the retention of transparent as well as opaque expressions. This means that by presenting students with mental imagery, the meaning of an idiomatic expression is made less arbitrary, and the mnemonic effect of this may be not only that the meaning of the expression is remembered more easily but also that students transfer receptive knowledge into productive knowledge. (For more on the effects of multimodal and visualisation techniques on L2 learners’ comprehension, the reader is referred to Karlsson, 2019, especially Chapter 3.) 4.3 The Parallel Tests Used for the Present Chapter
The parallel tests (one in English and one in Swedish) dealing with idioms and proverbs are both divided into three parts. Whereas Part A tests the students’ knowledge of general idiomatic expressions (50 test items, one point for each correct answer), Part B (30 test items, again one point for each correct answer) focuses on idioms that include animals as part of their metaphor. This source domain was chosen as idioms in this area appear to be incredibly prolific in both English (British and American) and Swedish. The reader is here reminded that studies have shown that a shared outlook on life has positive effects on L2 idiom comprehension, and it was therefore thought to be of great interest to investigate the Swedish students’ mastery of idioms in one such area. So as to achieve a situation that would correspond to what students experience in real life, almost all of the English idioms were randomly chosen from a list put together by the present author while watching TV, reading fiction, and listening to audio books. The rest were randomly
Idioms and Proverbs 117
picked from Collins Cobuild Idioms Dictionary (2002), which is based on the Bank of English and differentiates between frequent items, making up one third of the dictionary, and less frequent items. The Swedish idioms on the other hand were selected, again randomly, from three books (Jansson, 2006; Luthman, 2002, 2006), which all include extensive lists of Swedish expressions and their usage. On both tests, on both Parts A and B, the idioms were offered to the subjects in context with the expressions presented in bold. While the context for the English idioms was in the majority of cases created with the help of the BNC, and often simplified by the present author, the context for the Swedish idioms was taken, and only occasionally simplified, from the three books mentioned above. The students were asked to translate the English idioms into Swedish, either by providing a direct L1 counterpart or by explaining its meaning in their own words (the latter of which was what the students did for the Swedish idioms). Examples (1) and (2) are from Part A on the English test; (3) and (4) are from Part B of the same test. (1) to pass the buck = ____________________ (2) to paint the town red = ____________________
(3) to let the cat out of the bag = ____________________ (4) there are no flies on her = ____________________
You are responsible for the child at school and not the parents so don’t be tempted to pass the buck. Inside, the hotel is comfortably furnished and the hotel bar, which has TV, English video and video games, provides a lively and popular meeting place before painting the town red! You know something’s going on, we have a little surprise set up for you, and I think the time has come to let the cat out of the bag. I know there are no flies on her but I wonder how she found that out.
Furthermore, on both tests the students were requested to give information about the degree to which they thought they knew the idiomatic expressions. The students could either indicate that they did not know the idiom, that they recognised the idiom and thought they knew it, or that they were sure that they knew what the idiom meant. In addition, the students were also asked to fill in information about the extent to which they had used the context in which the idioms were offered. They could indicate either that they did not know the expression but had guessed its meaning with the help of the context provided or that they recognised the idiom but had still made use of the context when supplying its meaning. This forms the continuum seen in (5) below.
118 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(5)
to put paid to a thing = ____________________
The accident put paid to John’s thoughts of racing, but also rekindled his interest in the guitar.
□ I don’t know this expression. □ I’m guessing the meaning of the expression from the context of the sentence. □ I recognise this expression, but I don’t know what it means. □ I recognise this expression and I’m guessing its meaning from the context of the sentence. □ I recognise this expression and I think I know what it means. □ I am sure I know what the expression means. Part C of the tests focuses on the students’ knowledge of proverbs (20 proverbs on each test, yielding one point for each correct answer), all of which were tested in a decontextualised form. The L2 items were exclusively picked, in a random fashion, from a comprehensive dictionary listing, explaining, and comparing proverbs from English, French, German, Spanish and Swedish (Martinsson et al., 2004). Whereas some of the Swedish test items were also randomly selected from this dictionary, most were taken from (again randomly) a more practically oriented, but comprehensive, book of both idioms and proverbs (Edegran & Lönneborg, 2005). Two of the proverbs included in the English test are given in (6) and (7). (6) Once bitten, twice shy. = _______________________________________________ (7) One man’s meat is another man’s poison. = _______________________________________________
As in Parts A and B, the students were asked not only to translate the meaning of the proverbs (again either by giving a corresponding Swedish proverb (English test) or by explaining its meaning in their own words (English and Swedish tests)) but also to provide information about the extent of their knowledge. The students could indicate either that they did not know the proverb, that they recognised it but did not know what it meant, that they recognised it and thought they knew what it meant, or that they were sure that they knew what it meant. This is exemplified in (8). (8) The world’s coin is ingratitude. = _______________________________________________
□ I don’t know this proverb. □ I recognise this proverb, but I don’t know what it means. □ I recognise this proverb and I think I know what it means. □ I am sure I know what the proverb means. On both tests the students were also asked to evaluate the difficulty level of the test part at hand. As described in the previous chapters, the options available were ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’.
Idioms and Proverbs 119
Moreover, all the idioms and proverbs were presented in order of frequency in their respective parts, starting with the most frequent item in each case. These frequencies were based on searches in the BNC and Språkbanken. (See Chapter 1 for a more detailed description of these corpora.) However, it needs to be pointed out that searching for idioms in corpora is a very tricky business (Gardner, 2007: 260). One reason for this, as discussed in previous subsections, is that many idioms are not completely fixed (Liu, 2008). Thus, due to time constraints, the search performed in the present study was restricted to the idioms’ canonical forms. Before analysing the students’ results, the reader also needs to be informed that the idioms and proverbs were divided into three categories: transparent items, semi-transparent items, and opaque items. In addition to taking the students’ knowledge of the world into account, the degree of transparency of the English items was considered from a Swedish secondlanguage-learner perspective and based on two main criteria. Firstly, the idioms and proverbs for which exact literal and semantic translations could be given – that is, items for which cultural sameness was detected – were generally categorised as transparent items. Just a few idioms/proverbs were of this type. More items could be seen for which either an exact literal translation or an exact semantic translation could be given. These were categorised as transparent or semi-transparent items. (This criterion was for obvious reasons not considered when the degree of transparency of the Swedish test items was analysed.) Secondly, the idioms and proverbs for which literal readings helped understand their figurative meaning to various degrees were again generally classified as either transparent or semitransparent items. This criterion was of course also considered for the categorisation of the Swedish test items. However, here another factor comes into play for the L2 items. As pointed out in the theoretical background, certain key words, due to being highly infrequent or obsolete or both, may make the meaning of an expression more diffuse for a second language learner than for a native speaker. Items containing such key words were therefore categorised as either semi-transparent or opaque. Although it was considered, for obvious reasons this key word effect did not have an equally great impact when categorising the Swedish test items. For example, wear one’s heart on one’s sleeve (Part A), all his geese are swans (Part B) and there’s no place like home (Part C) were all categorised as fully transparent items. As for wear one’s heart on one’s sleeve and all his geese are swans, in addition to not containing any infrequent and/or obsolete key words, literal readings of these idioms give great help in explaining their figurative meaning. When it comes to the proverb, even though there is no direct literal translation, there is a direct semantic translation in Swedish – borta bra men hemma bäst – which appears to be not only commonly known but also comparatively frequently used. He mints gold (money) (Part A), a fly in the ointment (Part B) and in the eyes
120 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
of the lover, pock-marks are dimples (Part C) were all categorised as semitransparent items. In all three, there are key words – mints, ointment and pock-marks – that are either infrequent or obsolete, thus preventing the expressions from becoming fully transparent. Tongue in cheek (Part A), a red herring (Part B) and age and wedlock tames man and beast (Part C), finally, are all examples of test items that were judged to be opaque to the L2 learners. As for the former two, exact translations are not possible, nor do literal readings help the students to any great extent. When it comes to the proverb, the meaning is slightly more transparent due to its decomposability, but the key word wedlock is not very frequently used as well as rather obsolete and, together with the fact that there is no commonly used Swedish counterpart, it was thus judged to diffuse the meaning of the proverb to the degree of making it fully opaque. Tables 4.1 and 4.2 show the distribution of transparent, semi-transparent, and opaque items on the two tests used in the present study. Also, as discussed in the theoretical background (Section 4.2), context appears to play a great role in the L2 disambiguation process with idioms. Nevertheless, the context offered in connection with the items tested was not considered in the categorisation process as context is not part of the idiom itself. This does not mean, however, that context will not be considered at all; its impact will be discussed in detail in the results section (Section 4.4). Finally, as in previous chapters, in order to be able to make as fair a comparison as possible between the students’ L1 and L2 results, the total frequencies of the English and Swedish test items were contrasted, showing that the L1 test items are in total slightly more frequent (constituting Table 4.1 The number of transparent, semi-transparent, and opaque idioms and proverbs on the English test English test
Transparent
Semi-transparent
Opaque
General idioms (50 items)
3
20
27
Idioms with animals (30 items)
7
6
17
Proverbs (20 items)
11
8
1
Total
21
34
45
Table 4.2 The number of transparent, semi-transparent, and opaque idioms and proverbs on the Swedish test Swedish test
Transparent
Semi-transparent
Opaque
General idioms (50 items)
8
24
18
Idioms with animals (30 items)
3
13
14
Proverbs (20 items)
14
6
-
Total
25
43
32
Idioms and Proverbs 121
≈0.001031% of Språkbanken) than those forming the L2 test (making up ≈0.000741% of the BNC), as are the test items in Parts A and B separately (Part A: L1: ≈0.000765%, L2: ≈0.000537%; Part B: L1: ≈0.000229%, L2: ≈0.000156%). The Swedish test may hence be regarded as somewhat easier than the English test. (On Part C, the L2 test items were somewhat more frequent (≈ 0.000048%) than the L1 ones (≈0.000036%).) It should be added that the distribution of transparent, semi-transparent, and opaque test items may also have favoured the L1 test (see again Tables 4.1 and 4.2). None of the L1–L2 differences in frequency were confirmed statistically, which means that comparisons may be made between the parallel tests/test parts. Still the differences will be taken into consideration when discussing the results. 4.4 Results and Discussion
In the present section, the informants’ results on the two parallel tests will be presented. In the first subsection, the students’ mean scores and their evaluation of the difficulty level of the two tests are offered. The subsequent subsection explores the participants’ results on the individual idioms and proverbs. Here the effects of transparency, frequency and the use of context are in focus. The last subsection, finally, investigates the students’ individual results and makes comparisons between their L1 and L2 scores. The students’ results
Table 4.3 offers the students’ results on the tests as a whole as well as on each test part. The table shows that the native speaker surpassed the Swedish students on all three parts of the English test and thus on the test as a whole (native speaker: 82; Swedish learners (mean): 50.80). This agrees well with the model of dual idiom representation, which claims that fewer idiom entries will have had time to develop for the L2 learners due to less exposure. Furthermore, unsurprisingly, the Swedish students performed significantly better on the L1 test (mean: 82.73) than the L2 test (mean: 50.80), again agreeing with the model of dual idiom representation. (The L1 results also tally well with the results of the Trulsson (2007) study in which the Swedish subjects tested on L1 idioms had a correctness rate of 70%, and where, when age was considered, older speakers achieved 87% correct answers and younger speakers 54% correct answers.) If the test parts are considered separately, the informant group also did better in Parts A and B on the Swedish test (mean: 43.20 and 26.40 respectively) than in the corresponding parts on the English test (mean: 22.00 and 15.40 respectively). All these results are statistically confirmed at a
122 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 4.3 The students’ results on the idiom/proverb tests used in the present study Test part(s)
Students
N
Correctness rate
Mean
SD
Standardised scores
Parts A + B + C (general idioms + idioms with animals + proverbs)
Native speaker of English
1
82.00% (=82/100)
-
-
-
Swedish students – English test
15
50.80% (=762/1500)
50.80
16.02
Highest: 1.20 Lowest: −2.17
Swedish students – Swedish test
15
82.73% (=1241/1500)
82.73
9.41
Highest: 1.09 Lowest: −2.20
Native speaker of English
1
88.00% (=44/50)
-
-
-
Swedish students – English test
15
44.40% (=333/750)
22.20
7.66
Highest: 1.28 Lowest: 2.24
Swedish students – Swedish test
15
86.40% (=648/750)
43.20
5.05
Highest: 0.95 Lowest: −2.02
Native speaker of English
1
76.67% (=23/30)
-
-
-
Swedish students – English test
15
51.33% (=231/450)
15.40
6.25
Highest: 1.53 Lowest: −1.82
Swedish students – Swedish test
15
88.00% (=396/450)
26.40
3.02
Highest: 1.19 Lowest: 2.45
Native speaker of English
1
83.75% (=67/80)
-
-
-
Swedish students – English test
15
47.00% (=564/1200)
37.60
13.52
Highest: 1.36 Lowest: 2.12
Swedish students – Swedish test
15
87.00% (=1044/1200)
69.60
7.75
Highest: 1.08 Lowest: −2.27
Native speaker of English
1
75.00% (=15/20)
-
-
-
Swedish students – English test
15
66.00% (=198/300)
13.20
3.19
Highest: 1.51 Lowest: −1.94
Swedish students – Swedish test
15
65.67% (=197/300)
13.13
2.36
Highest: 1.22 Lowest: −2.18
Part A (general idioms)
Part B (idioms with animals)
Parts A + Part B (general idioms + idioms with animals)
Part C (proverbs)
significance level of 5%. The reader is, however, reminded that the English test items, on the whole as well as in Parts A and B separately, were less frequent than the Swedish ones, which may have helped widen the L1–L2 gap. In Part C, on the other hand, the students performed just as well with the English proverbs as the Swedish ones; that is, no statistical confirmation of a difference could be obtained here. The fact that Part C of the Swedish test was the only part where the total L1 frequency was lower than the L2 frequency, making the Swedish proverbs slightly more difficult than the English ones, may have played a part in this result. Confirmed statistically at a significance level of 5%, the Swedish students also did better on the idioms containing animals (mean: 15.4 (out of 30)) than on the idioms that displayed a mix of different source domains
Idioms and Proverbs 123
(mean: 22.0 (out of 50)). This may be due to the fact that animals form a domain that Swedish learners feel at ease with since, as pointed out earlier, it is a prolific domain not only in the English language but in Swedish too (see the discussion of Boers et al. (2004) in Section 4.2). (Compare this with Cieślicka’s (2006) study, which showed that literal translations (from L1 into L2) were very frequent when idioms with a shared source domain were encountered.) No such difference was ascertained statistically in the learners’ L1. It is also worth noting that the results achieved by the native speaker on the one hand and the Swedish speakers on the L1 test on the other hand are remarkably similar, on the three test parts as well as the tests as a whole. It may be that comprehension of idioms and proverbs in English and Swedish develop in a very similar fashion due to the cognate nature of the two languages. This will be explored further in the last part of the this section. The students’ self-evaluation aligns well with what they actually achieved on the tests. Whereas none of the Swedish participants thought that Parts A and B on the English test were ‘very easy’ or even ‘easy’, the majority of the informants thought so for the corresponding test parts on the Swedish test. The perceived L1–L2 difference is also confirmed in the correctness rates for items indicated as known (85.7% on the L2 test (the bottom of Table 4.4) versus 94.1% on the L1 test (the bottom of Table 4.5)) and items that were thought to be known (70.1% on the L2 test (the bottom of Table 4.4) versus 90.2% on the L1 test (the bottom of Table 4.5)). For Part C, on which the students did equally well in both languages, the majority indicated for both tests that the part was only ‘average’ in difficulty. Results on the individual idioms and proverbs
Tables 4.4 and 4.5 present the students’ results on Parts A (general idioms) and B (idioms with animals) for each idiomatic expression on the English and Swedish tests respectively; Parts A and B have been merged into one table for each language. 2 In both cases, the idioms are presented according to the number of correct answers, with the highest-scoring item placed first. (When two or more items received the exact same score, the one for which the students showed the greatest accuracy in their evaluation of their knowledge is placed first.) Furthermore, in both tables the idioms are colour-coded. Those that were judged to be of a transparent nature, according to the criteria discussed in Section 4.3, are in dark grey, those that were categorised as semi-transparent are in light grey, and those that were considered to be opaque in character remain in white. The students’ evaluation of their knowledge of the idioms is also included. For example, Idiom B30 (they are birds of a feather) is the
124 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 4.4 The students’ results on and evaluation of Part A (general idioms) and Part B (idioms with animals) on the English test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No. B13 B30 A25 A33 B6 A13 B12 B4 A19 A9 A50 A37 B7 A39 A38 B10 B28 A45 A5 A18 B14 A16 B15 A20 A30 B9 B16 A23 B24 A10 B17 B29 A12 A22 A42 B26 A26 A32 B22 A24 A44 A43 B20 A8 A14 B18 A29 A4
English idiom given in context (correct answers) look like something the cat dragged in (15) they are birds of a feather (14) paint the town red (13) he’ll get an earful from me (13) let the cat out of the bag (13) get a word in edgeways (13) he wouldn’t say boo to a goose (13) let sleeping dogs lie (13) as mad as a hatter (12) turn the tables on a person (12) wear one’s heart on one’s sleeve (12) he is a square peg in a round hole (12) count one’s chickens before they (are) hatch(ed) (12)
a busman’s holiday (12) cut/hurt a person to the quick (12) put a bee in a person’s bonnet (12) sell a person a pup (12) he mints gold (money) (11) pass the buck (11) bring down the house (the house down) (10) a fat cat (10) through the grapevine (10) it’s raining cats and dogs (9) every cloud has a silver lining (9) be a turncoat (9) gone to the dogs (9) birds of a feather flock together (9) as safe as houses (9) he thinks he’s the bee’s knees (9) when the chips are down (9) fine feathers make fine birds (9) that cat won’t jump (9) run (make) rings round a person (9) not on your Nelly! (9) get in Dutch with (9) play the giddy goat (9) put a person’s nose out of joint (9) get one’s comeuppance (8) get a person’s goat (8) cross one’s t’s and dot one’s i’s (8) has it at his fingers’ ends (7) give it a double take (7) all his geese are swans (7) be the worse for wear (6) take the mickey out of a person (6) have bats in the belfry (6) the pot is calling the kettle black (6) it sets my teeth on edge (6)
KN 7 2 4 1 1 1 1 1 4 2 4 (2) 1 1 1 (1) 2 2 (1) 2 1 4 3 1 1 1 1 1 2 (1)
1 1 1 (1) 1 2 (1)
TKR 3 1 2 3 3 3 3 1 3 (1) 8 (1) 4 4 (1) 3 1 1 1 1 2 4 (1) 1 6 4 4 (1) 4 (1) 2 2 1 (1) 3 2 1 1 1 1 1 (1) 1 1 1 3 (1) 2 (1) 1 1 (1) 1 1 2 (2)
RC 2 3 2 4 3 2
NKR 1 1
5 (1) 3 2 2 2 2 (1) 2 2 3 (1) 1 2 6 (2) 3 (1) 3 2 (1) 2 3 (1) 3 (1) 1 3 (1) 1 2 (1) 2 2 (2) 3 4 (1) 2 (1) 1 2 (1) 1 1 (1) 3 1 2 1 4 (3) 1 (1) 2 (1)
1 1 1 1 1
1 1
1
NKC 2 8 (1) 5 7 (2) 5 6 8 (1) 8 (1) 3 (1) 2 (1) 4 (1) 5 6 9 9 (1) 8 11 (1) 10 (3) 6 1 4 (1) 3 (2) 2 (2) 3 (2) 5 (2) 6 (2) 6 (2) 7 (2) 7 (1) 7 (1) 8 (5) 6 5 (2) 6 (2) 6 10 (3) 7 (1) 5 5 (1) 6 (1) 8 (5) 7 (2) 6 (3) 2 (1) 8 (4) 7 (4) 7 (2) 8 (3)
NK 1 1 2 1 1 1 2 2 2 2 1 2 1 2 1 4 (1) 2 2 2 2 3 2 2 4 3 1 3 3 3 4 (1) 3 4 6 6 (1) 3 2 5 3 8 2 2 6 2
(Continued)
Idioms and Proverbs 125
Table 4.4 (Continued) No. English idiom given in context (correct answers) A1 put paid to a thing (6) A7 a flash in the pan (5) B23 get one’s monkey up (5) A11 he is a chip off (of) the old block (5) B27 raise snakes (5) A15 sit (be) on the fence (4) B1 a red herring (4) B5 send a person away with a flea in his ear (4) A41 don’t teach your grandmother to suck eggs (4) B21 dressed up like a dog’s dinner (4) B2 a fly in the ointment (4) A34 pushing up (the) daisies (3) A48 look as if butter would not melt in one’s mouth (3) A28 he blows hot and cold (3) A46 he’s seventy (years of age) if he’s a day (3) A6 pay (give) lip service to (3) A31 be quids in (3) A40 … and Bob’s your uncle! (2) A3 tongue in cheek (2) A49 play gooseberry (2) A36 he has kissed the Blarney stone (2) B11 go to see a man about a dog (2) B19 the bulldog breed (2) A27 as keen as mustard (2) B8 there are no flies on her (1) A47 it’s all in a day’s work (1) B25 it’s a dog’s breakfast (dinner) (1) B3 have a bee in one’s bonnet (1) A2 give tit for tat (1) A17 give a thing (person) a wide berth (0) A21 go Dutch (0) Total Correctness rate
KN 1 (1) 1 (1)
2 1
3
1 (1) 1 2 (2) 2 (2)
TKR 4 (2) 1 2 (1) 1 (1) 4 (3) 2 (2) 2 (2) 1 (1) 1 1 (1) 2 (2) 1 (1) 1 2 2 (1) 1 (1) 4 (4) 4 (4) 1 (1) 1 (1)
63 (9) 85.7%
137 (41) 70.1%
RC
NKR
1 2 (1)
2 2 (1)
2 (1) 1 1 2 (1) 1 (1) 1 (1) 2 (2) 2 (1) 1 (1)
2 2
1 (1) 2 (2) 1 (1) 2 1 2 (1) 2 (1) 3 (3) 2 (2) 2 (2) 2 (2) 2 (2) 141 (48) 66.0%
2 1 1 1
2 2 1 1 1 (1) 1 1 32 (2) 93.8%
NKC 11 (6) 5 (2) 6 (4) 3 (1) 9 (5) 7 (6) 4 (3) 4 (3) 9 (7) 6 (4) 8 (5) 1 (1) 9 (8) 9 (7) 3 6 (4) 1 4 (4) 6 4 (4) 4 (4) 4 (3) 2 (2) 7 (7) 6 (6) 2 (2) 6 (5) 7 (6) 7 (7) 1 (1) 6 (6) 457 (192) 58.0%
NK 2 5 4 5 4 2 4 (1) 7 (1) 4 3 4 8 1 2 8 5 (1) 10 (1) 7 3 5 5 6 11 (1) 1 7 (1) 5 (1) 6 3 4 11 5 271 (10) 96.3%
expression that received the second highest score, 14 of the 15 students having supplied the correct answer. The idiom was also categorised as a transparent idiom, so the idiom itself is in dark grey. Also, two of the students indicated that they knew the idiom (column KN), while only one stated that it was recognised and thought to be known (column TK). Another three students indicated that the idiom was recognised and that they had used the context to guess its meaning (column RC), while another eight students wrote that they did not know the idiom but had used the context to guess its meaning (the plural colour-coded in grey, as these students were in the majority), seven of whom were successful (column NKC). The students’ mean scores, discussed in connection with Table 4.3, are of course also reflected in the results for the individual idioms. Whereas in the English material only eight expressions (out of 80) received the top three scores (15–13) (see Table 4.4), 59 of the Swedish expressions (also 80 in total) received the same top three scores (see Table 4.5). Similarly, whereas 15 idioms received 0, 1, or 2 correct answers on the L2 test, none of the idioms on the L1 test received the same low scores, the lowest here being 7 points. Furthermore, and more importantly, Table 4.4 shows that there is a concentration of transparent and to a lesser degree semi-transparent items among those test items that received the highest scores. In fact, none of the
126 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
eight English items that received the top three scores (15–13) were opaque in character. The reverse is also true; that is, in the bottom of Table 4.4 there is a clear concentration of idioms that are opaque in nature. Of the 15 idioms that received the three lowest scores (0–3), none are fully transparent and only three are semi-transparent. Also, it is not the case that the Table 4.5 The students’ results on and evaluation of Part A (general idioms) and Part B (idioms with animals) on the Swedish test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, RC = the item is recognised and its meaning guessed based on the context given, NKR = the item is recognised but its meaning is not known, NKC = the item is not known but its meaning is guessed based on the context given, NK = the item is not known). 3 Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No B3 B12 A35 A48 A10 A13 A21 A24 A1 A19 B11 B27 A22 A32 B2 A29 B24 A45 A50 A17 A11 B23 B14 B25 A3 B7 A30 A33 A28 A9 B21 A16 A47 A5 A34 B17 B10 A15 B30 B20 B18 B15 A49 B28 B9 A40 A6 A8 A37 A7
Swedish idiom given in context (correct answers) ha fjärilar i magen (15) (=have butterflies in one’s stomach) en hund begraven (15) (=something fishy) bli tagen på sängen (15) (=become very surprised) vara bakom flötet (15) (=be daft) ta sig i kragen (15) (=get a grip on oneself) ingen dans på rosor (15) (=not be all beer and skittles) lägga på ett kol (15) (=get a move on) gå åt som smör i solsken (1) (=sell like hot cakes) kronan på verket (15) (=the icing on the cake) sopa under mattan (15) (=sweep under the carpet) vara spindeln i nätet (15) (=be the one in charge, behind something) göra en groda (15) (=make a blunder/howler) lägga på is (15) (=put on ice) så (att) det visslar om det (15) (=very energetically) en ulv i fårakläder (15) (=a wolf in sheep’s clothing) inte för allt smör i Småland (15) (=not for anything in the world) sitta och uggla (15) (=sit up late and do nothing) inte vara född i farstun (15) (=not be born yesterday) veta hur landet ligger (15) (=understand the situation) gå på knäna (15) (=be totally exhausted) ligga någon i fatet (15) (=be handicapped by the fact that …) hälla vatten på en gås) (15) (=like water off a duck’s back) få bära hundhuvudet (15) (=be made the scapegoat) stå som en åsna mellan två hötappar (15)4 (=see note) kasta in handduken (14) (=throw in the towel) ana ugglor i mossen (14) (=there is something brewing) lägga benen på ryggen (14) (=run as fast as one can) ta något med en klackspark (14) (=not take a thing too seriously) ha tummen mitt i handen (14) (=be all fingers and thumbs) rinna ut i sanden (14) (=come to nothing) ge sig katten på (14) (=be absolutely certain about) slå näven I bordet (14) (=bring one’s fist down on the table) sitta med skägget i brevlådan (14)(=be in trouble) bära frukt (14) (=bear fruit) barka åt skogen (14) (=go to rack and ruin) gå som katten kring het gröt (14) (=beat about the bush) köpa grisen i säcken (14) (=buy a pig in a poke) göra slag i saken (14) (=go right ahead) sätta sig på sina höga hästar (14) (=get on one’s high horse) få sina fiskar varma (14) (=get a reprimand) ha en räv bakom örat (14) (=be a sly fox) sätta myror i huvudet på någon (14)5 (=see note) vara i gasen (14) (=be in high spirits; be tipsy) kasta ett getöga på (14) (=take a quick look at) ta tjuren vid hornen (13) (=take the bull by the horns) stå på näsan (13) (=fall on one’s face) ha is i magen (13) (=keep one’s cool) sitta med armarna i kors (13) (=not do anything) få blodad tand (13) (=one’s appetite has been whetted) krypa till korset (13) (=eat humble pie)
KN 14 12 11 11 11 11 11 11 10 10 10 9 9 9 8 8 8 8 7 6 6 5 5 4 14 12 (1) 12 (1) 11 10 10 9 9 9 10 (1) 8 7 8 (1) 8 (1) 8 (1) 6 5 5 5 3 12 (2) 11 (1) 10 (1) 9 7 7 (1)
TKR 1 2 4 4 3 3 3 3 4 3 3 5 5 4 6 6 5 5 2 5 4 5 4 4 3 3 2 4 (1) 2 5 6 (1) 4 4 3 6 6 5 4 4 5 6 (1) 4 6 3 3 (1) 4 (1) 3 6 (1) 6 (1)
RC
NKR
NKC
NK
1 1 1 1 1 1 2 1 2 1 2 2 2 3 2 2 3 1
2 (1) 1 2 (1) 1 1 2 (1) 1 2 3 2 3 2 2 2 1 1 1 (1) 1
1 1
1 (1)
1
3 1 3 2 3 5
1 1 1 (1) 1 1 (1)
1 (1)
1
2 1 1 2 2
2 (1) 2
1 1 1
1
(Continued)
Idioms and Proverbs 127
Table 4.5 (Continued) No A31 A42 A46 B1 B22 B26 A36 B4 A20 B19 A12 A38 B29 A26 A14 A44 A27 B6 A43 A41 A2 A18 B8 A23 B16 B13 A25 B5 A4 A39 Total
Swedish idiom given in context (correct answers) spotta i nävarna (13) (=roll up one’s sleeves) bli eld och lågor (13) (=become very enthusiastic) salta räkningen (13) (=salt the bill) något i hästväg (13) (=something quite extraordinary) ha en gås oplockad med någon (13)6 (=see note) det osar katt (13) (=you smell a rat) fika efter (13) (=hanker after) det fina i kråksången (13) (=the beauty of it) väcka ont blod (13) (=stir up bad blood) kasta pärlor för svin (12) (=cast pearls before swine) plocka russinen ur kakan (12) (=take the best plums) hamna på glasberget (12) (=be left on the shelf) lägga lök på laxen (12) (=make matters worse) det är hugget som stucket (11) (=it comes to the same thing) måla fan på väggen (11) (=make things worse than göra en tavla (11) få kalla handen (11) ingen ko på isen (11) bli lång i ansiktet (10) bita ihop tänderna (10) bekänna färg (10) lägga rabarber på (10) sila mygg och svälja kameler (10)7 sista skriket (9) gå kräftgång (9) en gökunge i boet (9) suga på karamellen (8) possible) göra någon en björntjänst (7) en het potatis (7) i bara mässingen (7) naked)
Correctness rate
they are) (=put one’s foot in it) (=be turned down flat) (=there’s no great hurry) (=pull a long face) (=keep a stiff upper lip) (=confess) (=take, walk away with) (=see note) (=the latest fashion) (=move backwards) (=a cuckoo in the nest) (=enjoy as long as (=do a person a disservice) (=a hot potato) (=in one’s birthday suit,
KN 7 8 (1)
TKR 6 (1) 5
RC
NKR
1 (1)
6 6 (1)
2 8
1 1 (1)
5 5 6 (1) 4 4 8 7 4 2 10
6 7 (1) 6 (1) 7 7 (1) 4 (1) 4 (2) 6 4 3 (2)
3 (1) 2 2 2 2 (1) 1 1 (1) 1 3 1 (1)
1 1 1
7
4 (1)
2 (1)
1
6 8 (2) 5 8 (1) 8 (1) 8 (3) 3 1 (1) 7 1 1 8 (5)
4 5 (1) 4 (1) 3 5 (3) 2 (1) 6 (1) 4 2 (1) 4 (1) 1 4
1 2 (1) 1 3 (3) 2 (1) 3 (1) 1 5 (2) 2 (1) 3 3 2 (1)
10 (4 ) 7 (3) 6 (2)
4 (3) 5 (2) 4 (2)
1 (1) 2 (1)
615 (16) 94.1%
336 (33) 90.2%
113 (23) 79.6%
NKC 1
NK 1
4
2
1
1
2 1 3 1 2
66.7%
2
2 2 1
2 (2)
2
2 1 (1)
1
1 1 4 (1) 3 (3) 4 (2) 4 (1) 1 (1)
12 (4)
1 1
4 1 1 3 4
3 (3) 2 (2)
1
75 (17) 77.3%
34 (0) 100%
transparent/semi-transparent items in the upper half on the one hand and the opaque items in the bottom half on the other hand are of the same type. For example, while let sleeping dogs lie (13 correct answers) was categorised as a fully transparent item due to the fact that there is a direct semantic counterpart in Swedish – väck inte den björn som sover (in which a bear is the proverbial animal, not a dog) – he is a square peg in a round hole (12 correct answers) was classified as an equally fully transparent item because of its decomposability. Paint the town red (13 points) and get a word in edgeways (also 13 points) were both considered to be semitransparent idioms, but, again, for different reasons. For the former, there is a semantic counterpart in Swedish – göra stan osäker – but this is considerably more literal in character than the English one, using the adjective uncertain instead of a colour to describe the situation. Nevertheless, it was deemed that since the colour red often signals some kind of danger, the students would, with the help of the Swedish counterpart, be able to figure out the meaning of the English idiom. Get a word in edgeways, on the other hand, was, in spite of its decomposable nature, categorised as a semi-transparent idiom due to the infrequency of the key word edgeways. Along the same lines, dressed up like a dog’s dinner (four correct answers),
128 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
play gooseberry (two correct answers), as keen as mustard (also two correct answers) and go Dutch (no correct answers) were all judged to be opaque idioms for different reasons. Dressed up like a dog’s dinner was considered to be diffuse for several reasons. Firstly, there is neither a literal nor a semantic counterpart to this idiom in Swedish. Secondly, it was thought to be a false friend in that the first meaning that comes to mind is not that you are dressed very smartly but rather the opposite. This was indeed the most common belief among the students who produced an incorrect answer for this expression. Play gooseberry was categorised as an opaque item due to its non-decomposability and because, even though there is a Swedish counterpart – vara tredje/femte hjulet under vagnen – this is only semantic in character and has no literal similarity. Additionally, the key word gooseberry is an infrequently used item. As keen as mustard, for which there is neither a direct literal nor a direct semantic counterpart, can really only be interpreted in one of three ways. Either you are very keen, not keen at all, or somewhere in between, but, when decomposing it, there is really nothing that tells the learner which of these is the correct one. Lastly, go Dutch, just like as keen as mustard, does not have a counterpart in Swedish (literal or semantic) and its constituent parts, unless its etymology is explained, do not help at all. The effect of transparency can also be seen with the errors made by the native speaker. Of the 13 incorrect answers given, almost half of them (six) involved items that were categorised as opaque (sell a person a pup, when the chips are down, that cat won’t jump, get in Dutch with, play the giddy goat and do a double take), four that were considered to be semitransparent (cut/hurt a person to the quick, give it a double take, raise snakes and a fly in the ointment), and only three that were classified as fully transparent (be a turncoat, fine feathers make fine birds and all his geese are swans). When the students’ results on the Swedish test are considered, a slightly different picture emerges. Whereas there is still a concentration of fully and, again, to a lesser degree semi-transparent idioms in the upper half of Table 4.5, the distribution is more scattered, especially with the latter type, the semi-transparent idioms appearing to be almost equally common among the expressions that received high scores as among those that received comparatively low scores. The results described above tally well with the model of dual idiom representation. According to the studies forming the basis of this theory, as L2 learners have to rely heavily on their mother tongue when trying to interpret L2 idioms, not having created as many idiom entries in their L2 as their L1, they are also more inclined to try to decompose L2 idioms than native speakers are. Compare this with the Cieślicka (2006) study discussed in the theoretical background (Section 4.2), where it was observed that, even for opaque idioms, the most common disambiguation strategy was to try to interpret them literally. Unsurprisingly,
Idioms and Proverbs 129
decomposition works well with expressions that are transparent and semitransparent, but not with opaque ones, hence the concentration of the former two types in the upper part of the L2 table, and the corresponding concentration of the latter type in the bottom part. The reverse is of course true for L1 idioms for which learners already have a lot of idiom entries and therefore do not have to resort to literal readings of the idiom constituents to the same extent. The students are thus not as dependent on transparency in their mother tongue as they are in their L2. It here needs to be pointed out that even though many of the students offered correct translations for quite a few of the English idioms, this does not entail that they knew their meanings. Based on Tables 4.4 and 4.5, it is quite clear that in quite a few cases the informants made use of the context in which the idioms were presented (columns RC and NKC), which, as discussed in several parts of Section 4.2, is one of the most common techniques used to disambiguate idiom meaning. In fact, for the English test items, the participants indicated that they made use of the idioms’ contexts in 619 cases, 358 of which resulted in a correct answer, yielding a correctness rate of 57.8%. This can be contrasted with the L1 data, where an even higher correctness rate was achieved (78.6%), the participants here indicating that they had made use of the context in 187 cases, 147 of which resulted in a correct answer. This L1–L2 difference in inferencing was confirmed statistically at a significance level of 5%. In Karlsson (2019), both context and transparency were regarded as facilitators, but of the two the former was seen as the main facilitator. In the present investigation, however, it appears that the two factors are more intertwined than what Karlsson (2019) was able to observe. The difference in approach to contextual support may account for this. While Karlsson (2019) took great care in separating between varying degrees of contextual support (supportive; semi-supportive; not supportive), the present investigation primarily made use of the authentic text in which the idiom was found. That the interplay is a reality can be seen in Table 4.6, which relates the students’ disambiguation success based on context to the idioms’ degree of transparency. On the English test, the students were most successful in their inferencing with transparent idioms, less so with semi-transparent idioms, and the least so with the opaque idioms. On the Table 4.6 Accuracy rate of disambiguation based on context in relation to degree of transparency English test
Swedish test
Accuracy rate Transparent idioms
77.8%
90.4%
Semi-transparent idioms
65.1%
75.0%
Opaque idioms
52.6%
77.1%
130 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Swedish test, the students were again most successful with the transparent idioms, while the continuum does not quite hold true for the semi- transparent/opaque distinction. Commonality also had an effect on whether the testees could provide the correct meanings on the L2 test. On all three tests parts, if the total number of scores for the first half of the idioms (the most frequent items) is compared to that of the latter half (the most infrequent items), the former is consistently higher. On the L1 test, this was only the case with Parts B and C, showing – tallying with the model of dual idiom representation – that frequency is more important in a second language than a first. (It must, however, be pointed out that there were quite a few exceptions within each test part.) This means that while in Karlsson (2019) the conclusion was drawn that frequency is a relatively poor predictor of whether an idiom is known or not, it appears to play a slightly more important role in the present investigation. This discrepancy may again be attributed to a difference in approach. Whereas Karlsson (2019), based on Collins Cobuild Idioms Dictionary, only made use of three frequency bands (especially frequent idioms; frequent idioms; not very frequent idioms), the present investigation, based on the BNC and Språkbanken respectively, took the number of occurrences of each idiom into consideration, thus offering a more fine-grained picture of the effects of frequency. Still, as in Karlsson (2019), when contextual support, transparency, and frequency are taken into account, frequency crystallises as the weakest predictor of the three. Tables 4.7 and 4.8 present the students’ results on the proverbs included in the present study and should be deciphered in the same way as the students’ results on idioms shown in Tables 4.4 and 4.5. The reader is here reminded that the proverbs were not offered to the students in context. As is shown in Table 4.7, which presents the students’ results on the English proverbs, there is a concentration of fully transparent items in the upper half. These are proverbs that can all be easily decomposed and for many of which there are semantic and sometimes very close literal counterparts in Swedish. Examples of this type are there’s no place like home (Swedish: Borta bra, men hemma bäst) and you can’t teach an old dog new tricks (Swedish: Du kan inte lära en gammal hund att sitta, in which the proverbial animal is also a dog). In the bottom half, since only one was categorised as fully opaque, there is a concentration of semi-transparent proverbs, for some of which there are no clear corresponding Swedish items (e.g. once bitten, twice shy) or only infrequently used Swedish counterparts (e.g. the early bird catches the worm (Swedish: Morgonstund har guld i mund, which is rather old-fashioned) and the end justifies the means (Swedish: Ändamålet helgar medlen, which is also rather old- fashioned)). Hence the students’ results on the proverbs also largely agree with the hypotheses put forth by the model of dual idiom representation.
Idioms and Proverbs 131
Table 4.7 The students’ results on and evaluation of Part C (proverbs) on the English test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, NKR = the item is recognised but its meaning is not known, NK = the item is not known). Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation No. Proverb (correct answers) C1 There’s no place like home. (15) C4 Honesty is the best policy. (15) C7 You can’t teach an old dog new tricks. (15) C6 Laughter is the best medicine. (15) C20 What is done by night appears by day. (15) C18 There are plenty of fish in the sea. (14) C9 The leopard cannot change its spots. (13) C11 Do as you would be done by. (12) C17 One man’s meat is another man’s poison. (12) C5 When in Rome do as the Romans do. (11) C13 He that fights and runs away, may live to fight another day. (9) C3 Once bitten, twice shy. (9) C16 Little wit in the head makes much work for the feet. (8) C15 In the eyes of the lover, pock-marks are dimples. (7) C2 The end justifies the means. (6) C14 He that lives next door to the cripple, will learn to halt. (6) C8 The early bird catches the worm. (5) C10 Age and wedlock tames man and beast. (5) C19 The world’s coin is ingratitude. (4) C12 Don’t cross the bridge till you get to it. (2) Total Correctness rate
KN 10 7 7 5 4 6 5 4 3 4 1 2 2 (1) 3 (1) 4 (2) 2 5 (4) 74 (8) 89.2%
TKR 5 8 6 9 6 5 5 6 7 7 (1) 6 (1) 7 4 (2) 3 7 (4) 3 5 (3) 2 2 (1) 5 (4) 108 (16) 85.2%
NKR
NK
2 2 (1) 4 (2)
2 1
2 1 1 4 3 (1) 4 3 (1) 3 2 4 4 39 (5) 87.2%
4 (1) 2 (1) 2 4 (1) 4 (2) 4 (2) 7 (2) 1 (1) 7 (1) 2 9 (1) 6 (1) 1 56 (13) 76.8%
The effect of transparency can be seen here too with the errors made by the native speaker, although this time this is not as pronounced as with the idioms. Of the five incorrect answers given, only one error was made with what is categorised as a fully transparent proverb (what is done by night appears by day), one with a proverb that is considered to be totally opaque (age and wedlock tames man and beast), and three with semi-transparent proverbs (in the eyes of the lover pock-marks are dimples, he that lives next door to the cripple will learn to halt, and the world’s coin is ingratitude). Since no Swedish proverbs were judged to be totally diffuse in character, it is not possible to make a comparison between transparent and opaque proverbs in the students’ L1. Nevertheless, while there appears to be a concentration of fully transparent items in the top of Table 4.8, the semi-transparent proverbs are in the majority in the bottom. Individual L1 and L2 knowledge
For 7 of the 15 students, there is a correlation between what they were able to achieve in their first language as compared to their second. Five of these seven learners display consistently high scores, all being among the top five on all test parts and thus on the tests as a whole (Students 2, 8, 9, 11 and 14), while the remaining two achieved only low scores, both being among the bottom five students on all test parts in both languages
132 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 4.8 T he students’ results on and evaluation of Part C (proverbs) on the Swedish test, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (KN = the item is known, TKR = the item is recognised and thought to be known, NKR = the item is recognised but its meaning is not known, NK = the item is not known). Figures in parentheses in columns NK indicate cases where the students were incorrect in their self-evaluation No. C17 C5 C7 C8 C14 C9 C10 C2 C4 C1 C13 C18 C15 C19 C20 C11 C6 C3 C16 C12
Proverb (correct answers) Man ska inte gråta över spilld mjölk. (15) (=It is no use crying over spilt milk.) Den som sig i leken ger får leken tåla. (15) (=If you give a jest you must take a jest.) Man ska inte döma hunden efter håren. (15) (=Appearances are deceptive.) Äpplet faller inte långt från (päron)trädet. (14) (=The apple never falls far from the tree.) Gräset är inte grönare på andra sidan. (14) (=He that goes far to be married, will either deceive or be deceived.) Han vänder kappan efter vinden. (14) (=He is a turncoat.) Solen har också sina fläckar. (14) (=Every man has his faults.) Allt är inte guld som glimmar. (14) (=All that glitters is not gold.) När katten är borta dansar råttorna på bordet. (13) (=When the cat’s away the mice will play.) Allting går att sälja med mördande reklam. (13) (=Anything can be sold.) Den illa gör han illa far. (12) (=He that mischief hatches, mischief catches.) Man ska inte gå över ån efter vatten. (11) (=Don’t give yourself a lot of unnecessary trouble.) Han skäller värre än han bits. (10) (=Barking dogs seldom (don’t) bite.) När fattig blir rik, blir han fan lik. (8) (=Muck and money go together.) Råttorna överger det sjunkande skeppet. (5) (=Rats desert a sinking ship.) Anden är villig, men köttet är svagt. (3) (=The spirit is willing but the flesh is weak.) Det börjar med en knappnål och slutar med en silverskål. (3) (=Crimes are made secure by greater crimes.) Kaka söker maka. (2) (=Birds of a feather flock together.) Ingen är profet i sitt eget (färdenes)land. (1) (=A prophet is hungry in his own country.) Av små grisar blir det stora svin. (1) (=Little pigs grow up to be big pigs.)
Total Correctness rate
KN
TKR
9
4
NKR
NK
8
7
7
6
1 (1)
9
4 (1)
2 (2)
8
5
1
9 (1)
5
1 (1)
9 (1)
5
1 (1)
7
8 (1)
8 (1)
7 (1)
2
8
1 (1)
4 (2)
5
5
2 (1)
3 (1)
4
5 (1)
4 (2)
1
5 (2)
6 (1)
3 (1)
7 (3)
2 (1)
5 (2)
2
6 (3)
3
4
3 (2)
3 (3)
1
8 (2)
1
6 (4)
1
7
2 (1)
3 (2)
6
4
1
3 (3)
4
7
1 (1)
4 (4)
2
7
100 (9) 91.0%
107 (27) 74.8%
32 (10) 68.8%
53 (8) 84.9%
(Students 1 and 10). It may be that learners who are good at processing idiomatic expressions in their mother tongue – that is, students who have developed successful strategies and thus created a great number of idiom entries in their L1 – are subconsciously able to transfer their techniques to a second language. Implementing context in the disambiguation process may, for instance, be a technique already acquired in their L1 and now made use of when faced with idioms and proverbs in an L2.
Idioms and Proverbs 133
It is also interesting to note that three of the five learners who did consistently well on idioms also did consistently well when tested on vocabulary size (see Chapter 2) (Students 2, 9 and 11), and that both learners who did consistently poorly on idioms also did consistently poorly when tested on vocabulary size. For a comparison, the reader is here referred to Tables 2.12 and 2.13, which display the informants’ individual results on the three tests in focus in that chapter (vocabulary taught at upper secondary school level, specialised uses, and advanced vocabulary). Additionally, four of the high-/low-scoring learners in the present chapter were also among the high/low achievers on the affixation tests discussed in Chapter 3, with Students 2 and 11 doing consistently well and Students 1 and 10 doing consistently poorly. There are also a few students who did consistently well in their native language only, but poorly in their L2. Student 12, for example, the oldest of the subjects (a 49-year-old male student), achieved high scores on all the Swedish test parts but produced only mediocre or low scores on the English test parts. As pointed out in previous chapters, this student had not studied English for quite some time before taking these vocabulary tests, and his English (spoken and written language) was indeed very poor. (For a more detailed description of the informants, the reader is referred to Section 1.2.) 4.5 Conclusions and Pedagogical Implications
Since idioms form a natural part of the L1 lexicon, and, as the present chapter has shown, native speakers usually have a good command of these types of expressions, they should be made to form an integral part of all courses aiming to teach L2 vocabulary. The main challenge is to find a strong enough incentive to acquire them, since L2 learners very often perceive these expressions as infrequent and hence perhaps not very important to learn. As discussed in Section 4.1, while idioms occur quite infrequently in some registers, they are comparatively frequent in others, with commonality, as shown in the present investigation, playing a part in whether an idiom is known or not. Based on dictionaries like Collins Cobuild Idioms Dictionary, which makes a frequency-based distinction between items, teachers could therefore start with the most common idioms, illustrating with examples from, for instance, the world of sitcoms or other types of programmes that produce a wide range of expressions, thus making the idioms even more accessible by contextualising them. (The reader is here also referred to Karlsson, 2019, especially Chapter 3, in which multimodal and visualisation techniques beneficial to L2 learners are discussed.) Based on the present author’s experience, this would make the students aware of the prolific nature of certain items, and so it would increase their eagerness to learn. This approach should be taken
134 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
not only with idioms in a second language but also with idioms in a learner’s mother tongue since, as shown in the present chapter, for many students, L1 idiom knowledge tallies with their mastery of idioms in their L2. Put differently, learners who know a large number of idioms in their native language –and thus have created a great many idiom entries in their L1 – have, most likely, during the process of acquiring this knowledge, developed idiom comprehension techniques that may be transferred and made use of when encountering idioms in a second language. Collaboration between instructors teaching students’ mother tongue and those teaching L2s is therefore called for. The same conclusion was drawn in Karlsson (2019). Furthermore, since students seem to approach idioms from a transparency perspective, teachers should start with idioms for which literal readings will help the students understand their figurative meaning and then move on to idiomatic expressions that are more opaque in character. Discussing source domains and offering etymological elaboration seem to be beneficial roads to take, starting with transparent idioms so that no gap occurs in the teaching continuum. Admittedly, many teachers most likely already make use of source domains and/or etymological elaboration from time to time, but this is probably done without considering it to be a specific technique. To be really useful the techniques should, in the present author’s mind, be implemented in a much more structured way, informing the students of the techniques and why they are used. 4.6 The Chapter That Follows
The aim of Chapter 5 is to explore the 15 students’ L1 and L2 mastery of prepositions that are used idiomatically on the one hand and multiword verbs on the other hand. Both types have clear affinities with idioms. While for the former type there does not, at least not superficially, appear to be any rhyme or reason for the choice of particle linked to a specific verb, noun, or adjective, for the latter type the parts of which these items are made up are not usually their sum. Chapter 5 will therefore again primarily be dedicated to delving into the learners’ vocabulary depth. Notes (1) The term ‘idiomatic expression’ (and other paraphrases) is here used to refer to idioms as defined in the literature review in the present chapter. (2) Some of the items tested are similes. On the English test, these are as mad as a hatter (A19), as safe as houses (A23), as keen as mustard (A27), and look like something the cat dragged in (B13); on the Swedish test, these are gå åt som smör i solsken (=sell like hot cakes) (A24), inte för allt smör i Småland (=not for anything in the world) (A29), så (att) det visslar om det (=very energetically) (A32), gå som katten kring het gröt (=beat about the bush) (B17), and stå som en åsna mellan två hötappar (=stand like a donkey between two bundles of hay) (B25). These expressions are comparisons that
Idioms and Proverbs 135
are highly institutionalised and fixed and are usually, but not always, transparent in character (Moon, 1998: 22, 150). Furthermore, a few of the items included in Parts A and B may be categorised as proverbs rather than idioms by some researchers. Disagreement concerning classification is only natural since, as pointed out in Section 4.2, the boundary between them is rather diffuse. (3) The best-known idioms in the Trulsson (2007) study (discussed in the last part of Section 4.2), receiving 80% or more correct answers, were ta tjuren vid hornen (take the bull by the horns), köpa grisen i säcken (buy a pig in a poke), and ha en räv bakom örat (be a sly fox). As the reader can see, these expressions also received high scores in the present investigation. (The expression det är ingen ko på isen (there’s no great hurry, literal translation: there’s no cow on the ice), on the other hand, was found to be comparatively difficult in Trulsson, with fewer than 50% of participants knowing its meaning. In the present study, 11 of the 15 subjects still offered the correct meaning for this expression.) Furthermore, if only the younger subjects are considered in Trulsson, there are still another two idioms that received very low scores: kasta ett getöga (take a quick look at, literal translation: take a quick goat look at), and göra en björntjänst (do a disservice, literal translation: do a bear disservice)). Only 15% and 19% of the informants provided the right meaning of these idioms respectively. Whereas the majority of the students in the present investigation had no problems with the former of these two, the latter caused major problems. Interestingly, as in Trulsson’s study, quite a few of the students who supplied an incorrect explanation thought that it meant to do someone a big favour, not the opposite. (4) The English translation of the idiom is to stand like a donkey between two bundles of hay. It is offered here for spatial reasons. (5) This idiom means to give a person something to think about. This translation is offered here for spatial reasons. (6) The English counterpart is to have a bone to pick with a person. It is given here for spatial reasons. (7) This idiom can be given a direct literal translation into English: to strain at a gnat and swallow a camel. It is also offered here for spatial reasons.
5 Idiomatically Used Prepositions and MultiWord Verbs
5.1 Introduction
In Karlsson (2002: 81), it is pointed out that ‘prepositions are often thought of as forming a rather difficult area of investigation’ and that ‘perhaps one reason for this is that they are borderline cases involving both grammar and lexicon’. This also appears to have repercussions in the L2 classroom, where teachers, at least in Swedish schools, either tend to avoid discussing prepositions altogether or, when they occasionally do spend time on the area, only give learners handouts that contain a jumble of different prepositional uses without any real structure. While visiting teacher trainees doing their practice, the present author has experienced this first-hand many times. The type of prepositional use focused on in the present chapter, idiomatic prepositional use, which has strong affinities with items of vocabulary rather than units of grammar, is especially prone to L2 errors. In Karlsson (2002), for instance, it was shown that this error type was, in both compositions and translations, by far the most common one of the three categories studied. Examples of errors made by first-term university students in the Karlsson study (2002: 109–110, 126–127) are given in (1)–(3) (other types of errors remain): (1) At (On) those occations, Swedes are really one people. (composition) (2) Of course he couldn’t count with (on) buying a house of his own the following two years. (translation) (3) When he arrived to (in) London he could have spent the days in the English Capital together with bookproducers and author-collegues. (translation) Moreover, the error type appears to persist. Examples (4)–(6) are errors made by second-term university students from the same study (other errors remain): (4) Thus, it is the right of human beings to survive on (at) the expense of other species. (composition) 136
Idiomatically Used Prepositions and Multi-Word Verbs 137
(5) Early in 1990 one British and one Swedish citizen were executed, because they had been accused for (of) spying. (translation) (6) It was worth continuing the struggle if he would just succeed to be (in being) declared famous before his death. (translation) After having studied English for more than 11 years, this kind of error is still very prolific among university students, as exemplified by the mistakes in (7)–(11) made by third-term students in the Karlsson study (again other errors remain): (7) She has been treated and operated – (on) for cancer and her boyfriend has recently left her. (composition) (8) During the latest twelwe years the average income of an American family has decreased with (by) 2.4% and consequently the U.S.A. have further two million poor people to take into consideration. (translation) (9) He also had a keen interest for (in) politics and became the youngest member of the House of Commons when he won an election in a Welsh election district. (translation) (10) In the 1980s, around 2 million people were, in (on) average, unemployed. (translation) (11) Then you will also have time for a visit in (to) the picturesque village La Palude with medieval traditions and yet another village in the far west. (translation) In addition to this being a highly error-prone area, studies have shown that native speakers react very strongly to the foreignness of these kinds of errors (Hultfors, 1986). As a result, there is certainly an urgent need for teachers to pay more attention to this type of prepositional use. Multi-word verbs,1 which can span from being highly literal/ transparent (he picked up the pen from the floor) to figurative/semi- transparent (he felt he had to step in (= intervene) and help her son) to highly idiomatic/non-transparent in nature (he was set up (= framed) by the police), 2 the latter type of which is in focus in the present chapter, are also prone to different types of errors, especially the highly idiomatic ones (Mondor, 2008). Sometimes the (superficially) idiosyncratic nature of the particle can create comical situations. Liu (2008: xiii) offers such an example in which a student heard his teacher say that their visit to a museum had fallen through. Thinking that this meant that they were going ahead with the visit, believing that fall through was on a par with go through, the student made a cheer of joy, probably with the result of being looked upon with suspicion by his fellow students. In Mondor (2008), a study that will be discussed in more detail in Section 5.2, a number of different error types with multi-word verbs are shown. In example (12), for instance, the
138 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
student has chosen the wrong multi-word verb altogether, confusing it with the verb–particle combination get on. (12) In the years of studying, the two disillusioned souls go on extremely well together. (Mondor, 2008: 171 (taken from SWICLE)3) In the sentence (13) I would say we are more apt to dream on now than ever before, since hope is the last thing we tend to give up. (Mondor, 2008: 169 (from SWICLE)) the single verb dream should be used instead, since dream on is really only used in sentences with a negative semantic prosody, as in ‘What a problem he was. A chip off the old block he was not. Bright, yes. Motivated, yes. But tough? Dream on’ (Mondor, 2008: 169, who in turn took it from the BNC). Similarly, in example (14), which involves a transfer-induced error, the student has again used a multi-word verb where it is not called for, the correct verb being the single verb stop. (14) Of course it is much easier not to stop up and reflect over the life we are living. It is easier to follow the main stream. (Mondor, 2008: 135 (from SWICLE)) A final example is shown in (15). (15) The setting of a story is very often used to help up the atmosphere of a story (as in Kew Gardens) but at the same time it can be more or less eliminated from the story. (Mondor, 2008: 39 (from SWICLE)) The student here seems to have thought that the phrasal verb help up goes together with the object atmosphere, along the same lines as with the object floor in He helped her up from the floor. Furthermore, there is a considerable amount of research that indicates that L2 learners, in comparison to native speakers, underuse multi-word verbs (e.g. Dagut & Laufer, 1985; Hulstijn & Marchena, 1989; Laufer & Eliasson, 1993; Sjöholm, 1995), hence overusing single verbs and sounding more formal than is appropriate in certain contexts: The plain fact is that what distinguishes the writing and, above all, the speech of a good foreign student from those of an Englishman is that what an Englishman writes and says is full of these expressions, whereas most foreigners are frightened of them, carefully avoid them, and sound stilted in consequence. Foreign students who enjoy being flattered on their English can best achieve this by correctly using masses of these compound verbs. (Bywater, 1969: 97)
Idiomatically Used Prepositions and Multi-Word Verbs 139
Since multi-word verbs are very common items, this underuse is serious (Celce-Murcia & Larsen-Freeman, 1999; Darwin & Gray, 1999; Gardner & Davies, 2007; Moon, 1997). Another complicating matter is that a great many multi-word verbs are polysemous in character. When investigating the 100 most frequent multi-word verbs in the BNC, Gardner and Davies (2007) found that each verb had an average of 5.6 meaning senses. The situation is made worse by the fact that L2 learners’ mastery of multiword verbs is a neglected field of research, especially when it comes to advanced students (Granger, 1999, 2003). Since idiomatically used prepositions and multi-word verbs appear to be difficult but important areas to master for L2 learners in order to be able to approach native-like behaviour, it seems justified to incorporate the subjects’ present knowledge in these two areas when aiming at comparing the students’ L2 knowledge to their corresponding knowledge in their mother tongue. 5.2 Theoretical Background and Previous Research
The present section is divided into three main subsections. In the subsection that comes first, definitions will be offered of idiomatically used prepositions on the one hand and multi-word verbs on the other hand. The subsection that follows will present a number of research articles focusing on L2 learners’ mastery of the two types of vocabulary, starting with idiomatically used prepositions. In the last subsection, the reader will be offered presentations of possible pedagogical approaches in both areas. Idiomatically used prepositions and multi-word verbs defined
In Karlsson (2002), prepositions were considered from a second-language-learner perspective and divided into three main types: ‘basic’ use (as in there’s a box on the floor), ‘systematic’ use (as in I’ll see you on Friday), and ‘idiomatic’ use (as in they commented on her article), and the criteria used to categorise these prepositional types are also mainly adhered to in the present investigation. The category in focus in the present chapter – idiomatic use – is most easily defined by putting it in relation to the other two. Basic prepositional uses can best be described in terms of native speaker perception (e.g. Herskovits, 1986: 54; Lindstromberg, 1997: 19), which appears to tally with prototype theory, both prepositional prototype theory specifically (e.g. Rice, 1996) and prototype theory in general (e.g. Coleman & Kay, 1981; Fillmore, 1975; Rosch & Mervis, 1975). Accordingly, most native speakers would agree that, of the three examples given above, there’s a box on the floor incorporates the best example of the meaning of the preposition on; that is, on has its most prototypical meaning in this sentence. Basic prepositional uses are also the ones that
140 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
are acquired first by L1 learners. In Piaget and Inhelder (1967), for instance, it was shown that children’s first representations of space are based on basic relations such as proximity (e.g. next to), order (e.g. between and behind), and enclosure and surrounding (e.g. inside and in), all of which are part of central meanings of prepositions. Finally, most basic prepositional uses can be depicted geometrically. This is strengthened further by the fact that different studies appear to agree on what these geometrical figures should look like. Compare, for instance, those suggested by Lindstromberg (1997: 27) with those proposed by Quirk et al. (1985: 674). Moreover, among languages that make extensive use of prepositions, there also appears to be agreement on what constitutes a basic prepositional use; that is, there are direct counterparts in all these languages for each of these basic uses. For example, the English preposition on in there’s a box on the floor corresponds directly to the Swedish preposition på in det står en låda på golvet. Similarly, the English preposition in in the sentence there’s a puppy in the box has its direct Swedish counterpart in the preposition i, as seen in det finns en valp i lådan, and so on. The second type of prepositional use – systematic prepositional use – is a use that from a second-language-learner perspective can be generalisable for one of two reasons. The first is that there are, based on contrastive differences, specific rules for its use, as seen in examples (16)–(18) (Quirk et al., 1985: 687–688; Svartvik & Sager, 1996: 120–123, 405–411): (16) at seven o’clock, at Christmas, at Easter, etc. (rule: when referring to a point in time use at) (17) on Monday, etc. (rule: when referring to days as periods of time use on) (18) in 1969, in May, in the morning, etc. (rule: when referring to periods longer or shorter than a day use in) The second reason is that there is some kind of co-occurrence, as exemplified in (19) and (20): (19) at the end/beginning, etc. of (20) for the first/second/third, etc. time in which at and of on the one hand and for and time on the other hand form the frames of the phrases in question. ‘As soon as a student has understood the co-dependence of the elements forming the frame, and filled the frame correctly, a number of similar phrases can be constructed (e.g. at the turn of the century, at the age of…, for the umpteenth time). This type of prepositional use can thus be said to be construction-dependent’ (Karlsson, 2002: 103; see also Erman & Warren, 2000).4 Idiomatically used prepositions are prepositions that are normally dependent on the choice of verb (as exemplified by comment on, congratulate on and succeed in), noun (as in on those occasions and an
Idiomatically Used Prepositions and Multi-Word Verbs 141
increase in), and adjective (e.g. dependent on and used to); that is, these uses are bound to specific verbs, nouns, and adjectives, forming tightly knit units. 5 In fact, the idea of ‘boundness’ versus ‘freeness’ appears to be adhered to in many discussions concerned with prepositional usage. In Biber et al. (2021: 76), for instance, ‘free prepositions’ are considered to ‘have an independent meaning; the choice of preposition is not dependent upon any specific words in the context’. ‘Bound prepositions’, on the other hand, are seen to ‘often … have little independent meaning, and the choice of the preposition depends upon some other word (often the preceding verb)’. Moreover, the notion that idiomatically used prepositions are interpreted mentally as units is substantiated by slips of the tongue, as in *go overing the exercises, *are we set asiding the rules? (Bolinger, 1971: 115– 116), *he go backs, *she washed upped the dishes and *I’d forgot abouten doing that (Aitchison, 1987: 109–110). To sum up, the three prepositional uses discussed above can thus be placed along a predictability continuum, where basic prepositional uses are the most predictable ones, systematic uses are less predictable, and idiomatic prepositional uses are the ones that are most difficult to predict. This is illustrated in Figure 5.1. (For an even more detailed description of basic, systematic, and idiomatic prepositional uses, see Karlsson, 2002: chap. 3.) As for multi-word verbs, a concept that Swedish learners – in contrast to many other L2 learners of English – are already familiar with (Moon, 1997: 46), there is no real consensus among researchers as to how they should be defined. For instance, while some researchers see them as one type of idiom (e.g. Fernando, 1996; Makkai, 1972), others regard them as vocabulary items in their own right (e.g. Grant & Bauer, 2004; Moon, 1998). Also, whereas some researchers think that only comparatively nontransparent items should be considered true phrasal verbs (e.g. Liu, 2008),6 others believe, due to the fact that these verb–particle combinations may be defined not only by semantic criteria but also by syntactic criteria, that quite or totally literal combinations should be classified as phrasal verbs too. Since almost all the verb–particle combinations included in the present investigation are non-transparent items, the vast majority being completely opaque, no stand will be taken in the matter by the present author. Most predictable, i.e. involving least effort in learning ‘Basic uses’
Least predictable, i.e. involving most effort in learning ‘Systematic uses’
‘Idiomatic uses’
Figure 5.1 The different prepositional types seen along a predictability continuum (Karlsson, 2002: 110)
142 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Forming a continuum, a number of criteria are used to distinguish more idiomatic items from those that are comparatively literal in character. Some are of a semantic nature. Firstly, idiomatic verb–particle combinations often have one-word synonyms, such as visit for call on, ring for call up and postpone for put off, which emphasise the semantic unity of the structure.7 From this it also follows that the verb in an idiomatic multi-word verb cannot be replaced by a near synonym without there being a change in meaning (The Oxford Dictionary of Phrasal Verbs, 1993). Similarly, most idiomatic combinations do not permit the particle to be exchanged or omitted. In Bill passed over the bread, in which the phrasal verb used is relatively non-idiomatic, it is possible to exclude the particle over (Bill passed the bread) and still retain the original meaning (The Oxford Dictionary of Phrasal Verbs, 1993: 424– 425). In They figured out that he was offended by what had happened, which contains the idiomatic phrasal verb figure out, meaning come to realise, it is on the other hand not possible to omit the particle, since the single-verb item figure means to believe/think, as in She figured that he was offended by what had happened (Mondor, 2008: 84). Furthermore, for many opaque verb– particle combinations, there also exist noun counterparts, such as a break-up and a break-down based on the multi-word verbs to break up and to break down (The Oxford Dictionary of Phrasal Verbs, 1993: 425). Other criteria are syntactic in character. For example, while literal verb–particle combinations permit objects to be placed between the verb and the particle, as in Mary picked the pen up instead of Mary picked up the pen, this is less acceptable in non-idiomatic combinations, as exemplified by Fido picked the scent up instead of Fido picked up the scent (Mondor, 2008: 82). Fronting of the particle is usually also only possible in non-idiomatic phrasal verbs. This means that while it is fully acceptable to say down it went, fronting the particle down in the literal combination go down (Mondor, 2008: 82), it is unacceptable to front the particle down as in *down the proposal he turned, turn down being an idiomatic phrasal verb (Liu, 2008: 22). Another syntactic criterion is gapping. This involves a pattern in which one and the same stem verb is used with two different particles and therefore not repeated with the second particle. This only works with literal combinations, as in Jones pulled the old table cloth off and Peters the new one on, but not with phrasal verbs that are idiomatic, as seen in *Jones pulled the deal off and Peters the money in (Mondor, 2008: 80). Furthermore, whereas particles in idiomatic phrasal verbs usually cannot be modified by adverbials, as in *He came slowly across a large sum of money and *She turned quickly down the proposal (Liu, 2008: 22), this generally works well with more literal combinations, as in She climbed slowly up (Huddleston & Pullum, 2002: 285). Also, while an action nominalisation of the verb stem is usually possible with non- idiomatic phrasal verbs, as exemplified in His throwing of the ball up was stupid, it normally cannot be done with idiomatic multi-word verbs, as seen in *His throwing of his dinner up was stupid (Fraser, 1976: 3).
Idiomatically Used Prepositions and Multi-Word Verbs 143
Finally, there is also one phonological criterion concerned with contrastive stress. In I said to carry the prop on not off, the particle off is stressed, thus signalling that the verb–particle combination is literal. In I said to carry the deception on not off, on the other hand, no contrastive stress may be placed on the particle off, thus signalling that the combination here is an idiomatic one (Fraser, 1976: 3). L2 learners’ mastery of idiomatically used prepositions and multi-word verbs L2 learners’ mastery of idiomatically used preposition
In the present subsection, two investigations – Karlsson (2002) and Mueller (2011) – will be presented. In the former, one of the book chapters focuses on advanced Swedish learners’ knowledge of and progression with prepositional use in English as a second language. The study involves 56 Swedish university students (46 women and 10 men), whose development was investigated during a period of at least three terms. (For those students’ who had to do resits, the study period was longer.) The data consist of 187 compositions and 251 translations of Swedish texts. The students’ performance was evaluated in terms of errors in relation to the number of potential errors, with the results showing that while there was progression with prepositional use between the first and third term in the students’ compositions, there was regression, statistically confirmed, during the same period in the students’ translations. This difference is perhaps not so surprising as, when writing freely, students are able to avoid constructions of which they are unsure to a greater extent. Furthermore, when the three prepositional categories were analysed separately (see the previous subsection where these types are discussed in detail), it could be shown that idiomatic prepositional uses, being the most difficult use to predict, caused the most errors; that systematic uses, being more predictable in character, caused fewer errors; and that basic prepositional uses, the most predictable type, caused by far the least errors. This result was observed in both composition writing and translation. Furthermore, in translations, in which, as mentioned above, students are to a greater extent unable to avoid structures of which they are uncertain, there was a statistically confirmed increase in the number of errors made for bound prepositions between the first and third term, once again substantiating the idiosyncratic nature of this prepositional type. Moreover, these findings were related to Hultfors (1986), which investigates native speakers’ reactions to a number of different types of L2 errors, one being idiomatically used prepositions. While native speakers in Hultfors’s study did not feel that errors with bound prepositions were very serious from a comprehension perspective, they did feel that these kinds of errors sounded very foreign and gave a negative impression of the learners’
144 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
performance as a whole. In view of this, it seems urgent to minimise these kinds of errors in learners’ repertoires. In Mueller (2011), the aim was to investigate the effect of frequency on the students’ mastery of ‘noncentral senses of prepositions’ (2011: 483). The reader is here made aware that although the vast majority of the prepositional uses categorised as non-central in Mueller correspond to idiomatic uses in the present investigation, not all of them do. Prepositions bound to nouns and verbs were focused on, that is, the same two types as investigated in the present study. Thirty native speakers of Chinese, Korean, and Spanish were involved in the experiment, none of whom had spent more than three months in an English-speaking country before the age of 18, but most of whom were comparatively highly educated. All subjects were given the same gap-fill test containing 36 items, half of which were distracters included to prevent the informants from spotting relationships between the items of interest but still testing prepositions. The remaining 18 items tested nine different non-central senses, with each sense represented by one high-frequency item and one low-frequency item based on the collocational frequency of the verb/noun–preposition combination as seen in the American National Corpus and the British National Corpus. In each pair the items appeared randomly on either the first or second half of the test. Furthermore, for each preposition the testees were given 15 alternatives, all of which were the same for the high-frequency and low-frequency item in each pair. In the pair laugh at/wink at, for instance, the former represents a highly frequent item in both corpora and the latter a comparatively infrequent test item. The results showed quite clearly that collocational frequency does have an effect on students’ mastery of bound prepositions. While the 90 subjects tested produced 86.4% correct answers on the high-frequency items, only 61.5% of the low-frequency items were correct. Also, whereas almost all of the high-frequency senses tended to receive scores higher than 80%, greater differences could be detected with the results for the low-frequency items, fluctuating between just below 30% up to just above 80%. Additionally, in cases where errors were only observed with one of the two items in a pair (which occurred in 40.5% of cases), 35.5% were made with the low-frequency item and only 5.3% with the high-frequency item, again corroborating the frequency effect. These findings are supported by quite a few other studies that also show that co-occurrence frequencies play an important role in second language acquisition (e.g. Ellis, 2003; Wray, 2002). It is also interesting to note that the participants’ mother tongues seemed irrelevant, with all three groups (Chinese, Korean and Spanish learners) showing approximately the same sensitivity to frequency. This result is corroborated by Lowie and Verspoor (2004), who investigated Dutch learners’ mastery of English prepositions in relation to their frequencies. The conclusion
Idiomatically Used Prepositions and Multi-Word Verbs 145
drawn is that L1 transfer has less of an impact than frequency on students’ mastery of bound prepositions. The results of the Mueller study become even more interesting when they are placed in relation to how native speakers are believed to acquire prepositional senses as children. In an initial stage, L1 speakers seem to learn all uses as unanalysed units, but they eventually subject the particles to semantic analysis. It is during this analysis that native speakers work out the different senses that make up the polysemous network of each preposition (Mueller, 2011: 482), which, in a first classification, may be along the lines of basic, systematic, and idiomatic senses, as described in Karlsson (2002). L2 learners, on the other hand, will, for a number of reasons, find it practically impossible to work out all the differences. Firstly, non-native speakers have usually received less input than native speakers, and the input they do receive may very often be from other L2 learners. Furthermore, the input non-native speakers receive from native speakers often contains many examples where more than one preposition is possible, as in I posted it on my blog versus or I posted it to my blog, and a lot of prepositions tend to occur in unstressed positions, yielding piece a cake instead of piece of cake. Also, even though using the incorrect preposition may sound foreign, these kinds of mistakes generally do not hamper comprehension (see Hultfors, 1986, discussed above), and native speakers may therefore not correct and/or give feedback on erroneous uses (Mueller, 2011: 482). Lastly and most importantly, even if languages that make use of prepositions seem to have similar categories for the more basic spatial situations, they tend to extend these senses in different ways; that is, ‘L1 semantic patterns serve as an imperfect guide to non-central spatial senses in the L2’ (Mueller, 2011: 482). This means that ‘the semantic opacity of prepositions’ various meanings may constrain L2 learners’ ability to acquire all their meanings through explicit learning and may thereby force learners to rely more heavily on implicit learning mechanisms’ (Mueller, 2011: 481) such as frequency, as investigated by Mueller. The present author, however, is of the belief that non-native learners, being experienced speakers of their first language, already have an inherent feeling for the different prepositional senses that may exist in an L2, at least when learning a language that is closely related to their mother tongue. The perception of prepositions as difficult held by many Swedish learners of English may thus be due to the fact that they do not believe that the system they already have will work for such a seemingly complex area in their L2. The key is therefore to point out similarities between the two systems and to convince students to make use of their subconscious L1 knowledge when approaching L2 prepositions. This will be explored in one of the subsections that follow below. The effect of frequency, as seen in the Mueller study above, also suggests that commonality in general can be a useful tool when teaching different prepositional uses. This will be discussed in more detail in the results section (Section 5.4).
146 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
L2 learners’ mastery of multi-word verbs
In the present subsection, two research projects – Olejniczak (2006) and Mondor (2008) – will be discussed. The former study focuses on Polish students’ productive knowledge of phrasal verbs. The selection of the verbs tested, taken from 10 American novels, a number of newspaper and magazine articles, and a great many dictionary sources, was frequency-based, only including phrasal verbs that occurred more than 10 times. The items were then considered from a Polish second-languagelearner perspective and ranked along a continuum. Three of the multiword verbs were categorised as transparent and considered to be peripheral items. Examples of this type are put (money) into and come back. Another three were items in which the particles were used in one particular sense, such as ‘continuity’ (e.g. walk on and drink away), ‘activity’ (e.g. switch something on), ‘exhaustive’ (e.g. wither away) and ‘completion’ (e.g. sober someone up and calm down), thus making the phrasal verbs comparatively transparent in character. Still another three items were categorised as highly lexicalised, rendering their transparency questionable (e.g. carry on). The last five items were categorised as completely opaque in nature and lacking a clear metaphorical core, such as act up, put up with someone, do someone in and be done for. In the experiment, the subjects were faced with sentences in which they were asked to insert a suitable phrasal verb. In addition to being given an explanation in their native language as to what they were supposed to do, the students were offered an example sentence and provided with a list of possible particles. Since not all students could take the test at the same time, two tests with different test items were created. However, the general structure remained the same for both tests: Type 1: transparent items, sentences 1–3; Type 2: items in which the particles conform to one sense, sentences 4–6; Type 3: lexically fixed items, sentences 7–9; and Type 4: opaque items, sentences 10–14. Furthermore, after each sentence, the Polish single verb counterpart was offered. Examples from the four different types are given in (21)–(24) (Olejniczak, 2006: 264–265): (21) Rabbit hates to any more money Japanese pockets but he knows from Consumer Reports that in this particular line they can’t be touched for quality. (correct phrasal verb: put into (Type 1)) (22) It was a very restless country, with people tearing around all the time. Every so often, somebody would stop to a monument. (correct phrasal verb: put up (Type 2)) (23) She had of college almost as soon as she’d begun. (correct phrasal verb: drop out (here: dropped out) (Type 3)) (24) She walks out with her dog before she , but she didn’t show up yesterday. (correct phrasal verb: turn in (here: turns in) (Type 4))
Idiomatically Used Prepositions and Multi-Word Verbs 147
Version 1 of the test was given to 27 informants and Version 2 to 20 subjects, all of whom were second-year teacher-trainee students taking part in the advanced level of English as a foreign language (EFL) education. Another 10 subjects, more advanced than those belonging to the experimental group, formed a control group. They took Version 1 of the test. Even though errors of tense were disregarded, the number of correct answers was comparatively low for both test groups: 33.1% (125 of 378 responses) for those students who did Version 1 and 31.4% (88 of 280 responses) for those who took Version 2; that is, both student groups received approximately the same low percentage of correct answers. This can be contrasted with the control group of slightly more advanced students, who received a percentage of 52.1% correct answers, this too being a comparatively mediocre result. Furthermore, it was only for those subjects that did Version 2 of the test that a difference could be detected between transparent/semi- transparent items on the one hand (sentences 1–3: 31.8% correct answers; sentences 4–6: 36.4% correct answers) and those that were marginally transparent or totally opaque on the other hand (sentences 7–9: 9.1% correct answers; sentences 10–14: 22.7% correct answers). Based on these observations, Olejniczak drew the conclusion that there does not appear to be a system for phrasal verbs based on transparency/opaqueness that decides the order in which they are learned. It is interesting to note though that the subjects in general tended to give more metaphorical responses to sentences 10–14 than they did for the more analysable multi-word units seen in sentences 1–9. Since phrasal verbs are quite frequent, especially so in spoken language, and are very often preferred to single-word items, Olejniczak concludes by saying that this gap in the students’ mastery of English as a second language needs to be taken seriously. In Mondor’s large-scale study from 2008, the main aim was to investigate advanced Swedish students’ mastery of phrasal verbs in terms of over- and underuse in comparison to native speakers. Literal combinations, figurative combinations, and completely idiomatic phrasal verbs were all considered, and the particles focused on were up, down, in, out, on, off and over, all of which are frequently used (Moon, 1997). In contrast to the majority of earlier studies, Mondor’s investigation is corpusbased, with the Swedish component taken from the International Corpus of Learner English (SWICLE) and the non-native component from the Louvain Corpus of Native English Essays (LOCHNESS), thus focusing on mastery of multi-word verbs in argumentative texts. Mondor also explores the types of errors made by the Swedish learners and what factors, such as transparency/opaqueness, may have played a role in their occurrence. These errors were then categorised by native speakers along a continuum according to whether they were ‘perfectly natural’, ‘neither perfectly natural nor wholly unnatural’, or ‘wholly unnatural’.
148 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
The errors made by the Swedish learners were divided into six categories. Firstly, there were errors in which the learners had used the wrong phrasal verb, such as look up at someone instead of look up to someone, thus blurring the intended meaning. Secondly, there were errors that involved an overuse of the particle – that is, cases in which a particle had been added to a simple verb where no article should be used – as in end up instead of end in a sentence like *Most novels/stories are read because the reader wants to know how it will end up (naturally!). Thirdly, there were so-called collocational errors. This category involved errors like *rub out thoughts, where the phrasal verb rub out, which is normally used with concrete objects (e.g. pencil marks), was used with an abstract object. This error type mainly appeared to occur for reasons of emphasis. Fourthly, there were contextual errors. These were mistakes in which the phrasal verb was used correctly on a sentence level but inappropriately in the paragraph or text as a whole. Fifthly, confusion errors involved errors that were caused by there being a similarity in phonetic shape or meaning, as between go on and get along in combination with well as seen in *go on well instead of get along well. Finally, there were also errors that could not be seen to adhere to any of the categories described above, thus forming a miscellaneous group, containing mistakes like *building up a red thread (instead of providing a main thread/connecting thought). Generally speaking, the errors in this group involved mixing-up of metaphors as well as direct translation of Swedish metaphors into English (Mondor, 2008). The results of the investigation show that the Swedish learners used all three types of phrasal verb – literal, figurative, and fully idiomatic – almost as often as native speakers, and that, in terms of overall frequency, the learners mimicked the native speaker pattern, thus contradicting the results of previous studies showing that L2 speakers tend to underuse multi-word verbs. However, whereas native speakers use a great variety of phrasal verbs, the advanced Swedish learners seemed to use more tokens of each type. Moreover, the types that were overused were mostly comparatively informal combinations that very often had direct Swedish counterparts, such as sit down and cut down. (One particle that generally seemed to be underused was off.) In most cases, it appeared to be the topic of the composition that was the determining factor as to whether a verb– particle combination was over- or underused. Errors were extremely common in the material, and the overwhelming majority involved either figurative or fully idiomatic phrasal verbs, rather than literal verb–particle combinations. The two most frequently made errors were collocational errors and particle overuse, the former being more common than the latter. However, native speaker disagreement concerning the degree of gravity of the errors in question suggested that the borderline between what is correct and what is not is not always as clearcut as one would hope. Mondor’s investigation therefore shows that, even though advanced Swedish learners’ errors with phrasal verbs are generally
Idiomatically Used Prepositions and Multi-Word Verbs 149
not considered to be very grave by native speakers, a picture emerges in which Swedish learners’ use differs greatly from native speakers’ use. As in Olejniczak’s study presented above, the conclusion to be drawn is therefore that more time needs to be spent on these types of combinations. Pedagogical approaches to idiomatically used prepositions and multi-word verbs
In the following two subsections, two different techniques will be described, both of which are believed to enhance students’ processing abilities, comprehension, and retention of the vocabulary items in focus. Whereas the first subsection deals with a categorisation technique that may be combined with a frequency perspective when approaching bound prepositions, the latter describes a cognitive approach to multi-word verbs. A pedagogical approach to idiomatically used prepositions
Prepositions, which are normally considered by teachers to be part of grammar, appear to form an area that is especially neglected. When students are ‘taught’ prepositions, they are normally only given handouts to work with, and no real instructions are supplied, perhaps for the simple reason that teachers do not have any instructions to offer. This is probably partly to do with the fact that English grammar books are not very helpful in this area, at least the ones that are most commonly used as reference books by teachers in Sweden today. Although many of them are contrastive in character, prepositions are most typically presented in alphabetical order, and their usage is mainly based on traditional categories like place, time, manner, and so on. As described before, prepositional uses can instead be made considerably more accessible by dividing them into categories that are based, from a second-language-learner perspective, on their predictability. In Schmidt (2004), supervised by the present author, the three prepositional types incorporated were adopted from Karlsson (2002) and tested on first- and second-year upper secondary school students (categorised as). The part of the study investigating whether teaching of these three prepositional uses would help increase learners’ understanding and production of prepositions in general is of major interest to the present chapter. The students involved in the Schmidt study were given one pre-test and one post-test, both of which involved 25 occurrences of each prepositional type, 8 that is, basic, systematic, and idiomatic uses. Furthermore, in each category 10 uses were tested in the form of a gap-fill exercise for which no Swedish counterparts were offered, whereas for the remaining 15 the students were asked to perform translations into English. In addition, 36 of the prepositional uses on the pre-test were randomly picked to reoccur on the post-test, the remaining ones being totally new to the students. After the pre-test, the students were taught prepositions in
150 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
accordance with the three types discussed above, for which a special handout had been prepared. About 10 minutes were spent on each of the three prepositional uses, after which the subjects were asked to do some exercises, including assigning each use to one of three types tested and explaining why. The categorisation exercise was mimicked in the posttest, in which the subjects were again asked to name the type of prepositional use they had inserted/translated. The students were also given a number of sentences to work with at home and a comprehensive key that explained each prepositional use. Of course there is no guarantee that the students actually did their homework. In connection with the pre-test, the informants were also given a questionnaire, the purpose of which was to find out whether they had received any earlier instruction on prepositions and about their subconscious knowledge of the three categories that they were to be tested on. Even though the teacher in the A-level group claimed that he/she had discussed prepositions, none of the students remembered this, although they did recollect having worked with some handouts. The students in the B-level group, on the other hand, acknowledged that they had been taught prepositions but stated that they did not remember what had been said. More interestingly, when asked to rank the use of the preposition on in sentences (25)–(27) below, the students showed an intuitively good grasp of the three categories in focus. (25) I’ll see you on Friday. (26) There’s a box on the floor. (27) They commented on her article. In the A-group, 44% of the students thought that sentence (26) included the best example of the meaning of the preposition on. The corresponding figure for the B-level group was even higher (53%). Moreover, 38% of the A-level students and 53% of the B-level students believed that example (25) contained the second-best example. Lastly, 63% of the students in the A-level group and 80% of the subjects in the B-level group thought that the use of on in they commented on her article offered the least good example of the meaning of the preposition in question. One week after the lesson, the students did their post-test. Of the 17 first-year students, 13 were present at the pre-test and 16 at the post-test. All of the 13 second-year students included in the investigation did both tests. The results of the tests are presented in Tables 5.1 and 5.2, with Table 5.1 showing the results for the A-level students and Table 5.2 the results for the B-level students. Except for one case (basic prepositional uses in the A-level group), there was an increase in the number of correct answers between the preand post-tests. Part of the reason for this may naturally be due to the fact that 36 of the uses tested on the pre-test were also incorporated in the
Idiomatically Used Prepositions and Multi-Word Verbs 151
Table 5.1 Correct answers (%) for basic, systematic and idiomatic prepositional uses in the A-level group (Schmidt, 2004: 19) Basic uses
Systematic uses
Idiomatic uses
Pre-test
83.74%
68.78%
41.85%
Post-test
80.74%
74.52%
54.89%
Table 5.2 Correct answers (%) for basic, systematic and idiomatic prepositional uses in the B-level group (Schmidt, 2004: 19) Basic uses
Systematic uses
Idiomatic uses
Pre-test
84.98%
69.93%
49.28%
Post-test
85.38%
80.32%
57.81%
post-test, with the students perhaps having memorised these particular instances. As indicated in Tables 5.3 and 5.4, this is partially substantiated by the fact that participants sometimes found it difficult to categorise the prepositional uses, especially those of an idiomatic character. It must be remembered, however, that in addition to these concepts being entirely novel to the students, the time spent on instruction was very brief and there was, as mentioned above, no way of guaranteeing that the students had actually done the homework they were assigned. Given these two facts, it seems that presenting students with prepositions from a second-language perspective is far more promising than adhering to the classification given in contrastive grammar books. This technique could perhaps also be enhanced further for idiomatically used prepositions by Table 5.3 The number of correct answers, the number of possible correct answers, and the percentage of correct answers for the A-level group for the categorisation of prepositional uses task (Schmidt, 2004: 21) No. of correct answers
No. of possible correct answers
Correct answers
Basic
84
150
56.00%
Systematic
80
150
53.33%
Idiomatic
34
150
22.67%
Table 5.4 The number of correct answers, the number of possible correct answers, and the percentage of correct answers for the B-level group for the categorisation of prepositional uses task (Schmidt, 2004: 21) No. of correct answers
No. of possible correct answers
Correct answers
Basic
68
120
56.67%
Systematic
61
120
50.83%
Idiomatic
20
120
16.67%
152 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
considering the collocational frequency of the noun/verb + particle, starting by teaching more common units and only moving on to more rare ones after these have been learned. As mentioned before, this will be explored in the results section (Section 5.4). A pedagogical approach to multi-word verbs
‘Phrasal verbs are,’ as stated by Condon, ‘widely believed to be particularly difficult to master because the choice of verb-particle combinations seems so unsystematic’ (2008: 134), with particles displaying a semantically random nature and the whole unit a lack of transparency in meaning (Condon, 2008). When approaching multi-word verbs from the perspective of cognitive linguistics (CL), the main aim is therefore to try to systematise the structure of these kinds of items by explaining the motivations behind their use. In fact, quite a few experiments have already pointed in the direction that this may indeed be a beneficial approach when aiming at enhancing learners’ comprehension and retention of these types of lexical items (e.g. Boers, 2000; Condon & Kelly, 2002; Kövecses & Szabó, 1996; Kurtyka, 2001; Rudzka-Ostyn, 2003). Three main research questions were addressed in Condon’s large-scale study from 2008. Firstly, does the implementation of the ideas of CL when explaining the meanings of various phrasal verbs have positive effects on the students’ mastery of such multi-word items? Secondly, if positive effects can be seen, will the students be able to transfer the taught CL motivations when encountering other phrasal verbs in other contexts? Lastly, are there certain types of multi-word verb that are more suitable than others to be explained by a CL approach? The collection of data went on for two years, including both a pilot study (49 informants) and a main study (111 informants), both of which were structured in the exact same way. All the subjects selected for the experiment were French-speaking first-year students of economics at a Belgian university, and they were all, based on the results of a placement test, judged to possess an intermediate level of knowledge of the English language. The subjects were divided into four groups. While Groups 2 and 4 were used as experimental groups, the subjects in Groups 1 and 3 formed control groups. Each week all of the students had two consecutive hours of English class. For one hour the focus was on listening comprehension and vocabulary knowledge in a language laboratory, for which the students prepared by listening to a CD on which a number of phrasal verbs were encountered. The focus of the other hour was oral proficiency, for which the students also had to prepare, this time by reading short articles. For the sake of time management, Groups 1 and 2, forming Class A, did their time in the laboratory before their interactive hour, while the reverse was true for Groups 3 and 4, who formed Class B. At the very beginning of the first week, the students did a pre-test in the form of a gap-fill exercise targeting 30 phrasal verbs that the subjects
Idiomatically Used Prepositions and Multi-Word Verbs 153
were to pick from a list of 40 items. Condon opted for a semi-productive test, because fully productive tests on multi-word verbs have been seen to be extremely difficult for L2 learners and using a receptive test would have meant that the students could have guessed the correct meanings. Before taking the test, the students were given a brief explanation of what a phrasal verb is and were told that they did not have to worry about getting the tense right when filling in the verbs. Moreover, for each test sentence a translation into French was offered so as to eliminate any ambiguities. The particles in focus were out and up, which are the two most frequently used particles in multi-word verbs (Moon, 1997), and their opposites in and down. These were then divided into six semantically different types of phrasal verbs, ranging from highly literal to totally opaque items. During their time in the language laboratory, all of the students worked with half of these multi-word verbs (referred to as taught phrasal verbs) for eight weeks, with a maximum of two multi-word verbs during one and the same learning session. While all four groups spent the same amount of time on dealing with the multi-word verbs, only the experimental groups received instructions that were based on CL motivations, as exemplified with worn out: Tell students that worn is the past participle of wear. Remind students that they have already seen that when the participle out is used in a phrasal verb, there is an image of something moving out of a container or a box. They have also seen that boxes (containers) can be both concrete (e.g. a house or an office) and abstract (ignorance). In this sentence, the ‘box’ image is again an abstract concept. Give the students a few seconds to absorb what you have just said. Then, ask them to think about what sort of container might be implied in the above sentence. After a few seconds, tell them that the container is Existence or Life. When we die, we move out of that container. (Condon, 2008: 154)
At this point, diagrams were drawn to illustrate what had been said so far. The CL approach then continued by explaining that [s]ometimes when we are speaking we exaggerate. When we are very tired, we compare ourselves to dying (crever). That is what has happened here. The woman in this sentence was so tired after all her physical efforts that she felt as if she had left the container of Existence, that she was no longer alive, and so we have a phrasal verb with the word out. (Condon, 2008: 154)
The two control groups, on the other hand, followed a more traditional approach, in which the teacher provided the students with paraphrases and translations, as in the following: Tell students that worn is the past participle of wear. Wear on its own means porter. Explain to students that the verb in this sentence consists of worn out, and not simply worn. By adding the word out to the verb wore, the meaning of the construction as a whole is épuisé. Other words for worn out in this context are exhausted, or very tired. (Condon, 2008: 154)
154 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Furthermore, each week during the interactive hour the students were also faced with one or two new phrasal verbs (referred to as encountered phrasal verbs). These new verbs were presented in the same ways as described above. After the eight weeks, the students were again tested, once immediately after and once six weeks later. The subjects were not forewarned about either post-test. In the pilot study, the results showed that the experimental group in Class A outperformed their peers in the control group on both post-tests with regard to both categories of phrasal verbs (taught and encountered), with the difference in performance proven to be statistically significant only in the former case. It thus appears that these students benefitted from the CL-oriented instructions they received. This was, however, not the case for the experimental group in Class B. The difference in result may, according to Condon, be explained by the fact that that the experimental B-students, in contrast to the experimental A-students, encountered phrasal verbs first and were only later, during the lab hour, taught the CL motivations for the use of the particle in question. In fact, the lab hour was also the last lesson of the day for this group, which may also have had a negative effect. In the main study the experimental students in Group A also did better than the students in the corresponding control group, significantly so on the delayed post-test with the phrasal verbs that were taught. In Group B, on the other hand, as in the pilot study, the students in the control group did the best, particularly with the taught phrasal verbs on the immediate post-test. The order of the interactive hour and the lab hour thus seems highly relevant to the students’ performance. In general, it seems that a CL approach may indeed be beneficial to students who are trying to understand and memorise the meaning of multi-word verbs, provided that instruction of CL motivations comes before encountering the verbs. What is perhaps even more interesting is that this kind of approach appears to have long-lasting effects. Disappointingly, however, there was no evidence of strategy transfer; that is, the subjects in Condon’s experiment did not appear to be able to generalise to any great extent about the CL motivations taught in class. Previous studies show contradictory results in this area. While Kövecses and Szabó (1996) did notice signs of strategy transfer, there were no such indications in Boer’s (2000) study. The reason why no strategy transfer occurred in the Condon study may, according to the researcher, be due to the fact that the CL motivations presented to the subjects were simplified versions, and that the explanations therefore were not memorable enough. Furthermore, based on both the pilot study and the main study, the experimental groups seemed, in terms of long-term learning, to have improved their results significantly in connection with only two of the six types of semantic categories tested. These two types were both very literal in character; that is, it appears that CL motivations may be most useful when concerned with multi-word verbs that are comparatively transparent in
Idiomatically Used Prepositions and Multi-Word Verbs 155
character, while for the more abstract ones there are no obvious gains in retention. In view of this result, Condon stresses the need to develop CL motivations further so as to be able to create more tangible links between literal and more figurative uses of phrasal verbs. It may also be, Condon writes, that the more opaque multi-word verbs must be presented to students in a context consisting of more than just one single sentence. 5.3 The Parallel Tests Used for the Present Chapter
The two tests used in the present chapter (as usual one in English and one in Swedish) are both divided into two main parts, with the first part (40 items, one point/correct answer) focusing on idiomatically used prepositions and the second part (60 items, also one point/correct answer) dealing with multi-word verbs. As the reader will see, both parts are concerned with the students’ productive knowledge, to varying degrees. Of the 40 items making up the first main part, 20 items involve prepositions bound to nouns. Ten of these were tested in a gap-fill exercise in which the preceding nouns were offered in bold (Subpart Aa), as in (29) and (30): (29) The first day’s publication of the photos produced an incredible increase sales. (correct preposition: in) (30) The main reason general practitioners’ enthusiasm for counselling may well be a desire to reduce contact with and responsibility for a very demanding group of patients. (correct preposition: for) The remaining 10 were presented in sentences that contained incorrect prepositional uses (Subpart Ab). In this part, the informants were asked to correct the errors and offer an accurate alternative, as shown in (31) and (32): (31) *Demand after his services exceeded his capacity and in the second year he acquired a car to provide greater mobility. (correct solution: demand for) (32) *Our local representative will make every effort possible to allocate separate beds when this is particularly requested but it may at occasion be impossible. (correct solution: on occasion) The other 20 items making up the first main part focused on prepositions bound to verbs, tested in the same way as described above; the first 10 items (Subpart Ba) form a gap-fill exercise, as exemplified by (33) and (34): (33) In 1990 he was charged the murder of his third wife. (correct preposition: with) (34) Most parents usually succeed coming to terms with letting their children go. (correct preposition: in)
156 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
The remaining 10 contain prepositional errors (Subpart Bb), as exemplified by (35) and (36): (35) *In the next 30 years, Britain expects the number of people over 65 to increase with 20%. (correct solution: increase by) (36) *They’re blaming the crisis on the district health authority, who they accuse for not providing the hospital with sufficient money to do its job properly. (correct solution: accuse of) The English test items were all chosen because they are, for contrastive reasons, especially prone to errors, and they were randomly picked from errors made by previous student groups taught by the present author. The items used on the Swedish test part were (again randomly picked) from Hallström and Östberg (2001), Holmes and Hinchliffe (1994), and Montan and Rosenqvist (1992, 1993), all four of which are exercise books dealing with, among other things, Swedish prepositions from an L2 perspective. The context in which the prepositions were presented was taken, and often simplified, from the BNC and Språkbanken respectively (see Chapter 1 for more details on these corpora), or, in the case of the Swedish prepositions, from the books listed above. Furthermore, as part of the very definition of idiomatically used prepositions, there is normally only one possible solution for each noun and verb. Nevertheless, with a few items, more than one preposition was judged to be acceptable (see Tables 5.6 and 5.7). In such cases a descriptive approach, based on the BNC and Språkbanken, rather than a prescriptive one, was taken. On this first main part of the test, the informants were also asked to evaluate their knowledge of the 40 test items. For the gap-fill exercises (Subparts Aa and Ba), the students were asked to give information about the meaning of the noun/verb as well as how certain they were about their choice of preposition, as in (37): (37) The rumours he had heard didn’t tally _______________ the cool, distant woman he was looking at now. (correct preposition: with) TALLY □ I don’t know what the word means. □ I think I know what the word means. □ I am sure I know what the word means. THE PARTICLE □ I don’t know what particle to fill in. □ I think I know what particle to fill in. □ I am sure I know what particle to fill in.
Idiomatically Used Prepositions and Multi-Word Verbs 157
Thus, for the noun/verb offered, the students could either indicate that they did not know the word, that they thought they knew the word, or that they were sure of the meaning of the word. As for the particle they were asked to insert, the subjects again had three choices. They could indicate either that they did not know what preposition to use, that they thought they knew what preposition to use, or that they were certain of what preposition they should use. Similarly, for the sentences containing prepositional errors, or, as in (38) below, 9 where the students had to fill in a missing preposition (which only applies to this one test item), the subjects were also asked to evaluate their knowledge. (38) *We are grateful for the opportunity to comment the revision of this important planning guidance note. □ I can’t find the error. □ I think I know what the error is. □ I am sure I know what the error is. THE PARTICLE □ I don’t know what particle to use. □ I think I know what particle to use. □ I am sure I know what particle to use. Here the informants could indicate either that they could not find the error, that they thought they had found the error (by underlining the perceived error), or that they were certain about having found the error (again by underlining the perceived mistake). The options for whether they knew what preposition to use instead of the incorrect one were the same as for the gap-fill exercise described above. In the second main part of the test, the focus was, as mentioned above, on multi-word verbs, all of which are, in accordance with the criteria discussed in the theoretical background (Section 5.2), comparatively nontransparent in character. While all the items for the English test were picked randomly among phrasal verbs encountered by the present author while watching TV, reading fiction and listening to audiobooks, and among multi-word verbs brought up in the classroom with previous student groups, the items used on the Swedish test were, again randomly, picked from Luthman (2002), which offers an extensive list of multi-word verbs and their meanings and usage. Presented with the Swedish translation of the multi-word verb tested and the first letter of the stem verb to be used, the students were asked to fill in the correct phrasal verb in a given context. Of the 60 multi-word verbs tested, 45 of the English items were phrasal verbs that involved oneparticle verbs, while the rest (15 items) involved two-particle verbs. On the Swedish test, 52 items were one-particle verbs while only eight test items
158 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
required the students to fill in two-particle verbs. On both tests, it was clearly stated when a one-particle verb was sought and when a two-particle verb had to be used. Furthermore, in order for the students to be able to give their full attention to the meaning of the multi-word unit, they were, as in Condon’s (2008) study, told that they did not have to worry about getting the verb form and/or aspect correct. Examples of four of the English test items are offered in (39)–(42), the first two being examples of phrasal verbs with one particle and the other two exemplifying two-particle verbs. (39) She is not well today, so I am for her. (meaning: vikariera, f… + one particle) (correct multi-word verb: fill(ing) in) (40) He was due to marry this weekend, but now the wedding has been until he’s released from prison. (meaning: skjuta upp, p… + one particle) (correct multi-word verb: put off ) (41) Unfortunately, in all the confusion of evidence, it seems to have one brother’s word against the other. (meaning: kunna reduceras till, c… + two particles) (correct multi-word verb: come down to) (42) By doing it this way, I’m certain we will encourage females to these posts. (meaning: söka, p… + two particles) (correct multi-word verb: put in for) Along the same lines as described above, the students were asked in this part too to evaluate their knowledge of the items tested, as exemplified in (43): (43) We taxied back with the plane while cars blocked other traffic for us to . (meaning: lyfta, t… + one particle) (correct multi-word verb: take off ) □ I don’t know what verb to use.
□ I don’t know what particle to use. □ I think I know what verb to use. □ I think I know what particle to use. □ I am sure I know what verb to use. □ I am sure what particle to use. If faced with two-particle verbs, information about the second particle was requested too. Based on the two corpora used for the present investigation, all test items were presented in order of frequency, in each case starting with the most frequent one. As for the idiomatically used prepositions, it was the frequency of the combination of noun/verb + particle that was considered.
Idiomatically Used Prepositions and Multi-Word Verbs 159
Also, the frequencies of the multi-word verbs were based on their base form. The reason for this is, as discussed in Chapter 4 (Gardner, 2007), that phrasal verbs are as difficult to find in corpora as idioms. Furthermore, for each subsection (Aa, Ab, Ba, Bb (idiomatically used prepositions) and C (multi-word verbs)), the students were asked to evaluate the difficulty level of each main test part, the options being ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. Lastly, in order to be able to make fair comparisons between the test sets, the L1 and L2 frequencies of the test items were calculated, the items on the English test being in total more frequent (constituting ≈0.1187% of the BNC) than those used on the Swedish test (making up ≈0.0386% of Språkbanken). The same can be observed if the two main test parts are considered separately (Part A: L2: ≈0.0685%, L1: ≈0.0245%; Part B: L2: ≈0.0501%, L1: ≈0.0144%). This means that, if level of difficulty is equal to frequency alone, the English test should be easier than the Swedish test/test parts. On the other hand, the L1 test involved fewer multi-word verbs with two particles (eight versus 15), which may have made the Swedish test easier in that respect. The reader is here also reminded that whereas the BNC contains both spoken and written language, Språkbanken is only made up of written newspaper language. This has most likely affected the frequencies of the multi-word verbs tested, since these kinds of items are generally more frequent in spoken than in written language. None of the L1–L2 differences in frequency offered above were confirmed statistically, which means that comparisons may be made between the parallel tests/test parts. Still, the differences will be taken into consideration when discussing the results. 5.4 Results and Discussion
In the present section, the students’ results will be analysed thoroughly. In the first subsection, the informants’ mean scores and their selfevaluation of the various test parts will be discussed. The second subsection will explore to what extent the frequencies of the test items affected the learners’ knowledge and will contrast the students’ results on the idiomatically used prepositions with those on the multi-word verbs. The final subsection will look into the students’ individual results. The students’ results
Table 5.5 offers the reader an overview of the students’ results for the different parts of the tests in accordance with the description offered in the previous subsection. The total mean score achieved by the Swedish students on the English test (37.93 out of 97) indicates clearly, when contrasted with native speaker’s score (68 out of 97), the unequivocally difficult nature of the test items in
160 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 5.5 The students’ results on the tests on idiomatically used prepositions and multi-word verbs used in the present study Test part(s)
Students
Idiomatically used prepositions10 + multi-word verbs11
Native speaker of English
Idiomatically used prepositions
Multi-word verbs
N 1
Correctness rate
Mean
SD
Standardised scores
70.10% (=68/97)
–
–
–
Swedish students – English test
15
39.12% (=569/1455)
37.93
13.24
Highest: 2.12 Lowest: −1.51
Swedish students – Swedish test
15
62.36% (=926/1485)
61.73
12.07
Highest: 1.85 Lowest: −1.80
94.59% (=35/37)
–
–
–
Native speaker of English
1
Swedish students – English test
15
63.60% (=353/555)
23.53
6.67
Highest: 1.57 Lowest: −1.43
Swedish students – Swedish test
15
79.00% (=474/600)
31.60
5.32
Highest: 1.58 Lowest: −1.99
55.00% (=33/60)
–
–
–
Native speaker of English
1
Swedish students – English test
15
24.00% (=216/900)
14.40
7.26
Highest: 2.42 Lowest: −1.57
Swedish students – Swedish test
15
51.07% (=452/885)
30.13
7.54
Highest: 1.97 Lowest: −1.48
question; that is, the present investigation corroborates earlier studies that have shown that these are indeed error-prone areas. Also, the native speaker did not only surpass the Swedish students on the test as a whole but also performed considerably better on each subpart. For idiomatically used prepositions, the native speaker achieved a correctness rate of 94.59% (35 out of 37), which can be compared to the correctness rate achieved by the Swedish students of 63.60% (23.5 out of 37). For the multi-word verbs, the native speaker produced 55.00% correct answers (33 out of 60), while the Swedish students only got 24.00% right (14.4 out of 60). As for the Swedish students’ scores on the idiomatically used prepositions, a comparison can be made with the results presented in Karlsson (2002), in which the university students achieved 95.99% correct answers (first term), 96.45% (second term), and 96.88% (third term) when writing compositions, where avoidance of unknown structures is possible, and 97.08% (first term), 93.58% (second term), and 93.72% (third term) when translating, where avoidance is not possible to the same extent. It appears that the gap-fill and correction exercises focused on in the present study are infinitely more difficult than the test types used in the Karlsson study from 2002. The scores achieved in the present investigation can also be compared with the results seen in Schmidt (2004). In her study, first- and second-year upper secondary school students received percentages of 41.85% and 49.28% respectively (25 items tested) before instruction took
Idiomatically Used Prepositions and Multi-Word Verbs 161
place, and 54.89% and 57.81% respectively after instruction. The lower scores of the informants in the Schmidt study can most likely be attributed to the fact that these students were upper secondary students having studied English for far less time than the subjects in the present investigation. The students’ score on the phrasal verbs can be compared with the results seen in Olejniczak (2006), where the two test groups consisting of second-year university students achieved percentages of 33.1% and 31.4% correct answers respectively, and where the most advanced informant group received a percentage of 52.1% correct answers. The Swedish subjects in the present investigation, first-term university students, thus received higher scores than the first-term students in Olejniczak’s study. This may be due to the fact that a phrasal verb is a concept that Swedes are much more familiar with than Polish learners, in whose first language the phenomenon does not exist at all (Swan & Smith, 2001). It is also interesting to compare the scores achieved for the phrasal verbs in the present chapter with the scores for the idioms tested and discussed in Chapter 4, where the subjects received a percentage of 47.00% correct answers (37.60 out of 80) for the L2 items and 87.00% (69.60 out of 80) for the L1 items, indicating that phrasal verbs may be more difficult units of vocabulary than idioms are. It must be remembered, however, that whereas all of the multi-word verbs tested in the present chapter were more or less non-transparent in character, there were many idioms that were categorised as either semi-transparent or fully transparent. Also, whereas the phrasal verbs in the present chapter were tested in a semiproductive way, the idioms were tested receptively. As discussed in Chapter 2, students most often show a higher level of receptive knowledge than of productive knowledge. As expected, and confirmed statistically at a significance level of 5%, the Swedish students also achieved considerably higher total scores in their native language than their L2 (61.73 out of 99 (62.36%) as compared to 37.93 out of 97 (39.12%) on the English test). Equally unsurprisingly, and again confirmed statistically at a significance level of 5%, the same trend can also be seen if the two main parts are considered separately (all idiomatically used prepositions on the Swedish test: 31.60 out of 40 (79.00%) versus 23.53 out of 37 (63.60%) on the English test; multi-word verbs on the Swedish test: 30.13 out of 59 (51.07%) versus 14.40 out of 60 (24.00%) on the English test). The validity of these L1–L2 differences is enhanced even further due to the fact that, based on frequency, the English test was more difficult than the Swedish one. The fact that the Swedish students achieved lower scores in their mother tongue than the native speaker of English did in his may also be explained by the fact that the items included in the Swedish test were, from a frequency point of view, more difficult than the items in the English test. Considering this difference, it seems that the native speaker on the one hand and the informant group on the other hand did almost as well
162 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
in their respective mother tongues, both on the test as a whole and on the two main parts separately. Interestingly, the same pattern was also discerned with the results on the idiom/proverb tests discussed in Chapter 4. The students appear to have a good grasp on their mastery of idiomatically used prepositions and multi-word verbs in English. On all subparts on the English test, the majority of the students either indicated they were ‘difficult’ (one student) or ‘very difficult’ (13 students),12 more so on multiword verbs than idiomatically used prepositions. (This intra-language difference in result was also confirmed statistically at a significance level of 5%.) The Swedish multi-word verbs were also considered to be difficult by the students, with the majority of the subjects indicating here too that the test items were either ‘difficult’ (six students) or ‘very difficult’ (three students). The idiomatically used prepositions, on the other hand, were, as the test results show, regarded as easier than the phrasal verbs, with only 1 student indicating that the task was ‘difficult’ and the rest thinking that these items were either ‘very easy’, ‘easy’, or ‘average’. (Again, this intralanguage difference was confirmed statistically at a significance level of 5%.) That there is an L1–L2 difference is also shown in the correctness rates in the students’ evaluation of their own knowledge, both for items indicated as known and those that were thought to be known, as seen in Tables 5.6 (L2 test) and 5.7 (L1 test) for idiomatically used prepositions and in Tables 5.9 (L2 test) and 5.10 (L1 test) for multi-word verbs. In the former case, there is a correctness rate of 74.4% (known) and 60.3% (thought to be known) on the L2 test, which can be compared to 90.5% (known) and 68.1% (thought to be known) on the L1 test. In the latter case, there is a correctness rate of 86.5% (known) and 44.7% (thought to be known) on the L2 test, which can be contrasted with 89.7% (known) and 69.2% (thought to be known) on the L1 test. Results on the individual idiomatically used prepositions and multi-word verbs
Although, when comparing the scores obtained for the first five idiomatically used prepositions (the most frequent items) with the remaining five uses (the least frequent items) on each test part, a (very) slight frequency effect can be observed in some cases,13 the conclusion that must be drawn is that high-frequency items are generally not much better known than low-frequency items in either language. Hence, frequency, at least on its own, does not appear to be a determining factor as to whether idiomatically used prepositions are mastered or not. (The reader is here reminded that it is the frequency of the combination of noun/verb + particle that is considered.) This also means that frequency, which seemed to be such a promising pedagogical tool in the Mueller (2011) study, where it was shown that students’ knowledge of high-frequency types was considerably better than of low-frequency types of the same sense (e.g. laugh at/
Idiomatically Used Prepositions and Multi-Word Verbs 163
wink at), does not appear to be equally useful when different prepositional senses are compared. Figures 5.2 and 5.3 offer the number of correct answers for each multiword verb in the English and Swedish tests respectively.14 As pointed out before, these were presented to the students in a frequency order, with the most frequent item appearing first. Also, the results offered in bold are items found on West’s (1953) general service list, indicating the general commonality of many multi-word verbs. (No corresponding list could be found for Swedish multi-word verbs.) The students’ results on the multi-word verbs paint a totally different picture than the one described in connection with the idiomatically used prepositions. In both languages, especially in the students’ L2, the impact of frequency is considerable. This is true for both one- and two-particle verbs. The reason there is such a strong frequency effect here but hardly any on the students’ mastery of idiomatically used prepositions will be discussed in connection with Tables 5.6 and 5.9. Tables 5.6 and 5.7 show the students’ results for each idiomatic prepositional use, with Table 5.6 presenting the English test part and Table 5.7 the Swedish test part. (For translations of the L1 items, see Appendix 2.) In both cases, the items are placed according to the number of correct answers, starting with the item that received the highest score. Furthermore, in both tables the prepositional uses are colour-coded. The test items that remain Total no of correct answers for each multi-word verb 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
5
10
15
20
25
30
35
40
45
50
55
60
Multi-word verb
Figure 5.2 The total number of correct answers for each of the 60 multi-word verbs on the English test
164 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
5
10
15
20
25
30
35
40
45
50
55
60
Multi-word verb
Figure 5.3 The total number of correct answers for each of the 60 multi-word verbs on the Swedish test
in white represent cases in which the students were offered the stem noun (Aa1–Aa10) or verb (Ba1–Ba10) and asked to fill in the correct preposition. The test items that are shaded in dark grey represent cases in which the students were asked to indicate the error and provide the correct preposition (nouns: Ab1–Ab3, Ab6–Ab8, Ab10; verbs: Bb1–Bb10). For example, for the verb succeed, 13 of the 15 students knew the correct bound preposition. (This is item Ba4, the fourth most common of the ten verb–particle combinations tested for which the students were to insert the correct preposition. It is also the 20th most common of all the idiomatic uses tested (shown in the first column in parentheses).) For this specific item, the majority of the students (11) indicated that they knew the meaning of the verb offered (see column KN under ‘Info about noun/verb’, mostly coloured in grey), while three students stated that they thought they knew its meaning and 1 that the verb was unknown. As for the preposition, the majority of the students (nine, one of whom was wrong) indicated that they thought they were sure of what preposition to insert (see column KN under ‘Info about particle’, again in grey). Only five stated that they were certain of what preposition to use, and one that the particle was unknown. Also, as discussed in Section 5.3, for a few items more than one preposition was accepted. All the alternatives and their individual scores are offered in the column titled ‘Noun/verb + particle’. For item Aa9, for example, having received a total score of 13 points, eight students offered cheque of and five cheque for.
Idiomatically Used Prepositions and Multi-Word Verbs 165
Table 5.6 The students’ results on and evaluation of idiomatically used prepositions on the English test, with the test items listed in order of correct answers, starting with the item that received the highest score. (Information about the noun/verb: KN = for the gap-fill task: the item is known, and also for the correction task: the incorrect prepositional use has been found; TK = for the gap-fill task: the item is thought to be known, and also for the correction task: think the incorrect prepositional use has been found; NK = for the gap-fill task: the item is not known, and also for the correction task: the incorrect prepositional use has not been found.) (Information about the particle: KN = the gap-fill task/the correction task: the particle to fill in/use is known; TK = the gap-fill task/the correction task: think the particle to fill in/use is known; NK = the gap-fill task/the correction task: the particle to fill in/use is not known.) Figures in parentheses in columns KN − NK indicate cases where the students were incorrect in their self-evaluation No. (frequency order) Ba8 (34) Aa10 (26) Ba6 (27) Aa1 (1) Aa9 (35) Ba4 (20) Ba5 (22) Aa2 (2) Aa9 (25) Ab7 (19) Aa7 (21) Ba3 (16) Ba2 (14) Aa6 (13)
Info about noun/verb KN TK NK 12 2 12 15 11 11 12 13 11 4 14 11 11 12
4 3 2 1 1 9 (2)
Aa4 (4) Bb8 (31) Ab6 (15) Aa8 (24) Bb4 (17) Bb7 (30) Bb2 (11)
11 6 3 (1) 8 4 (1) 3 6 (1)
3 5 (1) 11 (2) 3 7 (1) 9 (3) 4 (1)
1 4 (1) 1 3 4 (1) 2 5 (1)
Ab3 (10) Bb1 (7) Aa5 (9) Bb9 (33) Aa3 (3) Ab1 (5) Bb5 (18) Ba1 (8) Ab8 (23) Ab10 (29) Bb3 (12) Ab2 (6) Bb6 (28) Bb10 (36) Ba10 (37) Ba7 (32) Total Aa + Ba Total Ab + Bb
4 (2) 2 7 3 12 2 (1) 5 (1) 11 1 4 (2)
8 (2) 9 (3) 5 8 (4) 1 10 (3) 8 (5) 3 11 (6) 8 (4) 11 (7) 10 (7) 8 (6) 6 (4) 3 1 45 142 (61) 57.0%
3 (1) 3 (1) 3 3 (1) 1 3 2 1 3 3 3 3 4 5 6 13 49 53 (7)
Correctness rate Ab + Bb
2 (1) 3 (2) 4 (3) 6 202 56 (15) 73.2%
3 1 2
3 4 2
12 1 1 1 1 2 2 (1) 1 1 1
86.8%
Noun/verb + particle (bold) (correct answers)
remind(s) me of (15) surplus of (9)/in (4)/on (2) (15) operate on (14) end to (11)/of (3) (14) prevent them from (13) succeed in (13) agree on (13) increase in (noun + particle) (13) cheque of (8)/for (5) (13) map of (12) secret of (7)/to (5) (12) apply for (12) agree to (12) criticism about/of/for/on/against/at/with15 (12) advantage of (9)/with (2)/to (1) (12) translate into (11) shortage of (11) obstacle in (6)/to (4)/for (1) (11) suffer from (10) congratulate someone on something (10) prefer something to (5)/over (4) something (9) key to (9) apply to (9) fall in (noun + particle) (9) accuse someone of something (8) reason for (8) visit to (8) keen on (7) charged with (6) on occasion (6) suburb of (6) comment on (4) demand for (noun + particle) (4) increase by (verb + particle) (3) ply(ing) someone with something (3) run up16 (a debt) (3) tally with (2)
Info about particle KN TK NK 10
6 5 8 (1) 5 5 6 (1) 3 5 5 4 3 4 (1)
4 8 6 8 (1) 6 (1) 9 (1) 9 (2) 7 (1) 9 (1) 7 (1) 6 (2) 10 (3) 10 (3) 7 (2)
1 (1) 7 (7) 3 (2) 2 (2) 1 (1) 1 1 (1) 2 (2) 2 (1) 3 (1) 4 (3) 1 (1) 2 (2) 3 (3)
2 4 2 1 4 (1) 2 3 (1)
9 (2) 7 (1) 12 (4) 7 (2) 7 (1) 9 (3) 7 (1)
4 (3) 4 (1) 1 (1) 6 (5) 4 (1) 3 (1) 5 (1)
3 (1) 1 2 (1) 3 4 (2) 3 (2) 3 5 (1) 3 (1) 3 (2)
8 (3) 8 (2) 8 (4) 7 (3) 7 (2) 8 (2) 9 (6) 9 (7) 6 (2) 8 (3) 9 (5) 8 (5) 7 (6) 9 (6) 5 (5) 5 (4) 149 136 (54) 60.3%
4 (2) 5 (2) 5 (4) 4 (1) 3 3 (1) 3 (1) 1 6 4 5 7 (1) 5 (1) 5 8 (3) 9 (1) 66 71 (15) 78.7%
3 (2) 1 (1) 2 (2) 1 (1) 81 43 (11) 74.4%
As the reader can see, the students’ relatively low mean score on idiomatically used prepositions in their L2 is clearly reflected in the results presented in Table 5.6. For instance, while only nine of the uses tested received the three highest scores (15–13) on the English test, 17 received the same high scores on the Swedish test (Table 5.7). Similarly, whereas none of the Swedish prepositional uses received 0–5 points, six of the English test items received these low scores.
166 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 5.7 T he students’ results on and evaluation of idiomatically used prepositions on the Swedish test, with the test items listed in order of correct answers, starting with the item that received the highest score. (Information about the noun/verb: KN = for the gap-fill task: the item is known, and also for the correction task: the incorrect prepositional use has been found; TK = for the gap-fill task: the item is thought to be known, and also for the correction task: think the incorrect prepositional use has been found; NK = for the gap-fill task: the item is not known, and also for the correction task: the incorrect prepositional use has not been found.) (Information about the particle: KN = the gap-fill task/the correction task: the particle to fill in/use is known; TK = the gap-fill task/the correction task: think the particle to fill in/use is known; NK = the gap-fill task/the correction task: the particle to fill in/use is not known.) Figures in parentheses in columns KN − NK indicate cases where the students were incorrect in their self-evaluation No. (frequency order) Bb8 (35) Ba3 (17) Ba5 (23) Bb9 (37) Bb4 (24) Ba1 (6) Bb5 (31) Bb1 (10) Aa9 (15) Ba7 (30) Bb10 (40) Bb3 (20) Bb6 (32) Ba4 (22) Aa3 (7) Aa5 (9) Aa4 (8) Ba10 (39) Ab5 (19) Ba9 (38) Aa8 (14) Aa7 (13) Aa2 (5) Ab3 (3) Ba2 (16) Ab1 (1) Ba8 (33) Bb7 (34) Ab8 (27) Ab10 (36) Aa6 (11) Ab2 (2) Bb2 (18) Ab9 (29) Ba6 (26) Ab4 (12) Aa10 (28) Aa1 (4) Ab7 (25) Ab6 (21) Total Aa + Ba Total Ab + Bb Correctness rate Ab + Bb
Info about noun/verb KN TK NK 13 2 14 1 13 2 13 2 10 5 12 3 11 3 1 12 3 (1) 13 2 12 3 12 3 (1) 9 6 (1) 11 (2) 4 13 2 11 3 12 3 13 2 14 1 9 4 (1) 2 12 3 12 3 13 2 10 3 8 (1) 4 2 12 3 8 5 (2) 2 12 3 9 (1) 6 (3) 8 (1) 4 (1) 2 7 7 (3) 1 12 3 10 (2) 4 (2) 1 9 (2) 6 (3) 8 (3) 5 (1) 2 11 4 7 (2) 7 (4) 1 12 2 10 4 7 (3) 8 (5) 8 (4) 6 (4) 1 243 52 0 189 (21) 94 (32) 15 (0) 88.9% 66.0% 100%
Noun/verb + particle (bold) (correct answers) frysa om (händerna) (15) höra av sig (15) rätta sig efter (15) nicka(de) åt (7)/mot (4)/till (4) (15) akta sig för (15) ändra på (15) svära på (14) strunta i (14) (det nya) numret av… (tidskriften) (14) spionera på (14) misstänka(er) någon för något (14) förbereda sig för (8)/inför (6) något (14) vända kappan efter vinden (13) lida av (13) (bli) föremål för (13) känsla för (13) ansökan om (13) gripa någon för något (12) av erfarenhet (12) tala för (10)/inför (2) döva öron (12) ta initiativ till något (12) under kontroll (12) orsak till (12) ha hopp om (12) hyra ut (12) brist på (11) duga till något (11) svänga(er) om (orkestern) (11) ur funktion (11) synonym till (11) avtal om något (11) intresse för något (10) bestämma sig för något (10) ha anlag för (sjösjuka) (9) lida för något (9) chans till (8) ha en åsikt i frågan (7) efterfrågan på (7) (det finns) tendens(er) till att (7) (se) en fördel med ngt (6)
Info about particle KN TK NK 13 2 12 3 11 4 11 4 11 4 10 5 11 3 1 11 4 (1) 11 (1) 4 10 5 (1) 9 6 (1) 9 6 (1) 12 (2) 3 11 (1) 3 (1) 8 6 (1) 10 (2) 5 7 (2) 7 11 (1) 4 (2) 9 4 (1) 2 10 (1) 5 (2) 9 5 (3) 1 (1) 10 (2) 5 (1) 8 (1) 6 (1) 7 (1) 5 2 5 8 (1) 1 7 6 (2) 2 9 (2) 6 (2) 8 (1) 7 (3) 8 (1) 4 (1) 3 (1) 6 8 (3) 1 5 (1) 10 (3) 8 (2) 6 (2) 1 7 (1) 8 (4) 7 (2) 6 (2) 2 7 (3) 8 (3) 3 (1) 11 (5) 1 9 (4) 5 (3) 5 (2) 8 (4) 1 4 (2) 9 (5) 2 (1) 7 (3) 7 (5) 1 178 112 3 168 (16) 113 (36) 18 (2) 90.5% 68.1% 88.9%
If the students’ results on the English test are considered (Table 5.6), it can be observed that the test items for which the subjects had to find and correct errors (indicated in dark grey) were inherently more difficult than those for which the informants had to insert missing prepositions. In the Swedish material, on the other hand, there is no such clear difference between these two types, indicating that the learners are more sensitive to
Idiomatically Used Prepositions and Multi-Word Verbs 167
Table 5.8 A comparison between the students’ results on the gap-fill task and their results on the correction task Test part(s)
Students
Noun/verb + preposition (Aa + Ba) (filling in)
Native speaker of English
Noun/verb + preposition (Ab + Bb) (correcting)
N 1
Correctness rate
Mean
SD
Standardised scores
95.00% (=19/20)
–
–
–
Swedish students – English test
15
74.00% (=222/300)
14.80
3.12
Highest: 1.67 Lowest: −1.86
Swedish students – Swedish test
15
80.67% (=242/300)
16.13
2.42
Highest: 1.60 Lowest: −2.12
94.12% (=16/17)
–
–
–
Native speaker of English
1
Swedish students – English test
15
50.98% (=130/255)
8.67
4.32
Highest: 1.47 Lowest: −1.31
Swedish students – Swedish test
15
77.33% (=232/300)
15.47
3.27
Highest: 1.39 Lowest: −1.67
the degree of the productive nature of the test in their L2 than in their native language. This is made even clearer in Table 5.8. (The difference between filling in versus correcting was confirmed at a significance level of 5% in the learners’ L2, while it could not be confirmed in their L1. The L1–L2 difference was confirmed statistically at a significance level of either 5% (filling in) or 10% (correcting).) Another clearly noticeable difference can be observed if the students’ evaluation of their knowledge of the stem noun/verb on the one hand and the particle on the other hand is considered. On the Swedish test (Table 5.7), the majority of the students indicated that they were certain not only of the meaning of the noun/verb (all cases, shown in grey, column KN under ‘Info about noun/verb’) but also about what preposition to insert/ replace (33 out of 40 cases, also in grey). On the English test (Table 5.6), on the other hand, where most noun/verb stems were also known to the students, the overwhelming majority of the informants stated that they either only thought they knew what particle to fill in/replace or simply did not know at all. It thus appears that the noun/verb + particle combinations in Swedish may be psychologically real units that are stored together so that, when prompted with a certain noun or verb, the learners automatically think of the bound preposition in question. This does not appear to be the case with the noun/verb + particle combinations in English; that is, it does not seem that these combinations are tightly knit units stored together in the learners’ L2 mental lexicon (at least not at this stage). It should be added that, had a strictly prescriptive approach been adhered to, this conclusion would have been strengthened even further. There are several reasons why noun/verb combinations may not be perceived as psychological units by second language learners. One reason is of course that bound prepositions are not always present when stem
168 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
nouns and verbs are used (e.g. a clear increase could be shown (increase in), he had to operate immediately (operate on)). Also, in addition to receiving less input in general than native speakers, L2 learners often receive input from other L2 speakers, input that in many cases may be erroneous. Moreover, since prepositional errors rarely cause misunderstandings, native speakers generally do not react to these kinds of errors and, as a consequence, do not make corrections. Furthermore, as seen in the present investigation, for some nouns and verbs there may be more than one preposition possible, thus blurring the concept of idiomatic prepositional use. Additionally, input that comes from native speakers often contains prepositions in unstressed positions, making the particles difficult to hear (Mueller, 2011). Lastly, when browsing through some word lists in textbooks used in Swedish upper secondary schools today, the present author noticed that it is the exception rather than the rule that bound prepositions are presented together with their stem noun and verb. It is quite clear that all the above contribute to making noun/verb + preposition combinations into non-units, thus rendering collocational frequency useless when different prepositional uses are tested; that is, collocational frequency has in this context no pedagogical value. Here, rote memorisation should therefore not be discarded (Nation, 2001). Although very often thought of as an old-fashioned learning and teaching technique, it certainly has its merits. The important thing is that L2 learners need to start perceiving these combinations as tightly knit units so that, as in their L1, when prompted with a noun or verb in a context that requires a bound preposition, they automatically think of the correct one. As pointed out by Martin (1984: 132), ‘[N]ative speakers grow up absorbing these idiomatic collocations, combinations that elude most students who have not been exposed to massive amounts of the language.’ An analysis of the errors made by the students was also performed. This showed that interference appeared to play as insignificant a role as frequency, in that, with a few exceptions, no clear majority of transferinduced errors could be discerned. Rather, the students’ insecurity shines through via a jumble of suggestions that cannot be categorised as more than random guesses. Quite naturally, though, the more uncertain the students were of the meaning of the stem noun/verb, the greater the variation of erroneous prepositions. Tables 5.9 (the English test part) and 5.10 (the Swedish test part) present the students’ results on multi-word verbs in accordance with the number of correct answers, starting with the test item that received the most correct answers.17 These two tables should be deciphered in essentially the same way as Tables 5.6 and 5.7 discussed above. For instance, for test item 32 (turn down) – that is, the 32nd most frequent item based on the BNC (column one) – 11 of the 15 students supplied the correct answer. Here an equal number of students indicated, and were correct in their evaluation, that either they were sure they knew what verb stem to use (five
Idiomatically Used Prepositions and Multi-Word Verbs 169
students, column KN under ‘Info about verb’) or they thought they knew what verb stem to use (six students, one of whom was wrong, column TK under ‘Info about verb’). Another two subjects wrote that they could not come up with the targeted verb stem (column NK under ‘Info about verb’). As for the preposition, while the majority (five (six in the manuscript I sent in) of the 15 students) wrote they were certain that down should be used (column KN under ‘First particle’), five indicated that they only thought they knew what particle to use (column TK under ‘First particle’), and an additional three that they did not know the answer, one of whom actually did supply the correct particle (column NK under ‘First particle’). As before, when in majority, the squares are indicated in grey. Furthermore, in a few cases more than one phrasal verb was possible. All the correct solutions and their individual scores are listed in the column that indicates the multi-word verb in question. For example, for test item C24 both bring out (12 correct answers) and beef up (not offered by any of the Swedish students but used by the native speaker) were judged to be correct. Once again, the students’ low mean score is reflected in the results for the individual test items. For instance, whereas 11 of the multi-word verbs on the Swedish test received 15, 14, or 13 correct answers, only one phrasal verb on the English test received the same high score (cut down/cut back (13 points)). (It is interesting to note that this item is among those that are least opaque in character. It is also one of the phrasal verbs in Mondor’s (2008) study that was noticed to be overused, with the reason for its overuse thought to be the existence of a direct Swedish counterpart.) Similarly, whereas only 12 of the phrasal verbs on the Swedish test received 0–3 points, the vast majority (41) received these same low scores on the English test. Furthermore, with multi-word verbs, in contrast to what could be observed with the idiomatically used prepositions, the students’ knowledge of the stem verb on the one hand and the particle on the other hand appears, to a greater extent, to go hand in hand. That is, when the stem verb is known, the particle is also known (very few cases). When the students indicate that they think they know the stem verb, they also tend to indicate that they think they know the particle (slightly more cases). Finally, and less surprisingly, when the stem verb is unknown, the particle is also unknown (in the vast majority of cases). This result aligns with the basic definition of a phrasal verb, in which neither the stem verb nor the particle can be replaced without a change in meaning. On the other hand, this is not usually the case with idiomatic prepositional use, where an incorrect choice of preposition only affects the intended meaning on rare occasions (e.g. agree on/agree to). The Swedish students here seem to have understood that the concept of a phrasal verb in English is a tightly knit semantic unit, a perception that is most likely based on their native speaker experience, Swedish being a language that makes use of multi-word verbs to a great extent. The idea
4 3
1 6 (3) 1
C14 C15 C24 C18 C32 C13 C30 C26 C7 C22 C4
C29 C1
C31 C23 C11 C25 C10 C3 C27 C48 C47 C20
2 (1) 1 3 1 1
KN 5 5 3 6 5 4 3 1 3 (1) 1 3
No.
7 (2) 6 (3) 6 (3) 5 (1) 7 (3) 8 (4) 7 (1) 9 (4) 5 (3) 7
6 4
Info about verb TK 8 7 10 4 6 (1) 5 (1) 6 9 10 (2) 7 8
7 (1) 14 7 (4) 9 (2) 4 (1) 1 2 4 (1) 8 6 (1)
3 (1) 6 (3)
4 3 (2)
4 (1) 2 3 (1) 5 (1) 2 (1)
2
NK
Multi-word verb (correct answers) cut down (13), cut back (0) (13) fill(ing) in (for a person) (12) bring out (12), beef up (0) (flavour) (12) take off (airplane) (11) turn down (=say no to) (11) turn out (=end) (11) bring down (11), blow away (0) (=shoot) (11) come down to (11) get on with (5), get down to (3), get on to (1) (9) come off (=come loose) (9) take up (took up) (8), take to (took to) (0) (=start to spend time on) put someone through (phone) (7) come to (came to) (5), came (a)round (came (a)round) (2) (7) fall out (fell out) (=become enemies) (6) put off (=postpone) (5) put out (5), put about (0) (=spread) (5) set against (=weigh against) (5) come across (2), come by (2) (4) get(ting) on (4) pull in (2), pull up (1) (3) come down with (a cold) (3) hit it off (3) make out (cheque) (3) 1 1 (1)
1 (1) 3 5 (4)
6 (2) 6 (4) 5 (4) 5 (1) 8 (5) 9 (3) 6 (4) 9 (6) 4 (2) 8 (4)
5
3 1 4 1
6 (3)
8 (5)
4 4
First particle TK 11 (2) 7 (1) 9 6 (1) 5 7 (2) 7 9 (1) 10 (2) 8 (1)
KN 3 6 5 (1) 5 6 4 2 3 3
8 (1) 4 7 (4) 9 (1) 5 (1) 1 2 4 8 6
7 (1)
3
3 (1)
4 (2) 3 (1) 4 (2) 6 (2) 2 1 6 (2)
2
NK
3 4
KN
5 (2)
6 7 (2)
6 (1)
1
Second particle TK NK
Table 5.9 The students’ results on and evaluation of multi-word verbs on the English test, with the test items listed in order of correct answers, starting with the item that received the highest score. (Information about the verb: KN = for the gap-fill task: the item is known, and also for the correction task: the incorrect prepositional use has been found; TK = for the gap-fill task: the item is thought to be known, and also for the correction task: think the incorrect prepositional use has been found; NK = for the gap-fill task: the item is not known, and also for the correction task: the incorrect prepositional use has not been found.) (Information about the particle: KN = the gap-fill task/the correction task: the particle to fill in/use is known; TK = the gap-fill task/the correction task: think the particle to fill in/use is known; NK = the gap-fill task/the correction task: the particle to fill in/use is not known.) Figures in parentheses in columns KN−NK indicate cases where the students were incorrect in their self-evaluation
170 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Correctness rate
C50 C56 Total
C54 C57 C44 C46 C2 C52 C33 C39 C40 C12 C38 C49 C53 C60 C58 C59 C43 C41 C45
C36 C8 C35 C16 C42 C55 C21 C51 C34 C37 C6 C5 C17 C19 C9 C28
78 (13) 83.3%
1 (1)
1 (1)
1 (1)
2 (1)
2 2
1 1 1 1 (1) 3 (3)
4 (4) 5 (5) 343 (125) 63.6%
2 (2) 4 (4) 2 (2) 2 (2)
2 2 (1) 3 (2) 7 (6) 6 (6) 6 (3) 9 (3) 7 (2) 5 6 (3) 4 (1) 5 (3) 4 (2) 3 (2)
8 (4) 8 (6) 11 (4) 4 7 (5) 8 8 (5) 2 3 (2) 6 (5) 8 (7) 3 5 (2) 6 5 (1) 6 (3)
9 9 393 (71) 81.9%
11 11 11 (1) 6 7 (2) 6 (4) 4 (1) 7 (2) 8 (3) 9 (2) 9 (3) 9 (1) 8 9 (2) 14 12 10 10 (1) 10 (1)
6 (1) 6 (1) 2 9 (2) 6 (4) 5 (2) 2 (1) 12 9 7 4 (2) 7 (3) 8 (5) 8 (4) 10 (2) 7 (1)
catch on to (=understand) (3) set off (down the street) (3) get around (=move about) (2) take to (took to) (2) (=start to like) fall through (=fail) (2) cut in (=interrupt) (2) set (my dog) on (someone) (2) ring off (rang off) (2) round up (2), rustle up (0) (2) The play never caught on (2) set in (rainy season) (=start) (2) come into (fortune) (1) bring about (=achieve) (1) take out (insurance) (1) do (done) someone in (1) put something aside (1), put something by (0) (=evade) (1) make off with (1) wolf down (1) dish out (1), dish up (0) (=serve) (1) dish out (=give) (1) set up (1), strike up (0) (friendship) (1) put forth (strength) (=summon) (0) come in for (criticism) (=receive) (0) put across (=convey) (0) put down to (=accuse of, attribute to) (0) put someone out (0) (=bother) come out in (spots) (0) fall in with (=abide by, agree to) (0) catch out (=uncover) (0) put it over on (=fool) (0) fall in (up)on (=visit unexpectedly) (0) fob(bed) someone off with (a glare) (=dismiss) (0) live down (a criminal past) (=rehabilitate after) (0) put in for (=apply for) (0) do (did) someone out of, diddle (diddled) someone out of (0) boil down to (0) bump off (0)
44.7%
74 (10) 86.5%
91.0%
10 9 390 (35)
10 3 (3)
4 (2) 3 (3) 3 (3)
9 8 (1) 12 (5) 7 (1) 8 (2) 7 5 6 8 8 10 10 (1) 8 10 13 9 8 9
7
6 (5) 3 (2) 3 (1) 1 5 (2) 5 (4) 6 (6) 9 (9) 8 (8) 5 (5) 6 (6) 3 (3) 3 (3) 4 (3) 3 (3)
6 (1) 7 (2) 4 8 9 4 3 (1) 11 8 7 6 (2) 7 7 8 (1) 9
8 (5) 8 (3) 8 (7) 5 (4) 4 (4) 9 (6) 11 (5) 2 3 (2) 6 (5) 9 (9) 5 (4) 6 5 (5) 6 (5)
3 (2) 4 (3) 349 (193)
1 (1)
1
1 (1) 1 1 (1)
1
1 1 1
77.8%
9 (2)
1 (1)
1 (1)
42.6%
54 (31)
89.5%
76 (8)
5
10
6 (1)
2 (1) 2 (2)
5 5 5 (2)
8 6
2 (2) 3 (3) 2 (1) 1 (1) 4 (2)
5 (2)
5
5
4 (2)
5 (4)
6 (6)
3 (1)
6 (4)
Idiomatically Used Prepositions and Multi-Word Verbs 171
KN 10 10 9 11 11 7 5 9 8 8 (1)
6
8 7 6 5 3 10 8 (1) 6 8 5 6 (1)
6 (1)
5 (2) 7 (1) 6 5 5 (1)
C13 C16 C5 C14 C25 C33 C27 C45 C3 C6
C21
C55 C58 C57 C18 C10 C4 C54 C39 C11 C7 C12
C32
C28 C1 C15 C17 C9
No.
7 (1) 4 (1) 8 (5) 9 (5) 6 (1)
6 (2)
6 (2) 5 (1) 8 7 (1) 9 (3) 5 6 (2) 7 (2) 4 (2) 7 (1) 6 (1)
8
Info about verb TK 5 5 4 3 3 7 8 5 6 (1) 5
1 4 1 1 3
1
1 2 3 2 1
1 3 (1) 1 2 1 (1)
1
1 1 1
2 (2) 1
NK
Multi-word verb (correct answers) andas ut (15) varva ner (15) passa på (15) turas om (14) se upp till (14) brusa upp (12), blossa upp (2) (14) backa upp (14) frysa ut (13) dyka upp (13) tänka om (13) lägga av med18 (13) something) nicka(r) till (12) lappa(de) till (12) hålla (er) av (12) falla för (9), fastna för (3) (12) trappa ner (12) ge efter för (9), gå med på (2) (11) kasta upp (11) brås på (11) passa in (10) komma undan (10) gå under (10) tysta ned/ner (10) stifle) halka efter (10) gripa in (8), gå in (1) (9) smälta in (9) tända till (9) koppla av (9) (=fall behind) (=step in) (=blend in) (=get angry) (=relax)
(=doze off) (=slap) (=be fond of) (=fall for (someone)) (=take it easier) (=yield to, agree to) (=throw up) (=take after (a person)) (=fit in) (=get away) (=succumb) (=reduce to silence,
(=feel relieved) (=take it easier) (=take the opportunity) (=take turns) (=look up to) (=lose one’s temper) (=back (up), support) (=freeze (a person) out) (=appear) (=reconsider) (=stop doing
5 7 (1) 6 5 5 (1)
7 7 6 4 3 9 (1) 7 7 7 4 5 (1) 6 (1)
KN 10 10 11 10 11 7 5 9 (1) 10 7 (1) 6
6 4 (1) 7 (2) 9 (4) 6 (1)
7 5 8 (2) 8 (1) 8 (2) 6 (2) 6 (2) 6 (1) 5 (2) 8 (2) 7 (1) 6 (2)
First particle TK 4 5 4 4 3 7 8 5 3 (1) 6 8 (1)
2 4 1 1 2
2 2 2 1 1 1
3 (3) 1 2 2 (1)
1 1 1 1 (1) 1
1
NK 1 (1)
9
10
KN
6 (3)
2
Second particle TK NK
Table 5.10 T he students’ results on and evaluation of multi-word verbs on the Swedish test, with the test items listed in order of correct answers, starting with the item that received the highest score. (Information about the verb: KN = for the gap-fill task: the item is known, and also for the correction task: the incorrect prepositional use has been found; TK = for the gap-fill task: the item is thought to be known, and also for the correction task: think the incorrect prepositional use has been found; NK = for the gap-fill task: the item is not known, and also for the correction task: the incorrect prepositional use has not been found.) (Information about the particle: KN = the gap-fill task/the correction task: the particle to fill in/use is known; TK = the gap-fill task/the correction task: think the particle to fill in/use is known; NK = the gap-fill task/the correction task: the particle to fill in/use is not known.) Figures in parentheses in columns KN–NK indicate cases where the students were incorrect in their self-evaluation
172 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
3 (1)
1 1
C51
C35 C50 C41 C56 C46 C31 C42
Correctness rate
Total
C30
340 (111)
67.4%
274 (27)
90.1%
3 (3) 4 (4)
19
C22 C36
6 (3) 2 (1) 5 (3) 6 (4) 6 4 (3) 4 (3)
9 (8)
3 (2)
1 (1) 3 (3)
6 (3)
8 (5) 5 (1) 9 (6) 8 (4) 3 (2)
C59
1 (1) 1 1
1
2 (2) 4 1 3 (3) 5 (2)
C47 C34 C43 C44 C53
2 2
C29
6 (5)
5 (1)
C8
C48 C24
6 (2) 3 (1)
7 (2) 9 (2) 11 (5) 5 (1) 8 (1) 8 (3) 2 5 5 (2) 4 (1)
5 (2) 2 3 (1) 2 1 5 5 3 3 3 (2)
C20 C23 C19 C2 C37 C49 C52 C40 C26 C60
1
5
C38
88.8%
214 (24)
11 7
9 (1)
6 10 7 7 7 (2) 8 6
3
4 (2) 5 (3) 4 4 5
6
7 6
3
3 (1) 3 1 7 (3) 6 (2) 2 (1) 7 5 5 7 (3)
8 (2)
droppa av (5) ryckas med (5) brinna för (5) about) storm(ade) in (4), stävla(de) in (1) (5) talas vid (4) klappa ihop (4) duka under (4) tutta på (3) bulla upp (3) excess) backa ur (3) bolla över (2) blomma upp (2) pinna på (2) falla undan för (1) bryta upp från (1) fresta på (1) lägga an på (0) (romantically)) bädda för (0) bottna i (0) sadla om (0) profession)
dra till med (8) (something)) hosta upp (6), hala fram/upp (2) (8) tala ut (8) friska upp (4), fräscha upp (4) (8) hålla ut (5), hänga med (3), härda ut (0) (8) falla bort (5), falla ifrån (3) (8) fiska efter (7) runda av med (7) bomma igen (7) spåra ur (7) nita till (6) hoppa in (5)
(=prepare for) (=originate in) (=change one’s
(=back out) (=pass on (something)) (=blossom out, grow) (=get a move on) (=give in to) (=leave) (=strain) (=go for
(=trudge in) (=talk) (=collapse) (=die) (=set on fire) (=offer something to
(=cough up) (=clear the air) (=freshen up) (=endure) (=drop off) (=fish for) (=conclude) (=close, lock up) (=go astray) (=strike) (=step in (in a person’s place)) (=drop off) (=be carried away) (=be enthusiastic
(=come up with
271 (28) 89.7%
1 (1) 3 (3)
1 1 (1) 1
1 2
2 (2) 2 (1) 1 2 (2) 5 (4) 3 (1)
2 2 1
5 (2) 2 4 (1) 2 1 5 (1) 5 3 5 (1) 3 5 (1)
5
325 (100) 69.2%
3 (2) 4 (4)
6 (4) 3 (1) 5 (1) 7 (1) 4 (4) 5 (4) 2 (2) 3 (3)
7 (3) 5 (3) 5 (3) 8 (4) 3 9 (7)
6 (2) 3 (1) 6 (3)
6 (2) 9 (3) 8 (4) 5 8 (2) 8 (5) 2 5 (1) 3 (1) 3 4 (3)
1
216 (29) 86.6%
11 7
5 8 (2) 7 (2) 5 (2) 8 7 7 (1) 9
5 (1) 6 8 (1) 4 5 (3) 3
6 6 6
3 (1) 2 3 (1) 7 (3) 5 (1) 2 7 4 5 8 (2) 5
8 (3)
96.7%
30 (1)
1 1 (1)
5
4
63.2%
82.1%
28 (5)
7
1 (1)
19 (7)
6 (1) 4
4
8 (4)
4 (1) 4 (2)
1
1
Idiomatically Used Prepositions and Multi-Word Verbs 173
174 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
of phrasal verbs being tightly knit semantic units is also directly mirrored in the various kinds of input that L2 learners receive from native speakers (e.g. corrections) and from almost all other sources supplying them with different types of spoken and written English. As discussed before, the opposite applies to prepositions that are used idiomatically; that is, quite a lot of input may induce the L2 learner to not think of a stem noun/ verb + bound preposition as a unit. Considering this difference, it is not so strange that collocational frequency only appears to be a usable pedagogical tool for multi-word verbs and not for idiomatic prepositional use when a number of different senses are tested. Individual L1 and L2 knowledge
For six of the 15 participants, there is a correlation between what they were able to achieve in their mother tongue as compared to their L2, with three informants (Students 9, 11 and 14) consistently being among the top 5 students and the other three (Students 4, 6 and 10) consistently being among the bottom five on all test parts in both languages. It can here be added that all six learners also did either (very) well or (very) poorly respectively on the tests focusing on exploring the informants’ vocabulary size discussed in Chapter 2. Additionally, it is interesting to note that two of the high achievers (Students 11 and 14) and one of the low achievers (Student 10) also got consistently high and low scores respectively on the idiom/proverb tests discussed in the previous chapter. As idioms and multi-word verbs both involve units whose components most often cannot readily be added together to get their sums, and idiomatically used prepositions are implemented in idiosyncratic ways (at least superficially so), this is not so surprising. Research with larger student groups must of course be performed, but it may be that, provided with context, being skilled at disambiguating idiom meaning rubs off on students’ performance with phrasal verbs and vice versa. 5.5 Conclusions and Pedagogical Implications
As seen in Section 5.4, the students’ mastery of idiomatic prepositional use in English is not frequency-based, at least not when a variety of different prepositional senses are contrasted with each other. The main reason for this appears to be that students do not perceive these as forming units with their stem nouns and verbs. The most important aim for teachers must therefore be to do everything in their power to create tightly knit mental units so that students, similarly to what happens in their mother tongue, will automatically think of the correct bound preposition when faced with nouns and verbs that require these particles. Supported by textbooks, in the word lists of which stem nouns and verbs always are presented together with their bound prepositions, even if the prepositions are
Idiomatically Used Prepositions and Multi-Word Verbs 175
not present in the context given, teachers should make students aware of the existence of such prepositional uses, perhaps by first giving examples in the students’ mother tongue. This is affirmed by the Schmidt (2004) study, in which there was a dramatic increase in the percentages of correct answers between the pre- and post-test, the latter being preceded by instruction on prepositional uses categorised along a predictability continuum. Furthermore, rote memorisation, although often an ill-reputed method, may here also have certain benefits. Hinkel (2012) even states that ‘learning L2 in lexical and grammatical units (chunks), instead of discrete words or word elements’ may make ‘learning far more efficient’ and lead ‘to higher levels of accuracy and fluency for the learners’. For instance, the present author still remembers all the French verb forms she rehearsed over and over again during her years in upper secondary school. Similarly, having memorised, on an item-by-item basis, the correct definite article together with its respective noun now makes it inherently odd to hear mistakes such as *le bague (where la is the correct definite article) and *la bateau (where le should be used). The verb forms and definite articles seem to be forever engrained in memory. This is of course also the aim in connection with the items discussed here. Research is also being done on the benefits of approaching idiomatically used prepositions from a cognitive perspective (e.g. Lindstromberg, 1997). Once this type of research permeates through to the L2 classroom, its advantages and disadvantages can be assessed more easily. In contrast to idiomatic prepositional use, frequency was seen to have a considerable impact on whether the students were able to produce the tested phrasal verbs or not. Thus, in addition to emphasising the prolific (and more informal) nature of multi-word verbs in general as compared to single verbs, frequency should be regarded as a viable and important pedagogical tool when selecting multi-word verbs for the teaching syllabus. (Compare frequently used items such as fill in, take off and turn out, which the majority of the students in the present study knew, to comparatively common items such as put (someone) through, fall out and put off, which between one third to half of the subjects in the present investigation were able to produce, to fall in with, catch out and live down, which are highly infrequent items that none of the informants in the present study knew.) Hence, a frequency-based selection in combination with a CL approach, as described in the theoretical background (Section 5.2), thus possibly reducing the feeling that the choice of stem verb and especially the choice of particle are completely idiosyncratic in nature, will help teachers and students approach these items in a more systematic way. 5.6 The Chapter That Follows
Chapter 6 will investigate the 15 students’ L1 and L2 mastery of polysemous words, lexical fields of near synonyms, and false friends, all of
176 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
which are comparatively unexplored areas from a second-languagelearner perspective, especially the latter two. Acquiring these kinds of lexical items requires extensive networking building, both forming links within each subfield and with other subfields. Superficial cognates may be particularly troublesome, as, while connections between a learner’s first and second language are most likely continually being made, students’ knowledge of false friends relies completely on being able to make use of L1–L2 ties. Notes (1) (2) (3) (4)
(5) (6) (7)
(8) (9) (10) (11) (12) (13) (14) (15)
No further distinctions between the various types of multi-word verbs will be made in the present investigation. For a more detailed description of the different categories, see Quirk et al. (1985). As the reader will see in Section 5.2, not all researchers define multi-word verbs as having such a wide transparency span. SWICLE refers to the Swedish component of the International Corpus of Learner English. In Karlsson (2002), the use of the preposition of as a default preposition, as exemplified in no sense of humour (Hargevik, 1996: 232) and the advantages of this car (Svartvik & Sager, 1988), was also categorised as a systematic prepositional use. The present author has since rethought this decision. In contrast to the rule-governed and construction-dependent uses described, this use appears to be comparatively idiosyncratic in nature since, as the reader will become aware in Section 5.3, other prepositions are very often also accepted by native speakers here. In the present study, this type of prepositional use will therefore instead be categorised as an idiomatic prepositional use. In the present investigation, idiomatically used prepositions will also be referred to as bound prepositions and noun/verb–particle combinations with no difference in meaning. The term phrasal verb will be used interchangeably with the term multi-word verb without any difference in meaning. It is, however, possible to find one-word synonyms for non-idiomatic items too. In Mary picked up the pen and put it in her pocket, which includes the non-idiomatic multi-word verb pick up, the combination could easily be replaced by grab or take, for instance (Mondor, 2008: 83). Due to a construction error, there were in the end only 74 prepositional uses on the pre-test. Since the corresponding Swedish item is a single verb (kommentera), students here tend to omit the preposition on. Because of a construction error, three test items had to be removed from the English test (Subpart Ab). Due to a mishap, the students were given the solution to multi-word verb number 30 on the Swedish test. Only 59. Multi-word verbs could therefore be included in the investigation. One participant did not offer any details on this test part. The reader is again reminded that, due to a construction error, three items were excluded from Subpart Ab; these were concerned with noun + particle combinations and asked the subjects to correct errors. As explained in connection with the students’ mean scores, test item C30 on the Swedish test, shown in Figure 5.3 as having received zero points, was omitted due to a construction error.
Idiomatically Used Prepositions and Multi-Word Verbs 177
(16) The results for the various solutions to test item Aa6 are offered here for spatial reasons: criticism about (3), criticism of (3), criticism for (2), criticism on (2), criticism against (1), criticism at (1) and, finally, criticism with (0). This test item thus received a total score of 12. (17) Item C21 in.the English test part (set (my dog) on (someone)) has quite a few affinities with free combinations of verb and particle, and therefore separates itself slightly from the rest of the phrasal verbs included in the present investigation. (18) The present author forgot to ask for information about the second particle in this case. (19) As pointed out before, this item was not included due to a construction error.
6 Polysemous Words, Lexical Fields of Near Synonyms and False Friends
6.1 Introduction
Based on research on, for example, slips of the tongue, a learner’s mental lexicon is thought to consist of two collaborating networks (Viberg, 2000: 287), one dealing with words from a semantic perspective (in which the words crossword and tomorrow may be wrongly used instead of jigsaw and yesterday (Aitchison, 1987: 19)) and one dealing with words from a phonological and orthographic point of view (where porcupine and orgasm may be erroneously used instead of concubine and organism (Aitchison, 1987: 19)). As research has shown that different areas of the brain are activated when we hear or speak in contrast to when we read or write (Viberg, 2000: 288–289), the part of the mental lexicon dealing with phonological and orthographic aspects is also believed to form two smaller networks. In the semantic network, the meanings of words and some grammatical information such as word class are stored through a number of d ifferent types of semantic links, such as coordination (e.g. salt ↔ pepper ↔ mustard), collocation (e.g. salt ↔ water), hyponymy (e.g. animal ↔ mammal ↔ dog ↔ Cocker Spaniel), synonymy (e.g. surprise, amaze, astound, etc.), antonymy (e.g. cute ↔ ugly) and meronymy (e.g. hand ↔ finger). Of these various types of semantic links, the first two mentioned – coordination and collocation – appear to be especially strong; that is, most semantically induced slips of the tongue appear to fall into either of these two categories. Blends such as *Noshville (Nashville + Knoxville, Tennessee towns) and *taquua (tequila + kahlua, Mexican drinks) are examples of the former type (Aitchison, 1987: 76), and *a large smoker and *a huge eater (instead of a heavy smoker and a big eater) are examples of the latter type (Martin, 1984: 132). Although errors are also seen with near synonyms, as in I don’t know many brands (instead of species) of tree and The inhabitants (instead of occupants) of the car were unhurt (Aitchison, 1987: 83), these are not nearly as common, indicating that lexical ties between near synonyms are comparatively 178
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 179
weak (Aitchison, 1987: 84). These semantic connections are in turn intertwined into even more elaborate networks. For example, for each item in an antonymy pair there are usually a number of synonyms (Viberg, 2000: 294). Other types of intertwined relationships can also be seen. The various meanings of polysemous words, for instance, are defined in the mental lexicon by being compared to items in different synonym sets. However, for polysemous words the situation is even more complex, as a great deal of research indicates that their meanings, just like stems and their derivatives (see Chapter 3), are also stored together, forming small worlds of their own. In Mackay (1966), for example, it was shown that in a context in which more than one meaning of a polysemous word is plausible, all meanings are activated subconsciously. This is also confirmed by quite a few other investigations (e.g. Foss, 1970; Lackner & Garrett, 1972). Even in cases where one meaning is considerably more likely than others, there appears to exist a brief moment during which all meanings are activated (Swinney, 1979). The fact that the meanings of polysemous words are stored together in the mental lexicon is also the reason it is possible to involve these types of words in wordplay for comical effect. In the second film in The Twilight Saga, for instance, one of the boys who can transform into a wolf warns his friend not to reveal trade secrets to Bella since she runs with vampires, meaning that she spends time with them. Bella jokingly comments on this using the more literal meaning of the same verb, saying that you cannot run with them since they are too fast. As a second example, in an episode of a show on Animal Planet, two cats are desperately trying to get out of their fenced-in space. The cats are called Thelma and Louise, named after the two main characters in the movie with the same name. The movie’s Thelma and Louise become outlaws due to an unforeseen chain of events, and at the end of the movie they are faced with the choice of either surrendering to the police or killing themselves. They choose the latter, driving off a tall cliff. Animal Planet’s Thelma and Louise are finally able to get out by crawling under a chest of drawers and the host of the show says that they were indeed in a tight squeeze, thus playing with the ambiguous meaning of the word squeeze. (For more examples see Alm-Arvius, 2003: 141–143). The various semantic links discussed above connect with other links until entire lexicons are developed based on interconnections. Connections between words allow newly acquired words to be easily assimilated within these networks because new words are not learned in isolation, but through links to already learned words. As learners progress lexically, they build lexical networks that are strengthened by differentiating sense relations between words and within words. (Crossley et al., 2010: 576)
(See also Ferrer i Cancho & Solé, 2001; Ferrer i Cancho et al., 2004; Haastrup & Henriksen, 2000; Mecartty, 2000; Quian, 1999, 2002 and Verspoor & Lowie, 2003, for similar discussions.)
180 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
In the form network, words are organised based on the one hand on their initial syllable, so that words such as stab, stack, stagger, stagnate, stalactite, stamina, stampede, stand, stance and static are found together (Aitchison, 1987: 186), and on the other hand, to some extent, on the last syllable, the existence of erroneous uses of words like anecdote instead of antidote being evidence of this (Aitchison, 1987: 120). This phenomenon in which the middle part of words are obscure, referred to by Aitchison (1987: 119) as the bathtub effect – due to a person’s head and feet being out of the water in a bathtub but the main part of their body being underwater and thus blurred – was first detected by Brown and McNeill (1966) when they performed a tip-of-the-tongue experiment asking subjects to give as much information as possible about words they could not remember. It turned out that it was the first and, to some degree, the last part that the informants could most easily say something about. People learning a second language are faced anew with the challenging task of creating a form network that tallies with the semantic network. This is a process referred to as mapping, in which L2 learners transfer semantic knowledge of L1 words to the corresponding L2 words (Jiang, 2004), while at the same time making use of L1–L2 phonological and orthographic similarities. In many cases this results in positive transfer. For example, assigning the semantic features that make up the Swedish word hund to the English word dog is quite straightforward since these two combinations of letters agree completely on the semantic profile. Even in cases where agreement is not as clear-cut, learners rely on approximations of semantic meanings, approximations that very often appear to work. In other cases, the L1 interferes rather than helps. A Swedish learner who says *He is the proprietor of a Saint Bernard, for instance, has made an incorrect choice between the near synonyms proprietor and owner (Martin, 1984: 133), probably thinking that they are both equivalents to the Swedish word ägare. A Spanish-speaking learner of Swedish who says vinna mycket pengar (win a lot of money) when wanting to say tjäna mycket pengar (make a lot of money) has wrongly assumed that the Spanish polysemous verb ganar, which can mean both vinna and tjäna, can be used in the same way in Swedish; that is, the learner has unknowingly made use of a semantic profile that does not work in the L2. Also, a Spanish learner who believes that two cars are thought to chockera (shock each other) when they collide has wrongly assumed that the Spanish verb chocar (which means collide) has its direct counterpart in the phonologically and orthographically similar Swedish verb chockera (which means to shock someone); that is, the Swedish verb chockera and the Spanish verb chocar are false friends (Viberg, 2000: 297). In these three examples, the incorrectly used items all share some properties with the desired ones, but since a oneto-one relation does not exist, the learner has in all three cases made incorrect hypotheses about the links (semantic ones in the two former cases and phonological/orthographic in the latter case) (see also Wolter, 2006).
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 181
Since phonetic associations are developed faster than associations connected with meaning, errors that are induced phonologically seem to disappear more quickly than those that are semantically based (Meara, 1982). The reason for this is that the vocabulary system of an L2 beginner and a low-proficiency L2 learner, just like the vocabulary system of an L1 beginner, is primarily based on phonology (Meara, 1978). The evidence comes mainly from association tests where L1 and L2 beginners/low- proficiency L2 learners are likely to think of a word like hat when prompted with a word like cat, while L1 adult speakers and high-proficiency L2 learners are more likely to offer a semantically related word like dog (Cohen & Aphek, 1981; Meara, 1978; Piper & Leicester, 1980; Singleton, 1999; Söderman, 1993; Wolter, 2001). Based on the above, it is clear that mastery of polysemous words, lexical fields of near synonyms, and false friends involves extensive network building. Exploring the 15 university students’ knowledge in these areas, starting with polysemous words, will therefore help shed more light on their vocabulary depth. 6.2 Polysemous Words Introduction
There are two different types of multi-meaning connections between word sense and word form: homonymy and polysemy (Alm-Arvius, 1998). While homonyms, such as bank (of a river) versus bank (financial institution) and bat (flying creature) versus bat (used in various sports) (Yule, 1996), have separate dictionary entries, the meanings of polysemous words are found under one and the same entry. Polysemy is also the more frequent type of the two and is therefore more of a rule than an exception (Gyori, 2002; Murphy, 2004). The present investigation will only focus on the latter of these two semantic relations. A polysemous word is a word that has ‘multiple meanings which are related by extension’ (Yule, 1996: 121). More specifically, it is a word that ‘commonly has a seemingly more basic or primary literal sense and one or more related, transferred or figurative readings’ (Alm-Arvius, 1998: 57). This can be exemplified by the verb run in The children came running towards us, where run has a literal sense, versus She did not run in the last election and Do you know how to run this machine? in which the use of run is more figurative (Alm-Arvius, 1998: 57). The development of figurative meanings of polysemous words, referred to as ‘layering’ (Aitchison, 2003: 154), is a language phenomenon that mainly occurs because people prefer to take words and extend their meaning rather than create new ones (the law of least effort: Gyori, 2002; Murphy, 2004). Historically, new meanings of words have appeared because speakers could see conceptual links between the original sense
182 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
and a newer sense. The verb launch, for instance, originally had the meaning wield a lance, but it generalised over time and acquired the meaning throw (any object) forward with force. Nowadays the verb launch is probably more associated with rockets and ships than with lances; that is, this meaning would now be considered the core meaning of the word launch (Verspoor & Lowie, 2003: 555). The core meaning is the one that represents the most literal sense that the word has in modern usage. This is not necessarily the same as the oldest meaning, because word meanings change over time. Nor is it necessarily the most frequent meaning, because figurative senses are sometimes the most frequent. It is the meaning accepted by native speakers as the one that is most established as literal and central. (The New Oxford Dictionary of English, 1998: foreword)
New meanings may be formed through semantic extensions, for instance. In The houses had been gutted by grenades, the verb gut is based on its more literal use as seen in They gutted the deer (Verspoor & Lowie, 2003: 556). In many cases semantic extensions appear to begin with the human body and then move outward to other parts of the world, for example, the foot of the mountain, the ribs of the ship and the head of the organisation. In other cases, the opposite direction can be seen, using everyday bodily behaviour to describe internal events, for example, I see what Helen means, Peter held on to his point of view and Let us go over that plan again (Aitchison, 2003: 154). New meanings may also develop through specialisation and generalisation. An example of the former is the noun queen, which was originally used to mean woman but now refers to a particular kind of woman. An example of the latter concerns the noun grid, the literal sense of which is a perforated or ridged metal plate, but which may also be used to cover all networks of uniformly spaced and perpendicular lines (Verspoor & Lowie, 2003: 556). This means that links are not limited to the ones that occur between a core sense and a noncore sense, but the senses are all interrelated, as one peripheral sense may form the base for an even more peripheral sense. However, there may not be any direct conceptual links between all peripheral senses. (Verspoor & Lowie, 2003: 556)
By extension, this also means that there may not necessarily be a clear link between a polysemous word’s core sense and its most frequent meaning. (See also Lakoff, 1987, for the same discussion.) On the whole, layering is a process that primarily appears to take place with short and comparatively frequent words (Zipf, 1945), making polysemous words a very common vocabulary phenomenon. Thus, because frequent words have the most senses, learners will encounter polysemous words more often than other types of words. It is therefore unfortunate that L2 knowledge of polysemous words has not been a common topic in research (Crossley et al., 2010). The research that has
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 183
been done shows that learning the meanings of polysemous words is an exceedingly slow process (see Schmitt, 1998, which is discussed in more detail in the theoretical background). In Bensoussan and Laufer (1984), for example, it was found that L2 learners did considerably worse on guessing the meaning of contextualised polysemous words than on guessing the meanings of other contextualised items of vocabulary. Research has also shown that students will typically acquire the most frequent meaning senses first, and that, from a teaching perspective, it therefore makes sense to consider commonality, with mastery of the more infrequent senses indicating a more comprehensive knowledge of a word (Schmitt, 2010b). Theoretical background and previous research L2 learners’ mastery of polysemous words
In one part of Schmitt (1998), learners’ incremental acquisition of meanings of polysemous words was investigated. The study included three university students of English (from India, Lithuania and Taiwan) and lasted for a year, during which the learners’ acquisition of 11 words with multiple meanings was looked into. In order to make sure that the words were encountered during the subjects’ university studies, the words were all picked from the University Word List (Nation, 1990; Xue & Nation, 1984). The words were also required to have at least three different meanings. Moreover, in order to ensure insights would be gained into the beginning stages of acquisition, as the informants were advanced learners of English, some comparatively infrequent words were included. Both the informants’ receptive and productive knowledge was investigated. Information was gained in individual sessions with the subjects in which the researcher started by asking the participants to explain all of the senses that they knew for each word. The senses that the learners offered spontaneously were thought to represent their receptive knowledge. When the learners could not come up with any more meanings, the researcher gave prompt words in order to elicit additional senses. Meanings that were accessed in this way were thought to demonstrate the learners’ productive knowledge. The pre-test showed that only one of the three subjects knew all the senses of one of the 11 words tested; that is, in all the other cases the students had only partial knowledge. In fact, on average only 50% of the meanings were known, and one of the informants did not know any of the senses for five of the words. Schmitt expresses his surprise at the poor achievements of the comparatively advanced informants. Six months apart, two additional tests were administered in the same way as described above. The results from these two tests show that the students’ meaning knowledge remained the same for the vast majority of the 11 test items. This indicates clearly that acquiring knowledge of the
184 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
various meanings of polysemous words is a slow process. On the positive side, the students did not appear to forget meanings once they had been learned. While there was improvement in 20% of the cases, the informants only forgot about 8% of the senses they knew at the first test opportunity. This was especially true for the senses that were known productively. Moreover, of the 74 instances of progression, there was an equal number of cases in which the informants moved from receptive knowledge to productive knowledge and cases in which the subjects moved from not knowing any meaning to gaining receptive knowledge. Unsurprisingly, there were considerably fewer cases where the knowledge moved from not knowing any meaning to having productive knowledge. In Crossley et al. (2010), both L2 mastery (quantitative approach) and growth (qualitative approach) in the spontaneous production of polysemous words in spoken language were investigated. Six L2 learners between the ages of 18 and 29 were included (L1s: Arabic, Japanese, Korean and Spanish), all of whom were enrolled in an intensive English university programme, and starting at the lowest proficiency level among the six that were available. In order to get more specific information about their development with polysemous words, the informants were, during a one-year period, interviewed every second week. In their first analysis, using a quantitative approach, Crossley et al. investigated whether there was an increase in the informants’ use of frequent words. If this were the case, since frequently used vocabulary items most often have more senses than infrequently used ones (Zipf, 1945), it would, according to the researchers, demonstrate a movement toward the use of more words with multiple meanings. This would also indicate an increase in the development of lexical relationships – that is, lexical networks – in general. While the results showed that there was a clear increase in the number of words with multiple meanings during the first four months, the data also showed that, after having reached a plateau, the students’ performance began to level out. This was also noticed in the Schmitt study discussed above (1998). In their second analysis, having a qualitative approach, Crossley et al. (2010) focused on the subjects’ development of the use of six polysemous words that were commonly produced by the informants. The analysis was limited to 60 examples of each derivative form for each word. The results showed that the majority of the informants demonstrated word sense growth for most of the words. This means that ‘although learners’ production of polysemous words tapers off after an initial period of growth’, as seen in connection with the quantitative analysis discussed above, ‘the actual sense relations that L2 learners use in their discourse increase in type (more varied senses of a lexical item)’ (2010: 599). Crossley et al. argue that such an increase in word sense relations is the result of learners beginning to make associations between the multiple senses available within individual lexical items. Thus, L2 learners at an early stage are able to recognise and employ semantic extensions of core word meanings. These
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 185
extensions seem to appear rapidly after the fourth month of learning. After this point, word extensions seem to increase and learners seem to have both a better understanding of word sense relations and the ability to produce more senses for each individual lexical item. (2010: 596)
Put in relation to what was observed in connection with the first analysis, according to Crossley et al. (2010: 597), this indicates that L2 learners first produce words that have the capacity for multiple word senses and thus ambiguity; however, only after the learners have acquired the core senses of the word do they begin to acquire and produce the other word senses available to that lexical entry. Swedish learners’ mastery of L2 polysemous words
In Odefalk (2004), a term paper supervised by the present author, the aim was to investigate comparatively advanced Swedish students’ mastery of polysemous words. The 57 subjects included came from three different educational levels, thus enabling the researcher to explore differences in mastery between students of various knowledge levels. Thirty-seven of the students participated in the adult educational system, 20 of whom took the A-course in English, which corresponds to the first-year course of English at upper secondary school, and 17 of whom took the more advanced B-level course, which corresponds to second-year English in upper secondary school. Since these students took part in the adult educational system, they were of varying ages. The remaining 20 students were first-term university students. Although the majority of these were around 20 years of age, there were a few older students included here too. The test given to the subjects, focusing on 48 polysemous words, was divided into three main parts, each testing increasingly more infrequent items, picked randomly from Hargevik’s (1998) frequency word list of spoken and written modern English. Sixteen of the words were from lower secondary school level, another 16 were from upper secondary school level, and the remaining 16 were judged to be items typically introduced at university level. Additionally, for each of the 48 test items, the informants were offered six decontextualised meanings in Swedish. At least two of these were correct. Also, in order to deter the students from indicating all of the meanings given, the subjects were informed that at least one of the meanings was incorrect and that points would be deducted for erroneous answers. The incorrect options were created in one of four ways. The first category were displayed orthographic and/or phonological similarities with the word tested. Stabilitet (stability) and bagare (baker), for instance, were offered as incorrect meanings of the English words ability and bag. The second category were translations of French words that share phonological and/or orthographic similarities with the test word. As incorrect meanings for the polysemous words bag and suit, for example, the
186 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
students were offered ring (from the French word bague) and överleva (which is a translation of the French verb suivre, the past participle of which looks like the English word suit). The third category were based on Swedish translations of English words that have similar orthographic and/ or phonological characteristics. Tillagad (cooked) and tonåring (teen), used as incorrect meanings for crooked and tin, are examples of this kind. Lastly, the fourth category were simply picked randomly, thus showing no real links to the meanings of the polysemous words tested. Table 6.1 presents the subjects’ average scores on the three test parts. As can be seen, for all three informant groups the results are frequencydependent so that the average score is highest for the most common items (Part I) and lowest for the least common items (Part III), the difference between Part I and Part III being greatest for the A-level students and smallest for the university students. In Table 6.2, the results are put in relation to the number of correct meanings mastered by the students. At all three educational levels and within all three frequency bands, the majority of the students knew one meaning of the polysemous words. The students’ mastery then decreases dramatically when two or more meanings are considered. This too applies to all educational levels as well as the three frequency bands tested, the decrease generally being least pronounced with the university students and most visible with the A-level students. In fact, the students from upper secondary school level only knew two meanings for about half of the words tested as compared to three quarters for the university students. Odefalk also points out that Table 6.1 Average scores for the three student groups tested on the three different frequency-based test parts (Odefalk, 2004: 10, 14 and 17) Average score A-level (20 students)
(Total: 55)
Part I
24
Part II
17
Part III
14
B-level (17 students)
(Total: 68)
Part I
27
Part II
24
Part III
18
University-level (20 students)
(Total: 84)
Part I
31
Part II
27
Part III
25
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 187
Table 6.2 Percentages of correct answers for each correct meaning for the three student groups tested (Odefalk, 2004: 11, 14 and 17) Correct meanings given
Part I
Part II
Part III
1
90%
80%
64%
2
56%
40%
32%
3
23%
14%
13%
4
7%
7%
5%
5
–
–
–
1
92%
90%
73%
2
70%
52%
43%
3
19%
20%
12%
4
3%
7%
6%
5
–
–
–
1
99%
92%
84%
2
85%
72%
60%
3
40%
21%
29%
4
13%
11%
11%
5
–
5%
1%
A-level (20 students)
B-level (17 students)
University-level (20 students)
erroneous meanings that were indicated as correct senses were very often synforms, as exemplified by the Swedish word lastbil (truck), which quite a few of the students believed was one of the meanings of the polysemous word trunk, and smidighet (agility), which some students thought was one of the meanings of the word ability (2004: 11, 15). The informants in Odefalk’s investigation were not only asked to indicate which of the meanings they thought were correct but were also told to rank their choices according to how certain they were that these actually were correct senses of the word at hand. The results of this part are shown in Table 6.3. In all three groups within all three frequency bands, the students were most correct about the item they had ranked as number one, then showing a clear decrease in certainty for items number two and three. (However, when the first, second and third meanings respectively are compared across frequency bands within the same student group, there appears, in the majority of cases, to be an inexplicable increase in certainty for the most infrequent items (Part III).)
188 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 6.3 Percentages of correct answers for the frequency order of the polysemous words tested (Odefalk, 2004: 11, 15 and 18) Part I
Part II
Part III
1st
54%
42%
46%
1st, 2nd
17%
A-level (20 students)
28%
12%
1st, 2nd, 3rd
6%
3%
1st, 2nd, 3rd, 4th
1%
4%
–
–
–
–
–
1st
55%
40%
43%
1st, 2nd
30%
18%
16%
3%
3%
1st, 2nd, 3rd, 4th, 5th B-level (17 students)
1st, 2nd, 3rd
1%
1st, 2nd, 3rd, 4th
–
–
–
1st, 2nd, 3rd, 4th, 5th
–
–
–
1st
61%
41%
54%
1st, 2nd
40%
20%
26%
8%
3%
8%
University-level (20 students)
1st, 2nd, 3rd 1st, 2nd, 3rd, 4th
–
–
–
1st, 2nd, 3rd, 4th, 5th
–
–
–
Finally, the results presented in Table 6.3 were also put in relation to frequency. The findings indicate that the senses the students felt most confident about were the most frequent ones in 50% of cases, demonstrating a clear decrease for the second and third senses, down to the fourth and fifth for which there was no agreement at all. In Söderström et al. (2022), the use of puns to teach the meanings of polysemous words was investigated. Fifty-eight Swedish upper secondary school students belonging to three different educational levels (with ages ranging from 15 to 19) were faced with 10 puns, such as Why are teddy bears never hungry? They are always stuffed! In each case, the learners were asked to find the word whose meaning was polysemous (one point) and to offer the meanings relevant to the pun (one point each). The students received a total score of 68% correct answers, showing an increase in knowledge in relation to educational level (first level (mean): 18.3; second level: 25.5; and third level: 28.8). The findings further show that when the students were only able to supply one of the two senses, it was predominantly the core sense that was accurate. As humour has generally
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 189
been shown to strengthen learners’ acquisition (Al-Duleimi & Aziz, 2016), implementing puns in the L2 classroom when dealing with polysemous words thus appears, based on the above, to be one useful approach. Another one will be explored in the section that follows. A pedagogical approach to L2 polysemous words
The finding that students first learn core senses of polysemous words and do not move on to extended senses until these senses have been processed, as demonstrated in the previous subsection, is substantiated by Verspoor and Lowie (2003). Their investigation is based on the notion that when trying to learn the various meanings of polysemous words it might be useful to define the polysemous word with the help of the concept that runs through all the word’s senses (Nation, 2001: 51). Verspoor and Lowie’s research has its foundation in semantic networks theory (Anderson, 1983, 1990) on the one hand and in the observation that indepth elaborations are vital when guessing and trying to remember items of vocabulary (e.g. Stein et al., 1982) on the other hand. Semantics network theory states that ‘different senses of a polysemous word are by nature related to each other, with a core sense having given rise to the more figurative senses’ and that ‘the semantic link between a core sense (e.g. nugget as in gold nugget) and a figurative sense (e.g. nugget as in chicken nugget or as in a nugget of information) is usually one that can be easily (re)discovered and understood, but the link between two figurative senses may not be so clear’ (Verspoor & Lowie, 2003: 568). As for the latter, Hulstijn (2000: 270) writes that ‘processing new lexical information more elaborately will lead to higher retention than by processing new lexical information less elaborately’. Verspoor and Lowie examined whether supplying informants with the core sense of a polysemous word as compared to non-core word senses would help L2 learners figure out the more peripheral meanings of that word. In addition, Verspoor and Lowie investigated whether information about core word meanings improved long-term retention of the more peripheral senses. As described in the earlier discussion of the process of layering, it should be noted here that core meaning does not necessarily refer to the most frequent meaning. Verspoor and Lowie (2003) build on three pilot studies, the results of which gave the researchers incentive enough to carry on with the intended research. In the first of these three pilot studies (Verspoor, 1997), Dutch learners of English were offered the core meanings (in context) of previously unknown polysemous words and were asked to guess the meaning of a more figurative sense of the same word (again presented in context). Even though the results were not statistically confirmed, a positive effect was seen on retention. The second pilot study (Rijpma, 1999) investigated differences in correctness in guessing and retention when either a core sense, a figurative sense, or no sense was provided. The results showed
190 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
clearly that the subjects’ degree of accuracy in guessing the meaning as well as their retention was at its peak when they had been offered a contextualised core sense. In the third pilot study, Lowie and Verspoor (2001) were also able to show that offering a non-core-based sense made the informants less likely to remember the item in question than providing them with a core-based sense. The Verspoor and Lowie study from 2003 included 78 Dutch students, all of whom had studied at least three years of English and were taking part in a pre-university course. The informants were divided into two groups, each being tested three times. The polysemous words selected adhered to two main criteria. Firstly, each had to have at least three different senses: a core sense (referred to as S1), a figurative sense (S2), and a third even more figurative sense (S3). Secondly, only items in which the meaning extensions were clearly chained to each other, so that S1 had given risen to S2, S2 had given rise to S3, and so on, were included, leaving the researchers with a total of 18 polysemous words, the core senses of which all dealt with everyday concepts. On the first test, also referred to as the guessing test, both informant groups were faced with 18 sentence pairs. In each pair, one sentence was used as a cue while the other one included the test item, a figurative sense (S2), for which the subjects were asked to guess the meaning and provide a correct Dutch translation. Whereas Group 1 was prompted by sentences that contained core senses (S1) and a literal translation into Dutch, Group 2 was prompted by another figurative sense (S3) and its Dutch counterpart. Immediately after the test, the subjects were asked to memorise the correct translation of the figurative sense they had just been tested on. They were also asked if they could see any links between the different senses. The results of this test showed that providing students with a core sense yields higher correctness scores than prompting them with a noncore sense. It thus appears that knowing the core meaning of a polysemous word will help learners form meaningful links between the word’s core sense and peripheral sense(s), thus enabling them to comprehend the figurative sense(s) more easily. The second test, an unannounced short-term retention test, took place immediately after the students had had a class unrelated to the subject at hand. This test also consisted of 18 sentence pairs, all of which included the same target words with a focus on the S2 sense, but now appearing in a different order and in a different context. Again the informants were asked to give the Dutch equivalent of the S2 sense tested. The results showed that short-term retention was high irrespective of whether the subjects had been provided with a core meaning (S1) or a more peripheral meaning (S3), the mean score being only marginally better for the former situation. Lastly, two to three weeks later, both student groups were given the third test, which was identical to the second test described above. Here the
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 191
results showed a significant difference in mean score depending on whether the informants had been provided with a core sense or a peripheral sense, the former being by far the more beneficial of the two. An obvious additional problem when faced with polysemous words in an L2 is that the links between peripheral senses might be unclear because the core sense in the L2 has not developed figurative senses similar to those in the L1; that is, in many cases there is simply not a one-to-one relation between the existing peripheral senses. In the Verspoor and Lowie study, there was a clear difference in result for those senses that did not have any direct counterparts in the students’ L1 and those that did. This means that if the meaning extensions in the L2 are essentially the same as in the student’s L1, it is not so important for retention which of the two peripheral senses is introduced first. Verspoor and Lowie (2003: 570) sum up by saying that L2 learners can benefit from a brief introduction into the way that the different senses of a polysemous word may be related to each other and to a core sense, so that they can discover meaningful links among the various senses. This knowledge can be practised in classroom and textbook exercises in which students are to guess a nonliteral sense of a suitable polysemous word from a context, but with a core sense given as an additional cue. Eventually, students should realise that finding the core sense and its meaning relationship with the other senses is a useful strategy in learning vocabulary and then should apply this strategy when they look words up in dictionaries on their own. An improved insight into the polysemous nature of words should make learners aware of the ‘dangers’ of attaching only one meaning to a particular word form.
And the researchers continue by saying that their findings would suggest that this approach not only should help students to understand the ‘imagery’ of the more peripheral senses better but would also help them to remember them better. (2003: 570) Part A of the parallel tests used for the present chapter: Polysemous words
In both the English and Swedish tests, 40 words with multiple meanings were tested and presented to the students in order of frequency, starting with the most common item. Whereas the English test items were picked on a random basis from Norstedts engelska ordbok: Engelsksvensk/svensk-engelsk (CD-rom version), the Swedish words were chosen, also randomly, from Stora synonymordboken (Strömberg, 1979, 1998). Provided that all the correct meanings had been offered and none of the incorrect meanings had been indicated, each of the words was awarded one point. The reason for this rather harsh scoring system was to ensure that the students realised that it would not help to play it safe by indicating all of the senses given. Furthermore, as in Odefalk (2004), the students
192 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 6.4 The number of senses and the distribution of word classes among the polysemous words tested in the present study 2 senses
3 senses
4 senses
5 senses
Total
English test
14 (9 nouns; 3 verbs; 2 adjectives)
11 (5 nouns; 4 verbs; 2 adjectives)
13 (10 nouns; 3 adjectives)
2 (2 verbs)
40 (24 nouns; 9 verbs; 7 adjectives)
Swedish test
12 (6 nouns; 3 verbs; 3 adjectives)
13 (3 nouns; 6 verbs; 4 adjectives)
12 (7 nouns; 2 verbs; 3 adjectives)
3 (1 noun; 2 verbs)
40 (17 nouns; 13 verbs; 10 adjectives)
were faced with six decontextualised alternative meanings for each polysemous word, between two and five of which were correct. Table 6.4 gives an overview of the number of correct meanings for the words tested in each language, as well as the distribution of nouns, verbs and adjectives. As the reader can see, the two tests do not tally completely, with priority being given to the polysemous words themselves, the meanings that were to be tested, and what incorrect senses could be created. For the English test items, the correct alternatives were again based on Norstedts engelska ordbok: Engelsk-svensk/svensk-engelsk (CD-rom version), which lists meanings of polysemous words in order of frequency. The meaning referred to as Meaning 1 in the results section will thus in each case be more frequent than Meaning 2, Meaning 2 more frequent than Meaning 3, and so on. For the Swedish test items, Svenska Akademiens ordlista (online) and Norstedts engelska ordbok: Engelsk-svensk/svenskengelsk were used. Since a clear categorisation according to frequency could not always be attained for the Swedish items, the ordering of these senses is slightly more unreliable. As for the creation of incorrect senses on the English test, Odefalk’s approach was adhered to. Hence, some of the invented incorrect meanings showed orthographic and/or phonological similarities with the word tested, while others were Swedish translations of either English or French words that were also orthographically and/or phonologically reminiscent of the English word. Again, as in Odefalk (2004), one group of incorrect meanings were randomly picked senses that had no real connection to the word tested. The incorrect options on the Swedish test were created in five different ways. As in the English material, the first category of incorrect options were meanings of words that were phonologically and/or orthographically reminiscent of the polysemous word. The second category of incorrect alternatives were based on the fact that the tested word had a morphological make-up that could easily be perceived as having a very different meaning. For the Swedish word remiss meaning referral, for example, the erroneous alternative andra missen (the second miss) was given. The third category were based on translations of English words
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 193
that are phonologically and/or orthographically similar to the Swedish test word. For the polysemous word koloni meaning colony, for instance, the erroneous alternative tjocktarm (i.e. colon (body part)) was offered. A fourth category of incorrect alternatives consisted of senses of the antonym of the polysemous word in question. Lastly, as in the English material, the final category consisted of incorrect options that had no obvious link to the test word. Examples of items on the L2 test are offered in (1) and (2). (The definitions of the correct Swedish options, all taken from Longman Dictionary of Contemporary English (1987), can be found in Appendix 3. These definitions were also used on the test given to the native speaker. The definitions used on the L1 test can be found in the same appendix.) While only two of the meanings provided were correct for vocabulary (meaning both ordlista (=a list of words, usually in an alphabetical order and with explanations of their meanings) and ordförråd (=words known, learned, used, etc.), four of the offered senses were correct for board (meaning bräda (=a long thin flat piece of wood), styrelse (=an official group that has responsibility for particular organisation or activity), anslagstavla (=a flat piece of hard material used to put or write things on), and kost (=(the cost of) meals). (1) vocabulary
a) vokal b) sång (på skiva) c) ordlista
d) fallenhet e) kupong f) ordförråd
(2) board
a) bord b) bräda c) styrelse
d) jord e) anslagstavla f) kost
On both tests, the students were also requested to give information as to what degree they believed they knew the polysemous word and to what extent they were certain of the meanings they had indicated. This is illustrated in (3). (3) weight □ I don’t know this English word. □ I think I know this English word. □ I am sure I know this English word.
a) väntan b) konst c) vikt
d) brevpress e) börda f) syn
Of the Swedish words I have indicated above I am not sure about the following one/s:
194 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
I think I am sure about the following one/s: I am sure about the following one/s:
As in previous chapters, the students were also asked to evaluate the difficulty level of the test part, the options being ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. Lastly, in order to be able to make fair comparisons between the L1 and L2 test part, the total frequencies for each set were calculated, the English test items being more common (constituting ≈0.1100% of the BNC) than the Swedish ones (making up ≈0.0418% of Språkbanken). Hence, from a frequency perspective, the L2 part may be regarded as somewhat easier than the L1 one. Furthermore, as nouns usually form more links with other words than other word classes, they may be considered a somewhat easier type to deal with in terms of polysemy (Nation, 2001: 261). The English test part, as it involves more nouns than the L1 test (see Table 6.4), may thus be regarded as less difficult than the Swedish test in this respect too. The L1–L2 difference in frequency mentioned above was, however, not confirmed statistically, which means that comparisons may be made between the parallel test parts. Still, the difference will be taken into consideration when discussing the results. Results and discussion
The students’ results are shown in Table 6.5. As in Schmitt (1998), the students in the present investigation did very poorly on the English test; only a mean score of 13.20 out of 40 possible points (33.00%) was achieved. This can be compared to the results seen in Odefalk (2004), in which the A-level students produced an average score of Table 6.5 The students’ results on Part A: polysemous words Students
N
Correctness rate
Mean
SD
Standardised scores
Native speaker
1
52.50% (=21/40)1
–
–
–
15
33.00% (=198/600)
13.20
4.13
Highest: 2.13 Lowest: −1.26
15
38.33% (=230/600)
15.33
3.85
Highest: 1.99 Lowest: −1.91
Swedish students – English test Swedish students – Swedish test
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 195
18.3 out of 48 (38.2%), the B-level students an average score of 22.7 out of 48 (47.2%), and the university students, comparable to the present subjects, an average score of 28.0 out of 48 (=58.3%). It must, however, be pointed out that the scoring system used in Odefalk was considerably more lenient than the one implemented in the present investigation, where the students had to indicate all correct meanings and none of the incorrect ones in order to receive one point. Nevertheless, as seen in most research on polysemous words, the results here indicate that acquiring senses of words with multiple meanings is a laborious task, even for advanced students. Also, as evidenced by the results of the native speaker (21 out of 40 (52.5%)) on the one hand and the informants’ mean score on the Swedish test part (15.34 out of 40 (38.2%)) on the other hand, it is safe to say that even in an L1 acquiring meanings of polysemous words takes a great deal of time and effort, indicating the difficult nature of network building in general. (The L1–L2 difference in result could be confirmed statistically at a significance level of 10%, the validity of which is enhanced even further by the fact that the English test items were slightly more frequent than the Swedish ones.) It is interesting to note that, in contrast to all other tests in previous chapters, the students’ evaluation of the difficulty level of the polysemous word test does not tally well here with what they actually achieved, both the native speaker and the Swedish informants (on the English test as well as the test in their native language) being overly confident in their knowledge, not realising the number of meanings a polysemous word may actually have. In fact, the majority of the Swedish students indicated that they thought the test part either ‘average’ (nine students) or even ‘easy’ (three students), with only three thinking it was ‘difficult’ (two students) or ‘very difficult’ (one student). Unsurprisingly, this over-confidence is even more pronounced in the informants’ mother tongue, where 13 students believed the test part to be either ‘easy’ (four students) or ‘average’ (nine students) and only two thought it was ‘difficult’ (one student) or ‘very difficult’ (also one student). As stated above, the native speaker also misjudged his knowledge, indicating that the test was ‘easy’ even though he only knew all the senses sought in 21 out of 40 cases. The results also show that the frequencies of the polysemous words played little or no role in whether the senses were known or not. This holds true for the students’ L2 as well as their L1, and it is substantiated further by the fact that the words tested have a relatively wide frequency span, with the two most frequent items occurring 13,873 (L2 test) and 9923 times (L1 test) respectively, and the two most infrequent items only occurring five (L2) and four times (L1). (For the English test items, this finding is also corroborated by the fact that it does not seem to matter whether the BNC or West’s (1953) general service list is considered. In fact, quite a few of the words that received no points (board, sheet, square, grace, needle, rude, trunk and advance) are found on the latter.) This result does not tally with what was seen in Odefalk’s (2004) investigation, where frequency had a clear impact on the students’ mastery. One reason
196 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
for this difference may be that the words tested in Odefalk might have had an even greater frequency span, and that the boundaries between the three frequency bands tested were clearly delimited. In the present study, the degree of success with which the students were able to indicate the correct meanings rather seems to have to do with the number of meanings sought. This will be discussed in the paragraphs that follow. In Tables 6.6 (the English test) and 6.7 (the Swedish test), the results of the test items for which the students gave all the correct senses are Table 6.6 The students’ results on and evaluation of the English test on polysemous words with a focus on the items for which the students were able to give all the correct senses, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (S = sure that the word is known, TK = the word is thought to be known, NK = the word is not known) No. A36 A18 A35 A20 A10 A26 A12 A22 A4 A29 A23 A14 A33 A21 A3 A13 A34 A40 A7 A6 A32 A19 A38 A11 A15 A9 A5 A2 A17 A24 A1 A8 A16 A25 A27 A28 A30 A31 A37 A39
Polysemous word (info) S TK NK 6 7 1 (1) 8 6 12 1 1 (1) 8 4 8 4 9 3 8 3 8 1 1 (1) 9 7 1 1 (1) 7 2 5 3 5 2 4 3 5 1 3 2 1 4 2 3 2 2 4 1 2 (2) 3 1 2 (2) 1 1 1 1 2 1 1 1 2
WORD (no of correct answers) crooked (15) comfort (14) (n) nail (14) (n) expand (12) intelligence (12) vocabulary (12) journal (11) withdraw (10) weight (9) torch (9) (n) support (9) (v) exclusive (8) expose (7) pursuit (7) ability (6) resignation (5) suppress (5) brood (5) (v) credit (4) (n) administration (4) idle (4) (adj) dealer (3) haul (3) (v) peak (2) (n) wire (2) (n) keen (2) domestic (1) director (1) suit (1) ugly (1) board (0) sheet (0) square (0) (n) grace (0) (n) needle (0) (n) rude (0) trunk (0) brow (0) advance (0) (v) discharge (0) (v)
1 12 13 13 11 10 6 8 5 9 8 7 6 7 5 6 5 3 1 4 4 2 2 1 2 2 1 1 1 1 1
Sure 2 3 9 13 13 8 8 9 9 7 7 7 4 6 8 6 6 6 2 2 4 4 2 1 4 4 3 3 2 1 1 1 3 1 2 2 1 1 1 1 1
1 1 1 1 1
4
5
Meanings Think sure 1 2 3 4 5 2 3
1 4 2 4 2 2 1
3 3 3
2 1
1 2
2 2 2 1 1 2 2 1 1 1 4 2 1 1 1
1
2 1
1
1 1
Not sure 2 3 4 2 1 1 1
3
1
1
1 1 1
1 1
1 5
4
1
1
1 2
3
3
2
3
1 1 2
2
2
1
1 1
1 1
1 1
5
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 197
Table 6.7 The students’ results on and evaluation of the Swedish test on polysemous words with a focus on the items for which the students were able to give all the correct senses, with the test items listed in order of the number of correct answers, starting with the item that received the highest score (S = sure that the word is known, TK = the word is thought to be known, NK = the word is not known) No. A19 A26 A24 A12 A13 A29 A25 A11 A27 A31 A8 A9 A4 A21 A22 A17 A37 A15 A33 A30 A1 A7 A38 A2 A32 A35 A20 A28 A34 A39 A6 A36 A3 A5 A10 A14 A16 A18 A23 A40
Polysemous word (info) S TK NK 13 1 10 4 12 1 12 1 12 1 7 5 1 (1) 10 2 10 2 10 2 8 2 3 7 8 1 7 1 7 2 6 2 5 3 6 1 6 4 1 4 1 4 3 1 1 1 2 (2) 3 3 2 1 1 1 1 1 1
WORD (no of correct answers) greppa (14) krasch (14) dragningskraft (13) platt (13) samsas (13) huvudbry (13) revolutionerande (13) atmosfär (13) begrava (12) frälst (10) remiss (10) avbryta (9) tagen (9) nicka (9) gap (8) fiktion (8) grina (7) harmoni (6) flyktig (5) darra (5) börja (4) skugga (4) teknikalitet (4) ersättning (3) fullfjädrad (3) herravälde (2) brunn (1) dirigera (1) avig (1) krafsa (1) marginal (1) arbetsam (1) tillfälle (0) dryg (0) element (0) depression (0) jäsa (0) lukta (0) koloni (0) tetig (0)
1 14 8 11 11 11 11 11 10 12 7 4 8 6 6 4 3 6 4 3 4 4 3 1 3 1 2 1 1 1 1 1
Sure 2 3 14 12 11 10 10 10 8 11 11 9 11 8 9 7 8 6 3 5 5 7 7 6 5 6 5 6 6 5 5 3 3 4 3 3 2 1 2 1 3 3 2 2 1 1 1 1 1 1 1 1 1
4
5
Meanings Think sure 1 2 3 4 5 4 1 1 1 1 1 2
4 4 3 4 1
4
1 5 1 2 2 4 3
1 1 2 2 2 3 1 3 1 4 1 1 3 1 1
2 1 1
1
2
1
1 1
2
2 1 1 1 1
1
1
Not sure 2 3 4 5
1
2 1 1 2 2 2 1
1
2
1 1
1
1 1
1
1 2
1
1
1 1 2
1 2 1 2
1 1 1
1
1
1
1
presented, starting with the word that most of them knew. (When more than one test item received the same score, the polysemous word for which the informants showed the greatest accuracy in their self-evaluation is placed first.) To exemplify, for test item 12 (journal) – that is, the 12th most common item – 11 of the 15 students knew the two meanings sought and did not indicate any meanings that were incorrect. Eight of the students (indicated in grey since they were in majority) were absolutely sure that they knew the polysemous word (column S), while three students showed less certainty, indicating that they only thought they knew the word
198 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(column TK). When it comes to the different senses of this word, eight students indicated that they were sure of the most frequent meaning (‘Meanings/sure’, first column, and indicated in grey since they were in majority) and two that they thought they were sure (under ‘Meanings/ think sure’, first column). As for the second most frequent sense, nine of the students were certain that they knew it (under ‘Meanings/sure’, second column, and again coloured in grey since these were in majority), while one indicated a lesser degree of certainty (under ‘Meanings/think sure’, second column). Table 6.7, showing the students’ results on the Swedish test part, should be interpreted in the exact same way. The lack of knowledge depth in the students’ L2 as well as their L1, as indicated in the mean scores, is substantiated by the fact that there is a concentration of polysemous words for which only two meanings were sought at the top of the tables, and a similar concentration of items for which the students had to have acquired four or five senses to receive a point at the bottom of the tables. In fact, while the 7/6 top-most words on the L2/L1 tests involve polysemous words for which only two senses were sought, nine out of the 10 items receiving no correct answers on the English test and six out of eight items on the Swedish test involve polysemous words for which either four or five senses were sought. Despite their lack of vocabulary depth, the informants seem to be confident concerning their knowledge, stating for an overwhelming majority of the test items that they are certain that they know the polysemous words as well as their senses. This is especially true for the first and second senses (colour-coded in grey in Tables 6.6 and 6.7). Lastly, it is also interesting to note that word class affiliation plays some role in whether the senses of a word are known or not, with adjectives, described as a more difficult word class, generally receiving lower scores than nouns and verbs on the L2 test. In Tables 6.8 (English items) and 6.9 (Swedish items), the results of the words for which the students were unable to offer all the correct senses and/or indicated meanings that were incorrect are presented. While the word with the fewest number of incorrect answers here comes first, these two tables should essentially be deciphered in the same way as Tables 6.6 and 6.7 discussed above. There are only three additions. Firstly, there is one column indicating the total number of incorrect senses offered by the students for a particular word. For the word idle, for instance, a total of 14 incorrect meanings were indicated. Secondly, there is one column showing the distribution of correct meanings for the different senses. For the English test item nail, for example, the student who made the mistake got both meanings right but also indicated one incorrect meaning. Thirdly, if the most frequent sense tallies with the word’s core sense (as presented in The New Oxford Dictionary of English, 1998), an asterisk has been used to indicate this correlation. (No dictionary indicating the core senses for the Swedish words could be found.) As in Table 6.6, the items found on West’s service list (1953) are indicated in bold in Table 6.8.
A36 A18 A35 A20 A10 A26 A12 A22 A4 A29 A23 A14 A33 A21 A3 A13 A34 A40 A7 A6 A32 A19 A38 A11 A15 A9 A5 A2 A17 A24 A1 A8 A16 A25 A27 A28 A30 A31 A37 A39
No.
5 (5) 2 (2) 6 (6) 3 (3) 1 (1) 3 (3) 7 (7) 1 (1) 1 (1) 6 (6) 3 (3) 7 (7) 6 (6) 5 (5) 7 (7) 8 (8) 6 (6) 11 (11) 12 (12) 9 (9) 9 (9) 9 (9) 9 (9) 8 (8) 8 (8) 7 (7) 5 (5) 6 (6)
2 (2) 2 (2) 1 (1) 3 (3) 1 (1) 3 (3) 1 (1) 5 (5)
1 (1)
S
2 (2) 1 (1) 4 (4) 3 (3) 4 (4) 1 (1) 6 (6) 3 (3) 5 (5) 3 (3) 6 (6) 8 (8) 6 (6) 4 (4) 10 (10) 6 (6) 6 (6) 7 (7) 6 (6) 7 (7) 7 (7) 6 (6) 6 (6) 8 (8) 2 (2) 3 (3) 5 (5) 6 (6) 6 (6) 6 (6) 6 (6) 7 (7) 6 (6) 8 (8) 8 (8)
1 (1) 1 (1)
1
1 1
2
4
1 1
1
Polysemous word (info) TK NK 1 (1) crooked (0) comfort (1) (n) nail (1) (n) expand (3) intelligence (3) vocabulary (3) journal (4) withdraw (5) weight (6) torch (6) (n) support (6) (v) exclusive (7) expose (8) pursuit (8) ability (9) resignation (10) suppress (10) brood (10) (v) credit (11) (n) administration (11) idle (11) (adj) dealer (12) haul (12) (v) peak (13) (n) wire (13) (n) keen (13) domestic (14) director (14) suit (14) ugly (14) board (15) sheet (15) square (15) (n) grace (15) (n) needle (15) (n) rude (15) trunk (15) brow (15) advance (15) (v) discharge (15) (v)
WORD (no of errors) 1 * 1 1 3* 3* 1* 4* 3* 6* 6* 3* 5 7* 5* 9* 10* 5* 2 3 6* 2 11* 7* 11* 11* 4 11* 7* 14* 14* 15* 15* 14* 10* 15* 15* 4* 15 11* 8 1 5 10
4 4 5 10 2 13 4 10 14 7 15 10 13 2
3 4 5 4 2 6 4 8 4 4 1 6 9 9 10 2
1 1 3
2
1 2 7 15 7 6 11
15
12 3 6 8 5 4 3
8 3 2 11
5
2
5
2 2
3
4 3
13 2 6 14 1 6 12
10 8 7
12
9 6
4
Correct meanings
12 8
5
2 6
1
7 5
4
16 4 8
2 14
5 2 6 3 5 9 11 1 8
1 1 3 3 2 4
Incor. mean.
2 2 1 8 5 10 9 2 8 5 12 12 12 14 12 7 14 14 2 14 7 3
3 3 5 6 3 1 5 3 8 6 2
1 1 3 3
1
1 2 4
1 1 4 2 1 11 1 3 10 4 12 8 12 1
3 5 8 5 1
3 4 4 3 1 5 3 6 3 2
1 1 2
2 13 3 4 6
14
10 1 3 7 2 1 1
6 3 1 9
5
2
4
1
Sure 2 3
2 2
4 5
1
1 1 3 9
2
2 1 1
1 1 1 1
2
3 2 2
3 1 1
1
2 10
6 5
5
10
7 3 1
10
5 3
4
1 4
1
2 3 4 1 2 2
4
3 1 3
1 1
1
1
1
2 4
1 1 1
2 1
1
1
1
1
1 2
1
1
3
2
2 4 2
1
3 2
Meanings Think sure 1 2 3 4
5 2
5
1
1
2
1 1 1 1 1
1
1
1 1
2
2
3 2 1 4 1
1 1
2
1
1 1
1
4
1 1 4
1 1 1 1 2 2
1
1
1 2 2
2 4
3
1
Not sure 2 3 4
1
5
Table 6.8 The students’ results on and evaluation of the English test on polysemous words with a focus on the items for which the informants were unable to give all the correct senses (S = sure that the word is known, TK = the word is thought to be known, NK = the word is not known)
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 199
A19 A26 A24 A12 A13 A29 A25 A11 A27 A31 A8 A9 A4 A21 A22 A17 A37 A15 A33 A30 A1 A7 A38 A2 A32 A35 A20 A28 A34 A39 A6 A36 A3 A5 A10 A14 A16 A18 A23 A40
No.
S 1 (1) 1 (1) 2 (2) 2 (2) 2 (2) 2 (2) 1 (1) 1 (1) 2 (2) 2 (2) 2 (2) 4 (4) 5 (5) 4 (4) 5 (5) 3 (3) 5 (5) 6 (6) 5 (5) 8 (8) 10 (10) 10 (10) 4 (4) 10 (10) 6 (6) 3 (3) 11 (11) 12 (12) 7 (7) 8 (8) 8 (8) 10 (10) 14 (14) 13 (13) 10 (10) 11 (11) 11 (11) 13 (13) 7 (7) 3 (3)
1 (1) 5 (5) 1 (1) 5 (5) 10 (10) 3 (3) 2 (2) 5 (5) 5 (5) 5 (5) 4 (4) 1 (1) 1 (1) 3 (3) 3 (3) 3 (3) 2 (2) 8 (8) 7 (7)
1 (1) 1 (1) 1 (1) 2 (2) 3 (3) 2 (2) 1 (1) 2 (2) 2 (2) 4 (4) 3 (3) 3 (3) 5 (5) 2 (2)
4
1
1
1
1
2
1
Polysemous word (info) TK NK
greppa (1) krasch (1) dragningskraft (2) platt (2) samsas (2) huvudbry (2) revolutionerande (2) atmosfär (2) begrava (3) frälst (5) remiss (5) avbryta (6) tagen (6) nicka (6) gap (7) fiktion (7) grina (8) harmoni (9) flyktig (10) darra (10) börja (11) skugga (11) teknikalitet (11) ersättning (12) fullfjädrad (12) herravälde (13) brunn (14) dirigera (14) avig (14) krafsa (14) marginal (14) arbetsam (14) tillfälle (15) dryg (15) element (15) depression (15) jäsa (15) lukta (15) koloni (15) tetig (15)
WORD (no of errors)
2 1 2 2 1 1 3 5 2 5 3 5 4 6 8 3 5 9 11 5 6 12 2 12 14 14 12 14 13 14 11 9 11 15 13 9 11 12
1 1
2 1 1 2 2 1 2 1 2 1 4 5 6 5 2 6 5 2 6 7 4 11 8 8 10 12 8 2 13 9 7 12 3 14 14 5 14 7 10 13 2 14 11 10 3 6 8
12 1
6
4 5
4
7 8 2 9
4
3 5 15 1
5 11 12 6 13 6 9 3
6 9 4 2 4 9
1 4 5 6
2
2
3
Correct meanings
2
3
9
5
13
2 16
1 13 3 15 5 2
3 2 2 1 1 11 1 2 3
1 5 8
1 5 3
1
2 1 1 3
1
Incor. mean.
2 1 2 2 1 1 2 3 1 4 2 2 4 4 6 2 5 8 11 4 4 11 1 7 13 9 9 11 9 10 9 6 7 13 10 8 7 9
1
2 1 3 3 4 5 2 5 3 1 2 3 3 11 7 4 9 8 4 2 11 5 6 6 2 14 11 4 11 6 6 5 2
1
1 2 1
8
2 2 13
4 8 6 1 11 1 6 1
3 7 2 1 3 8
1 3 3 5
1
1
Sure 2 3
3 7
12 4 7 1
3
2 2
3
5
4 6
4
1
7
5
3 2
2 2 2 2 2 3 1 1
2
2
1
1 1
1 2
1
2 1 2 6
1
2 3 1 4 1
1 1 2 1 2 3
1 1 3 2
1
1
1 1
2 1 1
2 4 3 1 3 2 1
1 1 1 1
1 1
2 1
1 5 3 1
3
1
1
2 1 2 2
Meanings Think sure 1 2 3 4
1
2
5
1
1
1
2 2
1
1
1
1
1
1
1
1
2
1
1
1
1
2
1
1 1 1
1
1 1
1
1
2
1
1
1
2
1
Not sure 2 3 4
2
5
Table 6.9 The students’ results on and evaluation of the Swedish test on polysemous words with a focus on the items for which the informants were unable to give all the correct senses (S = sure that the word is known, TK = the word is thought to be known, NK = the word is not known)
200 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 201
As discussed in connection with Tables 6.6 and 6.7, the informants showed, somewhat surprisingly, great confidence in their mastery of words for which they knew all the correct senses and had indicated none of the incorrect senses. What is even more surprising is that the subjects also displayed great confidence with items for which they were incorrect. In fact, most of the students wrote that they did not know the polysemous word in only eight out of 40 cases (column NK), with the vast majority indicating that they were either sure of the item or thought that they knew the item. As before, this is not only limited to their L2. Although less pronounced, the same thing can be seen in their L1, where again the vast majority of the subjects wrote that they were sure of words for which none of them could give all the meanings sought. It is also interesting to consider the incorrect senses offered by the informants. As in Odefalk’s (2004) investigation, many of these were synforms. For example, quite a few of the participants thought that beundrande (idolising (from idol)) was one of the meanings of the polysemous word idle. Similarly, even more students believed that yla (howl) was one of the correct senses of haul, probably visualising the orthographic form of the word howl without being sure of its exact pronunciation. Another common group of incorrect meanings were those that were phonologically and/or orthographically reminiscent of the word tested. This appeared to be the case especially if the first syllable was similar, as exemplified by the word bredda (broaden), which a great many of the learners thought was one of the meanings of the word brood. The bathtub effect, according to which the first part of a word is most salient and thus remembered the best (see Section 6.1), appears to be at work here. In general, it seems that the subjects in the present study based many of their incorrect answers on phonological similarities. This, as explained in Section 6.1, is normally an indication of low proficiency. Since acquiring senses of polysemous words is equal to developing vocabulary depth, which is more difficult than developing vocabulary breadth, it may be that the students, not having created enough links, were subconsciously prompted to resort to the more basic approach of phonological similarity. Finally, in Table 6.10 the students’ results in relation to meanings (1st, 2nd, 3rd, 4th or 5th) of the words for which they were not able to give all the correct senses and/or gave incorrect senses are presented. As stated in Schmitt (2010c) and confirmed by Odefalk (2004), learners appear to have a better knowledge of the most frequent sense of a polysemous word than the less frequent ones, and this is true of the informants in the present study too. The only anomalies in Table 6.10 are the results on the fourth and fifth senses with the English items, where there is an increase in correctness. However, it must be assumed that the main reason for this is that the number of possible correct meanings here is relatively low (210 for the fourth sense and only 30 for the fifth sense) and can therefore not really be compared to the number of first (402 possible senses), second (again 402)
202 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 6.10 The total number of correct answers in relation to meaning: Swedish students Test
Polysemous word 1st meaning
2nd meaning
3rd meaning
4th meaning
5th meaning
English test
73.9% (=297/402)
53.5% (=215/402)
44.6% (=145/325)
53.8% (=113/210)
66.7% (=20/30)
Swedish test
77.8% (=288/370)
66.8% (=247/370)
51.0% (=156/306)
48.7% (=97/199)
34.1% (=14/41)
Table 6.11 The total number of correct answers in relation to meaning: native speaker Student
Polysemous word
Native speaker
1st meaning
2nd meaning
3rd meaning
4th meaning
5th meaning
66.7% (=12/18)
72.2% (=13/18)
73.3% (=11/15)
60.0% (=6/10)
100% (=1/1)
and third (325) meanings. For the same reason, the results on the fourth and fifth senses on the Swedish test may also be ascribed to mere chance. This is confirmed by the statistical analysis, which could not confirm the differences between the fourth and fifth meaning on the L2 test or the third and fourth as well as the fourth and fifth meanings on the L1 test. It is interesting to note that for the native speaker there is less sensitivity to frequency, as shown in Table 6.11. Finally, in the theoretical background it was suggested that offering the core meaning of a polysemous word may help students comprehend and retain other meanings of the same word. In order to find out if core meanings had a positive effect on the students’ mastery in the present investigation, a comparison between the most frequent meaning tested and the core sense was made. It was thought that if there was a concentration of correlations between core sense and the most frequent meaning in the top of the tables, where the words for which most students had achieved points were listed, the core sense could be shown to have had a positive impact on the students’ results. However, when The New Oxford Dictionary of English was consulted, it turned out that there was an agreement between the core sense and the most frequent sense in the vast majority of cases, thus rendering it impossible to draw any conclusions in this respect. The very fact that the majority of the most frequent meanings were also the core senses may of course have been a contributing factor as to why the students did the best on the most common meanings. The students’ individual scores will be discussed in Section 6.5, together with their results for lexical fields of near synonyms and false friends. We will now continue by exploring the learners’ quantitative and qualitative knowledge of near synonyms.
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 203
6.3 Lexical Fields of Near Synonyms Introduction
‘As the fluency of advanced language students increases, so too does the number of vocabulary errors they generate, both in speaking and in writing. Some of these errors are the result of first language transfer,’ such as false friends, which will be discussed in Section 6.4, ‘but the majority are interlingual, i.e. they reflect confusion between and among lexical items in the second language itself.’ Furthermore, all ‘vocabulary errors manifest one or more kinds of dissonance between the word and its context’ (Martin, 1984: 130). There are four types of dissonances that appear to be especially common in the language of advanced learners: stylistic (*dunk (instead of, for example, dip) the chicken pieces in the beaten egg mixture); syntactic (*the author purports that tobacco is harmful (the verb purport requiring a non-human subject)); collocational (*I used to be a large smoker (instead of a heavy smoker)); and semantic (*I was badly damaged in the accident (instead of injured, which is used for animate objects)) (Martin, 1984: 130). It is the latter of these types that will be in focus here. More specifically the present section will deal with lexical fields of near synonyms, also referred to as synonym sets or synsets. There are many different relations between items belonging to a synonym set. Accordingly, this affects the errors made by L2 learners. One common mistake is to use a semantically restricted term when a more general one should be used. This is exemplified in *I have to rectify (instead of correct) my younger brothers all the time. Rectify is restricted in the sense that it is only used with abstract nouns that denote something bad or wrong; that is, you can rectify abuses, errors, and mistakes, but not brothers. Correct, on the other hand, is a word that can be used to cover a wide range of actions. You can, for instance, correct a person, papers, or your watch. There are other near synonyms in this field that are also restricted in their use. For example, one can reform but not emend a person and one can remedy but not redress a situation. This thus means that ‘hyponyms cannot often be used in all the contexts open to the superordinate. By the same token, co-hyponyms (the entire set of specialised items related to a superordinate) may rarely be used interchangeably’ (Martin, 1984: 133). In other sets, different factors are involved. As already exemplified above with injured versus damaged, the semantic distinction may, for example, depend on whether the subject is animate or inanimate. It may also have to do with whether one is referring to something physical or non-physical. Fluctuate and wobble may be used to illustrate this distinction. Whereas the former verb is limited to non-physical subjects, the latter generally only goes with physical ones. This is the reason why the sentence *The air holes (of the clay sculpture) are filled, which may cause the head to fluctuate later on is semantically incorrect. Near synonyms in a lexical set may also be distinguished from each other
204 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
by intensity. In the following synset, in which surprise is the superordinate, there is a clear increase in intensity: surprise, astonish, amaze, astound and flabbergast, flabbergast being the strongest of all. The choice of synonyms can also reflect the speaker’s attitude. While the verbs persevere and enjoy denote approbation, persist and be subjected to involve negative feelings. Both continue and receive, on the other hand, are quite neutral in this respect. Thus, depending on the choice of verb in a sentence like This region enjoys/receives/is subjected to 30 inches of rain a year, different speaker attitudes may be conveyed. Other semantic relations between near synonyms are, for example, state versus event (e.g. exist (this trend exists on many college campuses) versus happen (*this trend happens on many college campuses)) and volitional versus non-volitional (e.g. deter versus keep (compare the drug deterred/kept him from falling asleep) (Martin, 1984: 133–134). Many items in synonym sets also involve more than one semantic contrast. In Rudzka et al. (1981), for example, a more detailed analysis than a mere increase in intensity is given of the synonym set surprise, astonish, amaze, astound and flabbergast discussed above. While all the items signal ‘affect with wonder’, surprise also means that something is unexpected, astonish that something is difficult to believe, and amaze that something is done to cause confusion, and astound and flabbergast both convey the additional meaning that the situation leaves someone helpless to act or think. Another example, offered in Martin (1984: 134), involves the near synonyms misplace and mislay. While mislay may only mean that you do it by accident, misplace may mean that you do it either by accident or by an act of will. Additionally, whereas one can only mislay physical objects such as one’s documents, one can misplace physical objects as well as abstractions such as one’s trust. The best way to determine how close in meaning two words are is to substitute one for another in a specific context (Church et al., 1994; Miller & Charles, 1991). Theoretical background and previous research
Whereas some more recent research in the area of near synonyms primarily appears to be based on various corpus analyses (Liu, 2010, 2013; Liu & Espino, 2012), there is, regrettably, still a dearth of studies focusing specifically on L2 learners’ mastery of near synonyms. Nevertheless, a few relevant investigations were found, and these will be presented in what follows. The frequency effect of near synonyms in lexical fields
A substantial amount of research has shown that native speakers have good intuitions about the relative frequency of different items of vocabulary (e.g. Strömqvist, 2009). More and more research also shows that commonality affects how words are processed in the mental lexicon of
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 205
non-native speakers (Milton, 2009; Nation, 2001, 2006; Schmitt, 2008). This will be explored here. In Schmitt and Dunham (1999), the effect of frequency was investigated in connection with lexical fields of near synonyms specifically. Their main study included five groups of non-native speakers of English, forming a heterogeneous set of informants both in terms of educational background and mother tongue. A group of native speakers of various educational levels were also incorporated. When selecting the lexical sets to be tested, to ensure the accuracy of the frequencies of the test items, Schmitt and Dunham used two corpora: the BNC and the COBUILD Bank of English Corpus. Only sets where the two corpora agreed as to the relative frequency of the test items in each field were included. Also, sets for which the frequencies of the individual test items aligned the best in the two corpora were prioritised. Thus 12 lexical sets were finally selected (three noun sets, three verb sets and three adjective sets). For each set, five near synonyms were chosen, all of which were similar in length. Moreover, one of the items in each set was picked to be the ‘anchor word’, against which the other four test items were to be rated. In the set given in (4) below, catastrophe was considered the anchor word. If an informant, for instance, thought that disaster was ten times as frequent as catastrophe, they were asked to write 10 after disaster. Similarly, if a subject thought that, for instance, calamity was one third as frequent as catastrophe, they were to write 1/3 after calamity. If the anchor word happened to be unknown, the subject was to assign the anchor role to another of the items in the set. A pilot study indicated that this was a useful approach. (4)
calamity cataclysm catastrophe disaster tragedy
1
In the main study, the informants were asked to make a number of frequency judgements regarding level of difficulty. The first task was to pick out the core word, that is, the most central word in terms of frequency. (Note that the core word is not the same as the anchor word.) The results showed that the native speakers were able to pick out the core word 77% of the time, with the accuracy rate being even higher – 85% – when the two lexical sets where the core word and the anchor word coincided were excluded. For the sets that were completed by the L2 learners, the corresponding percentages were 71% and 79% respectively. The second and more difficult task was, as in the pilot study described above, to rank the items in each lexical set according to their frequency in relation to each other. As expected, the native speaker outperformed the non-natives, who only achieved an accuracy rate of 10%.
206 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
On the third task, the most difficult one, the informants were asked to judge the absolute frequencies of the test items, that is, exactly how often each item occurs in the English language. Here the native speakers’ answers showed that they did not form a homogeneous group, with the more well-educated subjects scoring higher than those with less education. Based on the results, Schmitt and Dunham were able to draw the conclusion that L2 learners who have reached the same level of proficiency as native speakers are almost as good at picking out the core word and rating the absolute frequencies of near synonyms as native speakers. In Liu and Zhong (2014), comparing native speakers of English (undergraduate and graduate students) with Chinese L2 learners while implementing a corpus analysis, the results also show the important role played by commonality. Here the L2 informants did consistently more poorly with low-frequency synonyms than with the more frequent ones. L2 students’ mastery of near synonyms in lexical fields
In Haastrup and Henriksen’s (2000) longitudinal study spanning over three years, young Danish L2 learners’ acquisition of lexical sets in English was investigated. By studying how their receptive knowledge of adjectives of emotion (such as afraid, angry, happy and sad) and physical dimension (in subsets such as large, small and wide) developed in relation to other near synonyms within the same lexical field, the researchers were aiming at gaining insights into how L2 learners build networks, since ‘vocabulary acquisition is more a matter of system learning than of item learning’ (2000: 225); the ultimate goal was to add information to how L2 vocabulary acquisition may be enhanced. Their entire study rests on the assumption that learners build up semantic fields by adding terms to them as they elaborate their vocabularies by creating links between words they already know and new L2 words. During this process of network building, they continually restructure their semantic network in order to accommodate new words. (2000: 221–222)
More specifically, this entails ‘that a certain adjective such as happy finds its place in relation to other adjectives in its lexical field through a growing understanding of, for instance, paradigmatic relations’ (2000: 222), such as antonymy (happy versus sad), synonymy (happy and thrilled) and gradation (happy versus overjoyed and delighted), all of which have proved to be especially pertinent when studying lexical fields of adjectives (Miller & Fellbaum, 1991). The study includes 17 Danish 12-year-olds who, when the testing started, had been learning English for approximately one and a half years. All in all, there were three test opportunities, occurring at intervals of one year. As a reference corpus, the results of a number of native speakers, also aged 12, were gathered. As already mentioned above, the study
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 207
focused on two types of adjectives: those of emotion and those of physical dimension. While the latter group includes items that are some of the most frequently used adjectives in the English language and thus acquired very early by native speakers, the former group deals with adjectives of a more abstract nature, which are therefore less frequently used. Four tasks were put together, three of which were so-called sorting tasks of increasing complexity. In the first sorting task, the subjects were given a list of adjectives from the two sets tested. Their task was to place the adjectives in one of four categories: happy, afraid, temperature and size. In the second sorting task, adjectives of emotion were presented to the informants on cards in a decontextualised form. The subjects were faced with the task of sorting each card into one of four categories: sad, happy, angry or afraid. In the third sorting task, the learners were asked to grade eight adjectives from the two fields of happy and afraid in terms of their degree of intensity. However, due to a mishap this task became unintentionally easier than the second sorting task and was therefore excluded from the final analysis. Lastly, in the fourth task the informants were instructed to choose near synonyms from the fields of sad, happy, angry and afraid to fit in 16 different situations. To make it more difficult for the students, distracters were included, as illustrated in (5). (5) You are alone in a house. Around midnight you hear footsteps in the hall. You are/feel
afraid. frightened. alarmed. thrilled. gloomy.
Two main research questions were addressed. Firstly, when faced with the same adjective, how do learners perform on tasks involving lexical relations such as synonymy and gradation? Secondly, how does learners’ knowledge develop with items in a specific lexical field? In connection with these research questions, three hypotheses were put forth. The first hypothesis predicted that it would be easier for the L2 learners to make a distinction between an adjective of emotion and an adjective of physical dimension than putting a specific adjective in one of four lexical subsets that each denote a specific basic emotion (afraid, angry, happy and sad). This hypothesis was confirmed for the majority of the items. The second hypothesis held that it would be easier to ascribe the correct basic feeling of two items of the same lexical field than to decide differences in intensity between these two items. Lastly, the third hypothesis predicted that integrating an item into a lexical field is a slow process and that a contextbased category-providing task will create a situation that will increase the speed of the process. The second and third hypotheses were both tested from a developmental perspective. Only two of the 17 informants showed
208 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
improvement, confirming that, as in a native language (Aitchison, 1994), network building in a second language is a difficult and incredibly slow process. The two subjects who displayed progression acquired most of the items in the happy and afraid subsets first, indicating that these subsets were easier than those of angry and sad. Haastrup and Henriksen (2000: 236) mention several factors that may have had an impact on these results. In addition to word-specific features and learner motivation, the frequency of the adjectives and transfer from the informants’ mother tongue appear to have played significant roles as to which adjectives were acquired and which were not. In Wallin and Jaginder (2021), 64 Swedish secondary school students (with ages ranging from 14 to 16) were faced with a connotation test of English words. Each of the 20 words tested was offered in a context, and the informants were asked to rate their degree of appropriateness along a 5-point scale, with the optimal answer resulting in 2 points, thus yielding a possible total of 40 points. As a group the informants achieved a success rate of about 40%, corresponding to an average score of 16, with the highest-achieving learner garnering 33 points and the lowest-achieving accumulating only 5 points. The conclusion was drawn that the L2 learners had a comparatively poor knowledge of the appropriateness of a specific item within a field of near synonyms. Part B of the parallel tests used for the present chapter: Lexical fields of near synonyms
In both languages, 40 words belonging to eight different fields of near synonyms were tested. These were presented to the students according to the total frequency of the items in each field, starting with the most common one. Within each lexical field, on the other hand, the items were presented haphazardly. While the L2 items were picked, on a random basis, from the booklet English Synonyms and False Friends (Hargevik & Stevens, 1978), the L1 items were, again randomly, chosen from a book by Hallström and Östberg (2003), which, among other things, presents a number of lexical fields of near synonyms in Swedish. The number of test items in each lexical field and their word class affiliation are shown in Table 6.12. As can be seen, whereas the number of fields and the number of test items in each field are the same in both languages, the word class in focus varies. On the tests, the students were asked to pair up each item in a synonym set with its correct definition, with each correct answer awarded one point. While the English definitions were based on the Longman Dictionary of Contemporary English (1987), the Swedish ones were taken from the same book as the near synonyms themselves, that is, Hallström and Östberg (2003). This is exemplified in (6) and (7). (Definitions of the English items can be found in Appendix 3.)
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 209
Table 6.12 The distribution of the number of test items and word class affiliations in the lexical fields of near synonyms used in the present study 3 test items involved English test
5 test items involved
2 (1 field of nouns; 3 (1 field of nouns; 1 field of adjectives) 2 fields of verbs)
6 test items involved
7 test items involved
Total
2 (1 field of nouns; 1 8 1 field of verbs) (1 field of nouns)
Swedish 2 (1 field of verbs; 1 3 (2 fields of verbs; 2 (1 field of nouns; 1 test field of adjectives) 1 field of adjectives) 1 field of verbs) (1 field of verbs)
8
(6) a) bright
(of the shade of a colour), (the opposite is dark)
b) fair
(of a colour) strong, clear and easily seen, (the opposite is dull)
c) light
(of a colour) blond, used almost always about hair
(7) a) bound
to push oneself into the air or away from a surface by the force of one’s legs, i.e. the general term for hoppa
b) hop
to move in a light dancing way, with quick steps and jumps
c) jump
to move quickly and suddenly upwards or forwards as if by jumping
d) leap
1) to move along quickly from place to place by jumping movements, 2) to bounce back from a surface
e) skip
(of people) to jump on one leg
f) spring
to jump up in the air or over something
The students were also asked to state to what extent they were sure of the meanings they had assigned the various test items, and whether there were any test items that they did not know at all, as exemplified in (8). (8) a) carpet
1) thick, usually woollen, comparatively small, 2) large warm woollen covering to wrap round oneself
b) mat
heavy woven often woollen material for covering floors or stairs, i.e. the general term for matta
c) rug
a piece of rough strong material for covering part of a floor
Of the English words above, I have not seen the following word/s before:
210 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
I am not sure about the meaning/s of the following word/s: I think I am sure about the meaning/s of the following word/s: I am sure about the meaning/s of the following word/s:
As in Part A (polysemous words), the informants were asked to evaluate the difficulty level of the test part. As before, they could choose between ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. Also, as with the polysemous words in Part A, a comparison between the total frequencies of the test sets was calculated, again showing that the English test items were more common (constituting ≈0.0561% of the BNC) than the Swedish ones (making up ≈0.0142% of Speåkbanken). Again, the conclusion to be drawn, based on frequency alone, must be that the L2 test part is somewhat easier than the L1 one. The L1–L2 difference in frequency was, however, not confirmed statistically, which means that comparisons may be made between the parallel test parts. Still, the difference will be taken into consideration when discussing the results. Results and discussion
Table 6.13 presents the informants’ results on the synsets tested. As anticipated, the native speaker achieved better results, although not considerably so, than the Swedish learners (26 versus a mean score of 21.13), who in turn did better in their mother tongue than their L2 (31.47 versus 21.13). The L1–L2 difference could be confirmed statistically at a significance level of 5%, the validity of which is enhanced further by the fact that the English test items were more frequent than the Swedish ones. It is interesting to note here that the students’ L2 result (a 52.83% correctness rate) tallies well with the findings of Wallin and Jaginder (2021), who investigated secondary school students, achieving a success rate of around 40%. Table 6.13 The students’ results on Part B: lexical fields of near synonyms Students
N
Correctness rate
Mean
SD
Standardised scores
Native speaker
1
65.00% (=26/40)
–
–
–
Swedish students –English test
15
52.83% (=317/600)
21.13
5.88
Highest: 1.85 Lowest: −1.55
Swedish students –Swedish test
15
78.67% (=472/600)
31.47
7.07
Highest: 1.21 Lowest: −1.76
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 211
Furthermore, since both Part A – polysemous words – and the part discussed here tested an equal number of items, a comparison may be made between the two test parts. If Tables 6.5 and 6.13 are contrasted, it can be seen that the Swedish subjects did considerably better on near synonyms than on words with multiple meanings, especially in their mother tongue (31.47 versus a mean of 15.33 for polysemous words) but also in their L2 (21.13 versus 13.20 for polysemous words). This difference also applies to the native speaker, who received 21 points on polysemous words as compared to 26 on the synsets. It must, however, be remembered that the scoring system used for the polysemous words, in which the students had to indicate all the correct senses and no incorrect ones in order to receive one point, was probably more demanding than the scoring system used here, where each correctly picked definition conferred one point. Also, some of the correct answers offered for the near synonyms must certainly be ascribed to the method of elimination, whose effectiveness stands in direct relation to the number of items included in the synset; that is, the fewer the items, the more likely one is to get it right, and the higher the number of items of which the informant is certain. On the other hand, the total frequency of the words with multiple meanings is higher (L2: ≈0.1100%; L1: ≈0.0418%) than the total frequency of the tested synsets (L2: ≈0.0561%; L1: ≈0.0142%). This holds true for both languages. In terms of frequency then, the polysemous words must be considered easier than the near synonyms. In addition, the sometimes very finely tuned differences between the near synonyms in a synset certainly take away some of the advantage of being able to use the method of elimination. With the above taken into account, it appears that the students’ knowledge of synonym sets may be more developed than their knowledge of polysemous words, in their L2 as well as their L1. Put differently, the students’ restructuring of the semantic networks may have enabled them to incorporate more near synonyms than different senses of polysemous words. Moreover, in contrast to their self-evaluation of the polysemous words for which they clearly overestimated their mastery, the students here display a good take on their knowledge, with the majority of the L2 speakers (nine students) indicating that the English test part was either ‘difficult’ (six subjects) or ‘very difficult’ (three subjects), and that the Swedish test items were either ‘easy’ (six informants) or even ‘very easy’ (one informant). (For both tests, six participants found the test part at hand to be ‘average’.) (The native speaker, on the other hand, again appears to have overestimated his knowledge, indicating that this test part was somewhere between ‘easy’ and ‘average’.) The L1–L2 difference is also confirmed in the correctness rates for items whose meanings the informants stated they were sure about (72.1% on the L2 test (the bottom of Table 6.14) versus 86.8% on the L1 test (the bottom of Table 6.15)) as well as items they thought they were sure about (60.4% on the L2 test (the bottom of Table 6.14) versus 71.6% on the L1 test (the bottom of Table 6.15)).
212 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Table 6.14 The students’ results on and evaluation of the English test on lexical fields of near synonyms. Figures in parentheses in columns ‘Sure’ to ‘Not sure’ indicate cases where the students were incorrect in their self-evaluation. In the ‘Not seen’ column, a figure in parentheses means that the student was unable to provide the correct answer Lexical field
Word (correct answers)
Sure
Think sure
Not sure
Not seen
LF2 (38/45)
bright (13) light (13) fair* (12)
12 (1) 12 (2) 9 (2)
3 (1) 3 4 (1)
2 (2)
LF5 (37/45)
carpet (14) rug (12) mat (11)
3 1
8 9 (1) 7 (1)
1 (1) 2 (1) 4 (3)
LF4 (49/75)
solicitor (11) lawyer (11) attorney (7) barrister (12) advocate (8)
4 5 (2) 4 (2) 3 3 (1)
3 (1) 5 6 (3) 5 (1) 3 (2)
3 (3) 2 (2) 2 (1) 4 (3) 4 (2)
LF6 (38/75)
claim (4) maintain (14) state (10) assert (4) allege (6)
3 (2) 2 3 (1) 1 (1) 3
6 (4) 6 6 (1) 4 (3) 3 (2)
3 4 (3) 3 (1) 8 (2) 7 (2)
LF8 (26/75)
tremble (6) shiver (10) shudder (3) quiver (4) quaver (3)
2 (2) 2 (1) 1 (1) 1 (1)
6 (2) 8 (2) 5 (4) 3 (3) 1 (1)
4 (2) 2 (2) 3 (1) 3 (1) 4 (2)
1 4 (1) 6 (1) 10 (1)
LF3 (61/90)
accident (15) disaster (8) catastrophe (10) misfortune (13) calamity (5) mishap (10)
7 6 (3) 7 (2) 4
4 5 (1) 4 (1) 6 (2) 3 (2) 4 (1)
2 (2) 2 (1) 2 (2) 2 (2) 1 1 (1)
11 (4) 7 (3)
7 2 1 3
5 5 (4) 3 (3) 4 1 (1) 4 (3)
1 5 (2) 8 (3) 5 (1) 7 (1) 8 (1)
2 2 (1) 2 6 (2) 3 (1)
6 (5) 3 (3) 3 (2)
3 (3) 8 (7) 7 (4)
3 (1) 1 (1) 129 (36) 72.1%
1 (1) 6 (3) 5 (3) 182 (72) 60.4%
1 1 1 (1) 5 (1) 2 (2) 3 (2) 4 126 (56) 55.6%
LF7 (39/90)
LF1 (29/105)
(317/690)
jump (13) leap (6) hop (5) skip (9) bound (3) spring (3) disease (3) illness (2) sickness (6) infirmity (2) ailment (4) disorder* (9) complaint* (3) Total Correctness rate
2
3 (2) 3 (1) 1 2 (1)
2 (1)
10 (1) 11 (1) 1 87 (21) 75.9%
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 213
Table 6.15 The students’ results on and evaluation of the Swedish test on lexical fields of near synonyms. Figures in parentheses in columns ‘Sure’ to ‘Not sure’ indicate cases where the students were incorrect in their self-evaluation. In the ‘Not seen’ column, a figure in parentheses means that the student was unable to provide the correct answer Lexical field
Word (correct answers)
Sure
Think sure
Not sure
Not seen
1 (1) 3 (2) 1 (1)
3 (1) 1 (1)
2 (2)
4 (4)
lipa (13) snyfta (12) snörvla (14)
(=blubber) (=sob) (=snuffle)
11 (2) 12 (3) 12 (1)
3 2 2
LF5 (29/45)
förnärmad (9) förtretad (9) vred* (11)
(=offended) (=annoyed) (=furious)
4 (1) 3 (1) 5 (1)
6 (3) 3 (1) 4 (1)
LF4 (71/75)
snegla (14) kisa (15) light)) glo (14) plira (14) blänga (14)
(=glance) (=peer (e.g. bright
10 9
3 (1) 4
(=stare) (=peer) (=glare)
10 4 10
3 (1) 3 3 (1)
LF7 (67/75)
dugga (14) skvätta (13) strila (14) skvala (13) hällregna (13) dogs)
(=drizzle) (=splash) (=raining gently) (=pour) (=raining cats and
11 10 9 6 8 (1)
2 2 (1) 3 4 (1) 4
LF3 (48/75)
välmående (11) förmögen (6) stormrik (11) burgen (9) family) tät* (11) (slang))
(=flourishing) (=wealthy) (=immensely rich) (=from a wealthy
4 5 (2) 7 (1) 2
6 (4) 6 (5) 4 (1) 2 (2)
(=rolling in money
7 (1)
3
fnissa (11) flina (14) fnittra (11) flamsa (14) laugh) flabba (13) skrocka (15)
(=giggle) (=sneer) (=giggle nervously) (=fool around and
8 8 8 6 (1)
5 (3) 5 (1) 5 (3) 7
(=guffaw) (=chuckle)
5 (1) 5
6 (1) 6
dam (12) tant (13) donna (12) kär(r)ing (12) missus) fruntimmer (11) depreciatory) gumma (12) missus)
(=lady) (=nice lady; auntie) (=bird, broad) (=hag; old woman;
10 (2) 10 (2) 8 (2) 8 (1)
4 (1) 4 4 (1) 5 (1)
(=jokingly or
8 (3)
5 (1)
(=old woman;
10 (2)
4 (1)
diskutera (11) argumentera (12) konversera (10) babbla (6) orera (8) tjattra (12) samspråka (9) Total
(=discuss) (=argue (pro or con)) (=converse) (=babble) (=speechify) (=chatter) (=have a chat, talk)
8 (2) 8 (2) 6 (3) 8 (3) 1 6 5 (1) 295 (39) 86.8%
4 (1) 4 (1) 5 (1) 3 (2) 5 (3) 4 5 (3) 162 (46)
1 3 (3) 2 (1) 2 (2) 23 (20)
27 (19)
71.6%
13.0%
29.6%
LF8 (39/45)
LF6 (78/90)
LF2 (72/90)
LF1 (68/105)
472/690
Correctness rate
1 (1) 3 (3) 2 (1) 1 (1) 1 (1)
8 (5) 1
2 (2) 2 (2)
2 (2)
4 (3)
214 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Tables 6.14 (English test) and 6.15 (Swedish test) offer the L2 learners’ results on each of the synonym sets tested. The sets are presented according to (1) the number of items involved, starting with the synsets with the fewest number of members, and (2) the total number of correct answers, beginning with the set that received the highest score. (The items indicated in bold were found on West’s (1953) general service list.) On the English test, this means that the synonym sets bright–light–fair and carpet–rug–mat, involving only three test items each, are presented first, bright–light–fair preceding carpet–rug–mat because it received more correct answers. The synonyms within each set are also presented according to frequency, once again starting with the most common item. (There are three exceptions on the English test part (fair, disorder and complaint) and two on the Swedish test (vred (furious) and tät (rolling in money)), which all have more than one meaning, and for which, based on the two corpora used, it was therefore impossible to determine frequency. These were instead placed last in each synset and are indicated by an asterisk.) The Swedish test results are presented in exactly the same fashion. (As mentioned before, the definitions of the English near synonyms can be found in Appendix 3.) As already indicated by the mean scores (Table 6.13), the Swedish learners display considerably more confidence with near synonyms in their L1 than their L2, having accumulated a total number of 472 points in their first language as compared to 317 in their second. This result is shown in several ways in the tables. For instance, whereas the majority of the students indicated complete unfamiliarity with only two of the Swedish test items, as many as seven of the English words were not known to the majority of the learners, three of which (shudder, quiver and quaver) belong to Lexical Field 8, thus rendering this set the least best known of those containing five items. Similarly, while for the overwhelming majority of the Swedish test items most of the students indicated that they were sure of the definition chosen, this only occurred with seven of the English test items: bright, light and fair (Lexical Field 2), solicitor (Lexical Field 4), accident and catastrophe (Lexical Field 3) and jump (Lexical Field 7). Interestingly, these results largely appear to stand in direct relation to the frequency of the item in question. As discussed in the theoretical background, the effect of frequency was also observed in Schmitt and Dunham (1999), in which it was demonstrated that the non-native speakers were almost as good as the native speakers at picking out the most frequent word in a synonym set. The positive effect of frequency was also confirmed in Haastrup and Henriksen (2000) and Liu and Zhong (2014), also discussed in the theoretical background. With the Swedish items, on the other hand, of which the subjects were more certain, there is also less sensitivity to frequency. Moreover, the impact of frequency observed with the English test items becomes even clearer if the words are compared on an individual
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 215
basis, irrespective of which synset they belong to. Of the 21 English items for which the majority of the students gave the right definitions, 15 are among the 20 most frequent items. Consequently, of the 19 items for which the majority of the students did not succeed in offering the right definitions, only five are among the 20 most frequent items. The less pronounced effect of frequency in the students’ L1, on the other hand, makes the comparatively frequent items more evenly distributed. Additionally, on the L2 test, it is not only within synsets that the frequency effect appears to be at work. If whole synonym sets with the same number of items are compared, it can be seen that those for which the total frequency is the highest are consistently also those that the students do best with. Again, this does not apply to the L1 test. Finally, it is also interesting to note that for the three synsets that involve five items each (Lexical Fields 4, 6 and 8) as well as the two synsets that involve six items each (Lexical Fields 3 and 7), the subjects did better with those concerned with nouns than those dealing with verbs. This tallies with the fact that the former type is generally considered easier to deal with than the latter, which was also observed in Chapter 3 dealing with the learners’ affixation skills. (See also Schmitt & Zimmerman, 2002, for the same conclusion.) The students’ individual scores will be discussed in Section 6.5, together with their results for polysemous words and false friends. We will now continue by investigating the learners’ quantitative and qualitative knowledge of false friends. 6.4 False Friends Introduction
As discussed in the introduction to this chapter (Section 6.1), for someone trying to learn a second language one of the main tasks is to create a new form network by transferring already existing L1 properties, ‘the basic structure of a semantic and conceptual network’ already being ‘in place through L1 lexical/conceptual knowledge’ (Wolter, 2006: 743). Moreover, [t]his network acts as an integrated set of ‘placeholders’ for L2 lexical items, but the learner will still need to fill out the network with L2 words. In some cases their existing L1 lexical/conceptual network will suffice, and slotting L2 lexical items into the network will be fairly straightforward. In other cases, however, the network itself will need fundamental restructuring in order to accommodate divergent properties. (Wolter, 2006: 743)
This is the case with so-called false friends, 2 for which, if an error has been made, a learner may, for instance, have incorrectly assumed that a word in his/her L2 that is phonologically and/or orthographically similar to a word
216 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
in their L1 denotes the same thing (Wolter, 2006). While some false friends appear to cause problems to a more limited set of learners, that is, they are comparatively language-specific, others seem problematic across many L2s. Examples of the latter type are the English words actual, eventual and sympathetic, which in Swedish, German, French and Spanish may, due to similar-looking words, easily be misinterpreted to mean current, possible and nice. (See Swan & Smith, 2001, which contrasts English with various other languages from a grammatical, phonological and lexical point of view. For even more examples of false friends, see Hill, 1982 and the Cambridge International Dictionary of English, 1995: 1772–1773.) Theoretical background and previous research
Compared to syntactic interference, there is very little research focusing on negative transfer of vocabulary (Haastrup, 1991), and this is especially true for false friends. There does, however, exist a body of research on inferencing and communication strategies that tends to incorporate discussions on lexical transfer, the former of which, together with other research, will be discussed in what follows. In one study on vocabulary transfer into English (Ard & Homburg, 1983), L2 learners speaking two typologically distant languages – Spanish and Arabic – were investigated and compared. Two parameters, along which the students’ vocabulary transfer was measured, were developed, one indicating orthographic and morphological similarity between L1 and L2 words and one indicating semantic similarity. Both parameters had a scale from 1 to 6. On the parameter showing orthographic and morphological similarity, 6 indicated the highest degree of similarity. On the parameter showing semantic similarity, 1 indicated that the first meaning of the L1 word listed in the dictionaries investigated – that is, the most frequent sense – was basically the same as the meaning of the English word, followed by a decrease in similarity for 2, and so on. The two parameters were then combined so that each word tested could be assigned a slot in a 6 × 6 matrix. This means that words that scored 6 on both parameters were categorised as true false friends. For instance, deplazar, which means displace not displease, and sucesivo, which means successive and not successful, are partly false friends, their orthographic form not being entirely identical to the English spelling. The results of the investigation confirmed the researchers’ hypotheses. Firstly, the least transfer occurred with items that displayed high similarity on both scales. In Ringbom’s (1987: 53) words, [l]earning to understand a word is most obviously facilitated if the new item has full identity, both phonological and semantic, to the L1 item. When both phonological and semantic similarity work together, the effect is like that of a magnet attracting a new word to be stored in the learner’s mental lexicon when he meets it for the first time.
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 217
Secondly, due to the formal similarities between lexical items in their mother tongue and in English, the Spanish speakers made more use of transfer than the Arabic-speaking learners. In another study on transfer of idiomatic expressions (Kellerman, 1977), native speakers of Dutch from different proficiency levels learning both English and German were investigated. A number of idioms from their L1 were translated literally into both English and German, some of which had direct counterparts in the L2 and some of which had not. Three main results were obtained. Firstly, the informants indicated that they thought that German was more closely related to Dutch than English. In connection with this result, Kellerman emphasises that it is not the distance between languages per se but the psychotypology – that is, the learners’ notion of whether their mother tongue and a given L2 are closely or distantly related – that decides the extent to which the L1 is used. This means that if a student feels that the L2 they are trying to learn has few affinities with their L1, transfer will not be a prominent learning strategy. If, on the other hand, a student perceives the L2 to have strong affinities with their native language, transfer will be used extensively (cf. Faerch & Kasper, 1987). Secondly, the Dutch informants were able to make a clear distinction between language-specific items (i.e. expressions that could not be transferred) and language-neutral items (i.e. expressions that could be transferred from their native language into the L2). Lastly, due to the fact that the first-year university students accepted more transfer than the second-year students, Kellerman drew the conclusion that the informants’ perception of what is language-specific appears to change over time (cf. Jordens, 1977). In another investigation focusing on lexical inferencing procedures (Haastrup, 1991), vocabulary transfer was one factor looked into. The study included both high- and low-proficiency students, all of whom were taking part in the upper secondary school system. Differences were observed in the inferencing skills of low versus high achievers due to the fact that ‘learners proceed along a continuum leading from the use of formal similarity only to the use of formal + semantic similarity’, where ‘in the initial stages of learning clang associates are common’, whereas ‘as the learner progresses in the L2, she develops semantic, L2-based associative networks’ (Haastrup, 1991: 204). This was confirmed in Haastrup’s investigation, where the intermediate learners made more use of formal similarity than the advanced learners, thus making more transfer errors based on orthographic/phonological similarity than the high achievers (cf. Hatch, 1983; Meara, 1978) To sum up, researchers who have studied transfer in relation to lexis seem to agree that the main research question to be addressed is the relative importance that learners attach to phonological/orthographical similarity (formal similarity) as
218 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
compared to semantic similarity. Both forms of similarity play a role, but there are no precise predictions as to the balance between them for learners at different proficiency levels and with different L1s. (Haastrup, 1991: 54) Part C of the parallel tests used for the present chapter: False friends
On the English part, the test items, picked randomly from Hargevik and Stevens (1978), were presented in 10 pairs, with each correct item awarded 1 point (i.e. 20 in total). While one of the items in each pair was offered in an English context and the students were asked to give a Swedish translation, the other item, likely to be perceived as an equivalent to the test item given in English but actually a false friend, was offered in a Swedish context and the subjects were asked to give the corresponding English word. This is exemplified in (9) and (10); konsekvent and novell translate as consistently and short story respectively. (9) H e did not bother to show up and consequently we will not help him anymore. = (10) D id you read the novel I gave you last week? =
Han har konsekvent gjort på det viset.
= Den här novellen är underbart rolig! =
(The native speaker was instead given definitions in English for both of the items and asked to fill in the right words in sentence slots. This means that, from his perspective, this was not really a test on false friends.) On the corresponding Swedish test part, the informants were presented with 20 words (1 point/correct answer) that could easily be confused, orthographically and/or phonologically, with an English word, and asked to explain their meanings in Swedish. For example, the Swedish word sensibel, which may be thought to mean the same thing as the English word sensible, in fact means sensitive. Similarly, even though the Swedish word kreatur looks very similar to the English word creature, it does not share its semantic meaning but instead refers to cattle. (All the Swedish test items were also picked randomly from Hargevik and Stevens (1978).) Additionally, to prompt the students in a false way, after each L1 item tested, the English similar-sounding/similar-looking word was given in parentheses. Before taking the tests, the students were offered a definition of the term false friend and were told that that was what they were going to be tested on.
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 219
Furthermore, on both the L1 and L2 test parts, the items were presented in order of frequency, based on the BNC and Språkbanken respectively, starting with the most frequent item. Since the test items on the English test were presented in pairs, it was the total frequency of each pair that decided where it ended up on the test. As for the Swedish test part, on which items were presented one by one, it was the individual frequency that determined the position. The students were also asked to supply information about how well they knew the test items. Example (11) shows how this was done on the English test part. (11) Her sister works as a curator Kuratorn på vår skola är riktigt bra. of a museum in France. =
=
□ I don’t know this word/phrase. □ I don’t know the English word. □ I’m guessing the meaning of the □ I think I know the English word. word/phrase from the context □ I am sure I know the English word of the sentence. □ I recognise this word/phrase, but I don’t know what it means. □ I recognise this word/phrase and I’m guessing its meaning from the context of the sentence. □ I think I know what the word/ phrase means. □ I am sure I know what the word/ phrase means.
For both items in a sentence pair on the English test, the students could either indicate that they were sure of the test item, that they thought they were sure of it, or that it was unknown. These three alternatives were also used on the Swedish test. In addition, for the English test items, the subjects could also indicate either that that they recognised the word but still could not come up with an answer (also an alternative on the Swedish test), that they recognised it but had still used the context to arrive at their answer, or that they did not know it but had tried to guess from the context offered. As in the previous two test parts, the students were also asked to report on the difficulty level of Part C by indicating that it was ‘very easy’, ‘easy’, ‘average’, ‘difficult’ or ‘very difficult’. As before, to be able to make fair comparisons between the two languages, the total frequencies of the two sets were calculated, the L2 test items being more common (constituting ≈0.0294% of the BNC) than the L1 ones (making up ≈0.0083% of Språkbanken). Thus, this renders the English test part, at least from a frequency perspective, somewhat easier
220 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
than the Swedish test part. Similar to what was observed with the polysemous words and the near synonyms, the L1–L2 difference in frequency was, however, not confirmed statistically, which means that comparisons may be made between the parallel test parts. Nonetheless, the difference will be taken into consideration when discussing the results. Results and discussion
The students’ mean scores are presented in Table 6.16. As can be seen, the Swedish students’ mastery of false friends is quite poor, with them not even achieving an L2 correctness rate of 50% (mean: 8.93 out of 20). The reasons for this will be discussed in more detail in connection with Table 6.17. The students’ evaluation of the difficulty level of the test part reflects the students’ less than stellar results quite well, with the majority of them (nine students) thinking that the English test part was either ‘difficult’ (five students) or ‘very difficult’ (four students). (The L1–L2 results were statistically confirmed at a significance level of 5% and this is enhanced by the fact that the English test items were more frequent than the Swedish ones.) As he was offered definitions of the test items and asked to fill in the right words in sentence slots, the format given to the native speaker did not, as pointed out before, really test his mastery of false friends. Still, the native speaker’s result shows that he has, as anticipated, a better grasp of the meanings of these words than the L2 learners, but not by much. The Swedish test, in which similar-sounding/similar-looking English words were juxtaposed with the Swedish test items, is not exactly comparable to the English test either. The students’ results on this test, however, show that they were less prone to be prompted by their L2 than their L1 when faced with phonologically and/or orthographically similar-looking words. Tables 6.17 and 6.18 offer the informants’ results for each test item, with Table 6.17 displaying the English test part and Table 6.18 the Swedish one. The L2 test items are presented according to the number of correct answers for each sentence pair, starting with the one that received the highest score. Since the Swedish test items were presented one by one, they Table 6.16 The students’ results on Part C: false friends Students
N
Correctness rate
Mean
SD
Standardised scores
Native speaker
1
55.00% (=11/20)
–
–
–
Swedish students –English test
15
44.67% (=134/300)
8.93
3.63
Highest: 1.94 Lowest: −1.36
Swedish students –Swedish test
15
65.33% (=196/300)
13.07
3.90
Highest: 1.52 Lowest: 2.07
C1 C5 C10 C6 C2 C4 C9 C8 C7 C3
No.
critic (14) cheek (12) conserve (14) novel (7) by nature (14) consequently (11) physician (6) curator (2) all but (2) genial (0) Total Correctness rate
(=kritiker) (=kind) (=bevara) (=roman) (=av naturen) (=följaktligen) (=läkare) (=(musei)intendent) (=nästan) (=vänlig(t))
English word given in context to be translated into Swedish (correct answers)
36 (4) 88.9%
7 7 2 6 (1) 4 3 (1) 5 (2) 1 1
KN 3 2 3 5 (3) 6 5 (1) 1 1 (1) 4 (3) 1 (1) 31 (9) 71.0%
TK
2 (2) 1 (1) 2 (2) 5 (5) 13 (11) 15.4%
1 1 4 1 14 (5) 64.3%
1 1 1 (1)
2 (1) 3 (2) 1 1 (1)
NKC
1
NKR
4 1 (1) 3 (1) 4 (1) 4 (2) 5 (4) 3 (3) 3 (3) 28 (15) 46.4%
RC
1 1 6 1 3 15 (1) 93.3%
2 1 (1)
NK
Word to be translated (given in context) from swedish into english (correct answers) kritik (11) (=criticism) haka(n) (12) (=(the) chin) konservera (4) (=preserve) novell (8) (=short story) nature(en) (1) (=scenery) konsekvent (4) (=consistently) fysiker (5) (=physicist) kurator(n) (4) (=counsellor) allt annat än (2) (=anything but) genial(isk) (1) (=brilliant)
2 4 (4) 1 1 1 2 (1) 2 (1) 20 (6) 70.0%
3 4
KN
9 (1) 7 (1) 5 (3) 4 (1) 7 (6) 7 (5) 4 (2) 4 (2) 6 (5) 6 (6) 59 (32) 45.8%
TK
2 1 8 (1) 5 (3) 2 5 (1) 8 (2) 9 (1) 5 3 48 (8) 83.3%
NK
Table 6.17 The students’ results on and evaluation of false friends on the English test (KN = item known, TK = item thought to be known, NK = not known, RC = recognised and context used, NKR = recognised but not known, NKC = not known, context used). Figures in parentheses in columns KN−NK indicate cases where the students were incorrect in their self-evaluation
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 221
C3 C5 C9 C10 C2 C11 C7 C14 C1 C17 C15 C4 C19 C20 C6 C13 C18 C12 C16 C8
Swedish word (correct answers) kompositör (15) (=composer) permission (15) (=leave of absence) valv (15) (=arch, vault) likör (14) (=liqueur) bransch (14) (=line of business) kreatur (13) (=cattle) kompanjon (13) (=partner) vrist (12) (=ankle) eventuellt (11) (=possibly) graciös (11) (=graceful) självmedveten (11) (=self-confident) allé (10) (=avenue (with trees)) sensibel (10) (=sensitive) banna (9) (=scold) provision (9) (=commission) plump (4) (=coarse, rude) pulpet (4) (=desk) harm (3) (=indignation) rekvirera (3) (=order, send for) spirituell (0) (=witty)
Possible english fal se friend used as a prompt word compositor permission valve liquor branch creature companion wrist eventually gracious self-conscious alley sensible ban provision plump pulpit harm require spiritual Total Correctness rate 3 (3) 89 (9) 89.9%
1
KN 8 7 6 8 6 5 4 4 (1) 7 (2) 5 (1) 3 8 4 4 (1) 3 (1) 3
TK 5 4 6 2 6 6 6 6 7 (2) 5 (1) 6 (2) 3 (1) 1 4 (1) 8 (2) 3 (2) 2 3 (2) 4 (2) 3 (3) 90 (18) 80.0%
1 3 1 3 3 7 3 1 27 (0) 100%
1 1 2 (1) 1 1 (1) 2 4 (2) 3 (1) 4 6 (1) 3 7 5 41 (6) 85.4%
NK
1 1 1 1 1
NKR 1
Table 6.18 The students’ results on and evaluation of the Swedish test (KN = item known, TK = item thought to be known, NKR = item recognised but not known, NK = item not known). Figures in parentheses in columns KN−NK indicate cases where the students were incorrect in their self-evaluation
222 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 223
are presented individually here too, again starting with the test item that the students knew best. (When two or more items received the same number of correct answers, the item for which the students showed the greatest accuracy in their self-evaluation is placed first.) As an example, the English word physician was translated correctly into Swedish by six of the 15 informants. Here the majority (five subjects, two of whom were wrong) indicated that they were certain of the meaning of the test item (column KN, coloured grey as they were in majority). Its Swedish false friend fysiker was translated correctly into English by five of the 15 students. Here the informants displayed considerably more uncertainty, with the majority (eight students, two of whom actually did know the correct word) indicating that they did not know the English equivalent (column NK, highlighted in grey as they were in the majority). When the students’ answers are analysed, two main reasons why they were unable to provide the correct answer crystallise. With the more frequent error type, the subjects were simply unfamiliar with the meaning of the English word offered and/or were unable to provide an English word for the Swedish false friend. Examples of this error type were seen with almost all of the test items. The other type, which was surprisingly common, involved answers that were induced by the phonological and/or orthographic similarity between the L1 and L2 items; that is, despite the fact that it had been explained to the students that they were going to be tested on false friends, they were heavily prompted by the words’ similarity. This was especially true for the English items consequently, genial, novel, physician, and all but, which the students very often assigned the meanings of the Swedish false friends, and for the Swedish words natur, novell, and konsekvent, whose orthographic/phonological counterparts in English were very often offered as answers. As pointed out in the theoretical background, being prompted by phonological/orthographic similarities is typical of non-advanced learners. As reflected in the informants’ low mean score (8.93 out of 20), the majority of the informants in the present study thus appear to be low achievers when it comes to false friends. On the Swedish test, on the other hand, the subjects did not, as pointed out above, appear to be affected by the English similar-looking/ similar-sounding words to the same extent. It is also interesting to note that the students’ receptive knowledge (i.e. the translation of the English terms (e.g. critic, cheek, conserve, etc.) into Swedish) is more developed than their productive knowledge (i.e. the translation of the Swedish words (e.g. kritik, haka(n), konservera, etc.) into English). 6.5 Individual L1 and L2 Knowledge
For five of the 15 students there is an L1–L2 correlation, Students 2, 5 and 14 consistently being among those students who received the top five
224 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
scores on all three test parts in both languages and Students 4 and 13 being among those who received the bottom five scores on all three parts in both languages. It is interesting to note that those learners achieving well also performed (comparatively) well on the three tests in Chapter 2 (focusing on vocabulary size), and those learners performing poorly also performed poorly on the same three tests. It appears that developing vocabulary depth relies (partly) on having acquired a sizeable number of vocabulary items. It is also worth pointing out that in Parts A and B there were Swedish students who received higher scores than the native speaker on the English test. In Schmitt and Dunham (1999), it was observed that L2 learners are often able to assess the frequency of near synonyms almost as well as native speakers. The positive effect of frequency was also detected in the present study, both within and between synonym sets. Considering the fact that the definitions of the near synonyms often offered vital information as to their commonality (e.g. which of the items is the superordinate), it is very likely that the subjects’ sensitivity to frequency enabled some of the more advanced students to outperform the relatively low-achieving native informant. The relative frequencies of the different senses of the polysemous words tested, which were shown to have an impact on the learners’ knowledge, may also have aided the more advanced Swedish students in doing better than the native speaker. 6.6 Conclusions and Pedagogical Implications
As discussed in the literature reviews in each of the three subfields investigated in the present chapter, creating links between words is a difficult and time-consuming task. This is substantiated even further in the present study by the fact that even university students display comparatively poor mastery of polysemous words, near synonyms and false friends. In order to promote network building, teachers therefore need to make use of all the pedagogical tools available. For words with multiple meanings and synonym sets, the most useful tool appears to be frequency; that is, when selecting items for the teaching syllabus in these areas, commonality definitely needs to be taken into account. For words with multiple meanings, it was shown that, while the frequency of the polysemous word itself had little or no relevance as to whether the meanings of the word were known or not, the relative frequencies of the different senses played a significant role, the most frequent meaning being by far the best known. Starting with the most common senses may therefore be a good point of departure, perhaps by implementing puns, as discussed by Söderström et al. (2022). Furthermore, although it could not be either substantiated or refuted in the present study, since the core sense was also the most frequent sense in the majority of cases, other research has shown that explaining the core sense of a polysemous
Polysemous Words, Lexical Fields of Near Synonyms and False Friends 225
word may help learners comprehend and retain other, more peripheral, meanings of the same word to a greater extent. This may therefore also be a fruitful stepping stone when approaching polysemous words from an L2 perspective. The New Oxford Dictionary of English (1998), which to the present author’s knowledge is the only dictionary that consistently lists the core sense first, may be of great help to English language teachers here. As for synsets, ‘few indeed are the suggestions in the ESL literature on how to impart sensitivity to the distinctions between closely related lexical items’ (Martin, 1984: 135). Moreover, while most researchers appear to agree that near synonyms should be presented in a contextualised form (e.g. Channell, 1981; MacKay, 1980; Shioji, 1982; Wallace Robinett, 1978), there is disagreement on the exact way to do this, with some proposing a semantic approach (e.g. Holmegaard, 2000; Nagy, 1989) and others advocating a thematic approach (e.g. Waring, 1997). In the present investigation, commonality again proved to have a positive effect on the students’ mastery. This applied to the relative frequencies of the near synonyms as well as the synonym sets as wholes. Commonality could therefore certainly be used to guide teachers in their selection of items to be part of the teaching syllabus. Considering the fact that the frequency of a near synonym often tallies with its degree of generality – that is, whether the term is a superordinate or a hyponym with increased specificity – teachers should start with the most frequent item (i.e. very often the superordinate) of the more frequent synonym sets and then, only after these have been acquired, move on to more infrequent items within the same synonym set (i.e. hyponyms), making the more specialised items of infrequent synsets the very last ones on the teaching syllabus. Finally, as for false friends, what surprised the present author the most was that quite a few of the participants relayed the information that they had not heard the term before, and that they could not recall having discussed such items in the upper secondary school classroom. Admittedly, these items are less prolific than polysemous words and near synonyms, but since ignorance of these items may cause serious misunderstandings, they too warrant time in the L2 classroom. 6.7 The Chapter That Follows
In Chapter 7, we will explore the 15 university students’ L1 and L2 inferencing skills, both by implementing parallel tests constructed specifically for this purpose and by exploring the degree to which inferencing was used in some of the tests discussed in previous chapters. Notes (1) Due to a construction error, polysemous word number 27 (needle) was accompanied by five correct answers for the native speaker, not four as for the Swedish learners.
226 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(2) Sometimes the term false cognate is used interchangeably with the term false friend. However, while false cognates are defined as similar-sounding/similar-looking words that are thought to have a common historical origin but actually do not, the term false friend can be used in a more general way, irrespective of whether there is a false historical commonality or not.
7 Inferencing
7.1 Introduction
In the research literature, it seems to be widely agreed that vocabulary in a second language should be learned/taught in a contextualised form. Studies have shown that being able to infer meanings of words based on their contextual clues, which very often occurs in combination with the help of one’s general linguistic knowledge and general knowledge of the world (Haastrup, 1991), is one of the three most important skills to possess for a learner aiming to enlarge his/her L2 vocabulary, explicit teaching and knowledge of affixation rules being the other two (Nation, 2001). The implementation of inferencing seems especially instrumental to native speakers who have already developed a comparatively rich vocabulary (Schmitt & McCarthy, 1997). Regrettably, Nation (2001) states, this does not seem to be equally true for second language learners, since the conditions required for successful L2 inferencing are not always present. One such important condition is that the density of unknown words should not be too high (Horst et al., 1998; Sternberg, 1987; Swanborn & de Glopper, 1999). Research has shown that in order for an L2 learner to be able to guess the meaning of an unknown word about 95% of the text needs to be understood (e.g. Bensoussan & Laufer, 1984; Laufer, 1989a; Laufer & Sim, 1985; Liu & Nation, 1985). This means that there is around one unfamiliar word in every 20 running words. Some researchers (e.g. Nation, 2001) even claim that 98% coverage is necessary for L2 learners to be successful inferencers. This corresponds to there being one unknown word in 50 running words. In addition to the density of unknown words, there are a number of other so-called mediating variables (Jenkins & Dixon, 1983) that can strengthen or weaken a learner’s chances of guessing the meaning of a contextualised word (Nation, 2001; Webb, 2008), such as the degree of similarity between the first language and second language (which will be explored further in the following section) (Palmberg, 1988), the degree of importance of the unfamiliar word for the understanding of the text as a whole (Kim, 2003; Nation, 2001; Sternberg, 1987), the number of occurrences of the unfamiliar word (Horst et al., 1998; Stahl & Fairbanks, 1986; Sternberg, 1987) and the proximity of recurrence (Nation, 2001), whether the unfamiliar word occurs in many different contexts (Nation, 227
228 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
2001; Sternberg, 1987), whether the unfamiliar word occurs in texts whose topics are familiar to the learner (Nation, 2001), and whether the concept expressed by the unfamiliar word is known to the learner (Daneman & Green, 1986; Nagy et al., 1987; Shefelbine, 1990). Typographical aids such as the use of italics, quotation marks, bolding, and figures and diagrams have also been shown to help the inferencing process (Artley, 1943). That being said, the most useful information appears to come from clues in the immediate context (Chihara et al., 1977; Paribakht & Tréville, 2007; Paribakht & Wesche, 2006; Rye, 1985). Not only is the presence of clues important for successful inferencing (Nation, 2001), but the number of relevant clues (Nation, 2001) and their explicitness (Carnine et al., 1984) are as well. If, in addition, the ideas in the clues are familiar to the learner (Jenkins & Dixon, 1983) and concrete rather than abstract (Nation, 2001), the inferencing process is even more likely to be successful. Nation (2001: 262) concludes adamantly that in any list of vocabulary learning strategies, guessing from context would have to come at the top of the list. Although it has the disadvantages of being a form of incidental learning (and therefore being less certain) and of not always being successful (because of lack of clues), it is still the most important way that language users can increase their vocabulary. 7.2 Theoretical Background and Previous Research L2 Inferencing strategies
In a study by Qian (2004), inferencing techniques used by young adult L2 learners were looked into. The investigation included 61 Chinese and Korean students of English (with an average age of 24) who participated in intensive intermediate/high-intermediate ESL programmes at two Canadian universities. In the initial stages of the study, all of the subjects were asked to fill out a questionnaire in which they were asked to rank the frequency with which they thought they used different inferencing strategies and what kind of information they thought they made use of when guessing the meaning of contextualised words. Twelve of the 61 informants were then randomly picked and subjected to a test, during which they were asked to guess the meaning of previously unknown contextualised words. The test consisted of a one-page text in which 10 words, accounting for about 7% of the total number of words, were chosen as test items. Based on this test and a number of interviews, Qian was able to draw conclusions about what inferencing techniques learners actually utilise when reading. With the help of the subjects’ answers in the questionnaire, Qian was also able to address three additional research questions. Firstly, what resources do L2 learners think they make use of when they are faced with words that are unknown to them in a written text? Secondly, to what degree do L2
Inferencing 229
learners think they use different inferencing techniques? Thirdly, do the preferred inferencing techniques reported by the learners tally with the ones they actually use? The vast majority of the learners in Qian’s study, 62.3%, indicated that they often used a word’s context to guess its meaning. Another 34.4% indicated that they sometimes did so. This means that guessing words’ meanings from context was by far the most common tool used of all the ones considered. This result thus strengthens Nation’s view, expressed in Section 7.1, that contextual clues are vital in the inferencing process. Contextual inferencing was followed by looking up the word in a dictionary, which was also found to be a common strategy in Schmitt (1997), in which 600 Japanese learners of English were investigated. Still another 23.0% of the subjects often tried to see if the morphological form of the word could give some clues about its meaning, and as many as 49.2% sometimes did this. The strategies reported to be employed the least were asking a friend or the teacher for assistance. Based on the results on the second part of the questionnaire, in which the informants were asked what kind of information they thought they made use of when guessing the meaning of a contextualised word, Qian drew the conclusion that the informants’ inferencing strategies were mainly top-down, focusing on global meaning (the most common strategy) and making use of their knowledge of the world (the second most common strategy). Qian then made a comparison between the subjects’ perceived strategies and the strategies they actually used. Global meaning and world knowledge, which were perceived by the students as important strategies, were not the ones they really employed, with making use of the meanings of other words in the same sentence instead seeming to be the most frequently used strategy. Also, when asked about their perceived strategies, using morphological clues was not a technique the learners appeared to prioritise. However, as mentioned above, the test and the interviews actually showed that this was the second most common technique implemented. This means that the top-down strategies that the students believed they used (global meaning and world knowledge) were not what they relied on when faced with an actual text, the syntagmatic and morphological cues favoured being evidence of the opposite. Qian posits that there may be several reasons for this discrepancy. For instance, the teachers, who were observed to repeatedly tell the learners to use top-down techniques because of their superiority to other types of strategies, seemed to have convinced the subjects that they made more use of global meaning and their knowledge of the world than they actually did. Qian also believes that the discrepancy may have been due to the structure of the text itself. For example, for some of the words tested, all the strategies could not be applied to an equal extent. Other investigations have come up with results that contradict the ones presented in Qian. In Harley and Hart (2000), for instance, which
230 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
focuses on secondary school learners of French, asking the teacher or a friend for help were strategies that were perceived as favoured by the informants. Discrepancies like this may indicate that cultural background and the age of the learners play major roles in what strategies are implemented. This is shown in Schmitt (1997), in which strategies that were beneficial to certain age groups were considerably less so to other age groups. Moreover, it could be shown that learners usually mature into using different techniques. This means, Schmitt concludes, that ‘we must take our learners’ cognitive maturity and language proficiency into account when recommending strategies’ (1997: 225–226). In Gu and Johnson (1996), investigating the strategies of advanced L2 students, five different types of learners could be discerned. Firstly, there were the readers. They formed the most advanced group based on vocabulary size and general proficiency and put a lot of focus on learning through reading. They actively sought words they believed to be important and used a lot of contextual clues. Secondly, there were the active strategy users who achieved the second best results on the vocabulary size and proficiency tests. These students used a more varied sample of techniques than any of the other groups. The readers and the active strategy users made up about 11% of all the informants included in the study. Thirdly and fourthly, there were the encoders and the non-encoders respectively, accounting for 87% of all the learners. The main difference between these two groups is that the former displayed more use of memorisation techniques, such as association, imagery, visualisation of the form of a word, and breaking the words into stems and affixes. Lastly, there were the passive strategy users, who made up only 2% of the subjects. These were by far the least successful learners, using the least number of strategies. Although a substantial amount of research has focused on L2 vocabulary learning strategies, there is, as hinted at above, a great need for more studies focusing on variables like age, cognitive maturity, cultural background and level of proficiency. L1 and L2 inferencing compared
Firstly, as alluded to in the previous section, it needs to be pointed out that, even though there are commonalities across languages as to how learners approach unknown words in context (Wesche & Paribakht, 2009), research indicates that the typological distance between a student’s L1 and L2 may determine how a learner is able to approach an L2 inferencing situation and may consequently be an important factor in whether the learner is successful or not. This is one of the conclusions drawn in a large-scale study by Wesche and Paribakht (2009). Focusing, among other things, on the inferencing success of native speakers of French and Persian learning English as a second language, the researchers were able to show that the former, having a first language more closely related to English
Inferencing 231
than the latter, achieved considerably better inferencing scores. The impact of learners’ L1 has also been reported in several other studies (e.g. Koda, 2005; Odlin, 2003; Paribakht, 2005; Ringbom, 1987). Thus one highly relevant investigation to the present study is Albrechtsen et al. (2008), whose informants’ L1 is Danish, a language closely related to Swedish. In addition to making a comparison between learners’ L1 and L2 (English) inferencing skills, the study also focuses on how these inferencing skills relate to the learners’ mastery of L1 and L2 vocabulary size, vocabulary depth (in the form of network building), and reading skills. The informants were from three different educational levels: 30 students were considered to be beginners of English (having studied English for about three years only), 30 students were categorised as intermediate learners (having studied English for about six years), and 30 students taking English as a single subject course at university level were considered to form the most advanced group of informants (having studied English for at least nine years, making them comparable to the learners in the present study). The main difference between these three student groups, according to the researchers, is that while the members of the youngest group were still in the process of acquiring L1 literacy skills, the students making up the intermediate group had had some time to refine these skills, and the university undergraduates, having entered the world of academia, had most likely already acquired very elaborate L1 literacy skills. Two aspects of declarative knowledge (‘knowing that’) – vocabulary size and network organisation – and two aspects of procedural knowledge (‘knowing how’) – inferencing strategies and writing – were in focus. The reason for including tests on both types is, as explained by Albrechtsen et al. (2008), that they are interconnected. For example, research has shown that students’ vocabulary size appears to go hand in hand with their ability to draw conclusions about the meanings of unknown words in context (see also Nation, 2001). This means, Albrechtsen et al. (2008: 24) claim, that ‘deficiencies in declarative knowledge in the foreign language often prevent learners from transferring their procedural potentials to demanding L2 communicative situations’, such as lexical inferencing. Put differently, ‘[L]earners below a certain threshold of L2 vocabulary knowledge are unable to transfer the higher order skills they may have developed in their L1 to L2 tasks’ (Albrechtsen et al., 2008: 24). Furthermore, ‘[N]ot only do language learners need a vocabulary of a certain size, but also the organization of their lexicon must be of a certain quality to ensure efficient language use’ (Albrechtsen et al., 2008: 24). This is confirmed by Nassaji (2004), for instance, who found a clear correlation between vocabulary depth and inferencing success among the L2 learners included in his study. In the inferencing part of Albrechtsen et al.’s (2008) investigation, three main aims, mimicking the major aims of the entire study, were put forth.
232 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Firstly, the researchers wanted to find out if different inferencing strategies were used in the students’ L1 as compared to their L2, and if there were noticeable differences between the three educational levels. Secondly, the researchers wanted to investigate whether the level of inferencing success differed in the two languages and, again, whether educational level played a role. Lastly, Albrechtsen et al. (2008) wanted to explore whether a correlation existed between lexical inferencing success, L2 reading skills, and quantitative and qualitative vocabulary knowledge. The inferencing test itself consisted of a reading comprehension task. This task was based on short factual texts with similar topics in the two languages. In both the students’ L1 and L2, 30 items were tested. These were of three kinds: (1) items for which there were no linguistic clues as to their meaning, (2) items for which there were potential morphological clues in the form of affixes, and (3) items for which there was at least one central clue in the form of a word or a word stem and a prefix and/or a suffix. The data collection consisted of three parts. First there was a pre-test to determine which of the test items were already known to the informants. The words were here presented to the students in a decontextualised form. Next came the inferencing task itself, on which the students were asked to verbalise their thoughts while inferencing. Lastly, the informants were faced with a retrospective task, in which they were asked to state clearly what had helped them arrive at their answers. The clues drawn on by the students when inferencing were divided into three categories: contextual, intralingual and interlingual. Contextual clues are clues that come from the text surrounding the test item (either from the narrow or broad context) or from the student’s knowledge of the world. These two types of clues often work together. Intralingual clues are clues that come from the test item itself, while interlingual clues come from the student’s knowledge of other languages. These clues can be put along a continuum from a more top-level processing approach (advanced processing) to a more bottom-level processing approach: context (the text and knowledge of the world), semantics (meaning considerations), lexis (word form, morphology and orthography/phonology) (Albrechtsen et al., 2008: 80). The results of the study were then analysed in terms of how advanced the learners’ processing was, whether adaptation to word type (the three categories mentioned above) occurred, and the degree of inferencing success. Advanced processing – that is, the use of top-ruled processing with integration of linguistic clues – was far more prominent in the subjects’ L1 than their L2, and this held true for all three student groups, with a steady increase according to educational level. The same results could also be seen in the way the informants were able to adapt their inferencing strategies to the situation; that is, in their L1 the students did not only display a wider range of processing types but also applied the strategies in the appropriate places to a greater extent than in their L2. Again there was
Inferencing 233
also an increase according to educational level. As for inferencing success, once again the same result pattern could be discerned; that is, there was a higher success rate in the mother tongue than in the second language, as well as a clear increase from one educational level to the next. For the university students, who are of special interest to the present investigation, there was a success rate of 48% in their L2 as compared to 58% in their L1. It thus seems, Albrechtsen et al. (2008) claim, that processing skills acquired and made use of in a native language are not completely transferred into a learner’s L2, not even at as advanced a level as university. Based on their results, Albrechtsen et al. (2008) were also able to give clear descriptions of the typical mature, intermediate, and immature inferencer. The mature inferencer displays frequent use of advanced processing and very little use of bottom-ruled processing, and adapts easily to the inferencing situation. The intermediate inferencer also resorts to advanced processing, but only about half as much as the mature inferencer. Moreover, although the intermediate inferencer often activates linguistic clues, these clues are not really integrated into the processing. The ability to adapt to the inferencing situation is also lower for the intermediate inferencer than for the more mature inferencer. The immature inferencer, finally, generally displays more bottom-ruled processing than the intermediate inferencer; that is, low achievers, in contrast to high achievers, seem to focus on formal features. An immature inferencer is also most typically even less able to adapt to the inferencing situation than an intermediate inferencer. It thus seems, according to Albrechtsen et al. (2008), that learners generally move from a bottom-ruled approach to a more top-ruled approach. Finally, as predicted, Albrechtsen et al. (2008) were able to show that the success or failure with which a learner is able to infer the meaning of a word relies heavily on the learner’s L2 vocabulary size (see also Shefelbine, 1990). As for vocabulary depth, a clear correlation could only be observed for the lowest educational group, but this occurred in these informants’ L1 as well as their L2. The results of other investigations have, however, been able to demonstrate such correlations at other educational levels too (e.g. Laufer, 1997b; Paribakht, 2005). Also, just like Albrechtsen et al. (2008), other studies have shown correlations between advanced students’ reading skills and their ability to infer meanings of contextualised words (e.g. Herman et al., 1987). 7.3 The Parallel Tests Used for the Present Chapter
The L1 and L2 tests aimed at investigating the 15 informants’ inferencing skills in the present study were both divided into two main parts. In one part of the English test, consisting of four texts taken from Newsweek (Texts B (Huus, 1993) and F (unknown author, 1993)) and
234 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Time magazine (Texts A (Gibbs, 2008) and E (Jaroff, 1991)), the students were given the Swedish translations of the English words/expressions sought (the native speaker was here offered definitions in English) and were asked to find the corresponding English items in the text offered. These Swedish translations were given in chronological order so that if the first and third words/expressions were detected, the student would know that the second item could be found between these two. This is exemplified with an excerpt from Text A, in which the learners were instructed to find the English words for skillnaden (the disparity) (test item A8), slingrade sig igenom (wiggled through) (test item A9), and kvinnohatare (misogynists) (test item A10): While educators debate whether there is a ‘boy crisis’ that warrants a wholesale change in how to teach, colleges are quietly stripping the pastels from brochures and launching Xbox tournaments to try to close the gap in the quality and quantity of boys applying. ‘It’s a gross generalization that slacker boys get in over high-performing girls,’ says Jennifer Delahunty, dean of admissions at Kenyon College, ‘but developmentally, girls bring more to the table than boys, and the disparity has gotten greater in recent years’. Of course, admitting this is taboo, as Delahunty learned two years ago. She was in marathon committee meetings, stacking glorious girls on the waiting list while less accomplished boys wiggled through, when she got an e-mail informing her that her own daughter had been wait-listed. The experience inspired her to write a confessional Op-Ed, ‘To All the Girls I’ve Rejected,’ for the New York Times, responses to which lit up her inbox. ‘It pissed off the feminists and the misogynists. I got both sides of the spectrum,’ she told me. ‘The misogynists said women already have too many advantages. And the feminists said, How dare you not treat women like men? But what most amazed her was the reaction of young women: by and large, they assumed this is just how things work. ‘Why aren’t they marching in the streets? That’s the part that slays me,’ Delahunty says. ‘It isn’t fair, and young women should be saying something about it not being fair.’
The Swedish texts A, B, E and F were chosen from Cederholm and Danell (2007), which includes a collection of extracts from Swedish magazines, newspapers, novels, short stories, and so on whose main purpose is to teach difficult words. Here definitions of the Swedish words sought were offered. The English texts for this part were chosen first. These are texts that the present author has worked with in many student groups before in courses focusing on reading proficiency and vocabulary. The items finally selected for testing were on the one hand words/expressions that are infrequent and consequently most often found not to be known by students, and on the other hand words/expressions whose context provided clues as to their meanings (either in the immediate proximity or in the larger context or both).
Inferencing 235
Table 7.1 The titles and topics of texts A, B, E, and F included in the present study Text A
Text B
Text E
Text F
English texts
Affirmative action for boys
Call of the wild (preservation of endangered species)
It happens in the best circles (hoaxers creating crop circles)
Mayhem in Miami (tourists in danger in Miami)
Swedish texts
No title (people suffering from a compulsive disorder)
No title (differences and similarities between chimps and humans)
No title (UFO stories)
No title (the Swedish psyche)
The aim for the Swedish texts was to find similar topics to those selected for the English test part, but this proved difficult as the test items themselves as well as the clues offered had to be prioritised. Table 7.1 offers titles/descriptions of the texts finally picked. Moreover, the 44 words selected to be tested were, in both languages, picked randomly. (Their distribution as well as the number of words in each text can be seen in Table 7.2.) However, as the participants in the present study are considered advanced students, no high-frequency words were selected. This is, for instance, evidenced by the fact that none of them were found on West’s (1953) general service list. Furthermore, in order to be able to compare the test results in a fair way, the total frequencies of the words tested were contrasted. Based on the two corpora implemented, the Swedish test items were shown to be slightly more infrequent (constituting ≈0.0020% of Språkbanken) than the English ones (making up ≈0.0041% of the BNC), rendering the L1 test part, from a frequency perspective, more difficult than the English one. The L1–L2 difference in frequency was, however, not confirmed statistically, which means that comparisons may be made between the parallel test parts. To gain explicit knowledge about when inferencing had taken place, the students were also asked to provide information about the words/ expressions, as exemplified in (1). Table 7.2 The number of words sought and text size for each of the four texts making up one part of the inferencing test No. of words sought/no. of words in the text Text A
Text B
Text E
Text F
A + B + E + F
English texts
12/788
12/1257
11/901
9/1190
44/4136
Swedish texts
12/1713
8/290
15/1142
9/314
44/3459
236 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(1) klaga = __________________________________ A □ I can’t find the English word. B □ I think I have the right English word, but I haven’t seen the word before. C □ I am sure I have the right English word, but I haven’t seen the word before. D □ I recognise the English word and I think it is the right word. E □ I recognise the English word and I am sure I have the right word. F □ I know the English word and I am sure I have the right word.
Only answers B and C will be discussed in more detail in the results section (Section 7.4), since they are considered to involve true inferencing situations. There were two main reasons for including this kind of test format. Firstly, since, as discussed in Section 7.2, good reading skills appear to be a prerequisite for good inferencing skills, the present author wanted to use a testing method that would obligate the students to read the entire text, thus forcing them to take in all contextual clues available. The hypothesis is that this test type will consequently produce a large number of correct answers. Secondly, since it is difficult to construct inferencing tests that can be used for both L1 and L2 learners due to the superiority of native speakers as to vocabulary size, the present author wanted to use a testing method that would tap the L1 learner’s knowledge in a way that would make it slightly less possible for him to simply provide meanings for words already known to him. In the other part, the informant group as well as the native speaker were given a more traditional inferencing test in which they were asked to explain the meanings of words indicated in bold. The English texts (Texts C and D) used for this part consist of two short stories: The Bully by James Reaney (New, 1997), which tested 20 items, and To Room Nineteen by Doris Lessing (Övergaard et al., 1992), which included 36 items. The Swedish test part also consists of two texts. Whereas one of them (Text C), taken from Cederholm and Danell (2007), is a conversational article, testing 24 items, the other text (Text D), incorporating 32 items, is a Swedish translation of the beginning of the novel Perfume: The Story of a Murderer by Patrick Süskind (1987, trans. U. Wallenström). Both texts and test items were selected in the same manner as described for Texts A, B, E, and F. In this part of the test too, to be able to make fair comparisons, the total frequencies of the items were considered. Again, the Swedish test words/expressions were slightly more infrequent (constituting ≈0.0021% of Språkbanken) than the English ones (making up ≈0.0052% of the BNC), even more so than is the case with Texts A, B, E and F. The L1–L2 difference in frequency was, however, not confirmed statistically, which means that comparisons may be made between the parallel test parts.
Inferencing 237
Table 7.3 The number of test items and text size for each of the two texts making up the other part of the inferencing test No. of test items/no. of words in the text Text C
Text D
C + D
English text
20/4892
36/14,071
56/18,963
Swedish text
24/1960
32/5012
56/6972
Furthermore, while the total number of test items on this test part is the same for the two languages (56 items), the text mass for Texts C and D differs greatly between languages, meaning the students had considerably more text to absorb in their L2 than their L1 (as shown in Table 7.3). It is difficult to determine the exact effects of having different text sizes. On the one hand, it may be helpful with a longer text since it is then easier to get a feel for the story and thus perhaps easier to draw conclusions about the meanings of words. On the other hand, reading a long text in an L2 can be exhausting and may result in students giving up. Again, as with Texts A, B, E and F, to gain explicit knowledge about when inferencing had taken place, the informants were asked to provide information about the meanings they had offered. This is exemplified in (2), taken from Text C in the English material. (2) Every Saturday night we children all took turns bathing in the dish-pan and on Sundays, after Sunday-school, we would all sit out on the lawn and drink the lemonade that my father would make in a big glass pitcher. The lemonade was always slightly green and sour like the moon when it’s high up in a summer sky. A □ B □ C □ D □ E □ F □
I don’t know this word. I am guessing the word’s meaning from the context. I recognise this word, but I don’t know what it means. I recognise this word and I’m guessing its meaning from the context. I recognise this word and I think I know what it means. I am sure I know what the word means.
If you used the context to figure out the meaning of the word in question, what was it in the text that helped you?
As can be seen from the above, in this part of the inferencing test the students were also asked to supply information about what it was in the context, if anything, that had helped them arrive at the meaning offered.
238 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
The main focus in the results section (Section 7.4) will be option B, since this is considered to involve a true inferencing situation. The D option will also be considered, representing a weaker form of inferencing. On both tests, the texts occurred in alphabetical order, starting with Text A and ending with Text F. This means that the more conventional inferencing part made up of Texts C and D occurred in the middle of the test, with Texts A and B preceding it and Texts E and F appearing last. As described in Chapter 1, the whole English test was taken before the one testing Swedish words/expressions, and the students were allowed to sit with each test as long as they wanted to; no time constraints were put on either test. (All the test items, L1s as well as L2s, can be found in Appendix 4.) The participants were also asked to evaluate the test/test parts, choosing from the options ‘very easy’, ‘easy’, ‘average’, ‘difficult’ and ‘very difficult’. 7.4 Results and Discussion
In Table 7.4, the distribution of the students’ answers on Texts A, B, E and F is presented. The reader is reminded that these were the texts in which the informants were asked to find words for which Swedish translations (L2 test) or Swedish definitions (L1 test) were offered in chronological order. It can be seen here that more attempts at making inferences were made in the students’ second language (29.70%: 21.06% successful; 8.64% unsuccessful) than their mother tongue (12.12%: 8.79% successful; 3.33% unsuccessful). The lower percentage of inferences in Swedish can mainly be ascribed to the fact that a great many more of the L1 test items were already known (compare 51.97% on the Swedish test to 27.12% on the English test). Table 7.5 presents a more detailed analysis of the inferences made by the students in Texts A, B, E and F. As can be seen, the informants display similar high success rates in both languages (English: 70.92%; Swedish: 72.50%) (and consequently similar failure rates (English: 29.08%; Swedish: 27.50%)), with the learners’ L1 having a slight advantage over their L2, though no statistical Table 7.4 The distribution of answers on Texts A, B, E, and F Out of 660 tokens
No info offered about test item
Indicated as known
Indicated as not found
Items inferred
Correct info given
Incorrect info given
Correct info given
Incorrect info given
Successful
Not successful
English test
19.55%
27.12%
8.03%
14.24%
1.36%
21.06%
8.64%
Swedish test
17.27%
51.97%
7.12%
11.21%
0.30%
8.79%
3.33%
Inferencing 239
Table 7.5 The students’ inferencing results on Texts A, B, E, and F Successful
Not successful
%
Mean
SD
Standardised scores
%
Mean
SD
Standardised scores
English test
70.92% (=139/196)
9.27
7.68
Highest: 2.18 Lowest: −1.21
29.08% (=57/196)
3.80
5.68
Highest: 2.67 Lowest: −0.67
Swedish test
72.50% (=58/80)
3.87
4.37
Highest: 2.32 Lowest: −0.88
27.50% (=22/80)
1.47
2.97
Highest: 2.53 Lowest: −0.49
difference was obtained. (Nor could any statistical L1–L2 difference be obtained for those cases where the learners failed in the inferencing process.) The difference between the standard deviations confirms the more difficult nature of finding the correct item in a second language than a first. Compare 7.68 (L2) with 4.37 (L1) for successful inferences. The students’ results can also be contrasted with the result of the native speaker, who was only able to find 20 of the 44 words sought (=45.45%). In the 24 cases he did not score any points, there was an even distribution of words he simply could not find, words taken from the texts that were incorrect and words that were not in the texts but would have worked otherwise. Also, he clearly overestimated his ability to find the words, stating in his self- evaluation that he thought these parts were only average in difficulty. Table 7.6 shows the distribution of the students’ answers in relation to Texts C and D. These were the texts that represented a more traditional inferencing test, in which the students were asked to give the meanings of words indicated in bold. As with Texts A, B, E and F, the informants here made more inferences in their L2 (30.36%: 15.24% accurate; 15.12% inaccurate) than their L1 (20.95%: 9.05% accurate; 11.90% inaccurate). Again, the L1–L2 difference can mainly be attributed to the higher percentage of already acquired items in the learners’ first language (31.43%) versus their second (17.38%). Another reason may be that the students made more use of a weaker form of inferencing in their L1 (7.14%) than their L2 (5.12%), indicating that they recognised the item in question but still made use of its context in order to ascertain its correct meaning. Table 7.7 focuses entirely on the inferences made by the students in Texts C and D. Only the ones where the informants indicated that the test item was unknown to them – the stronger form of inferencing mentioned above – will be discussed. The findings show that the students made more successful inferences in their L2 (50.20%) than their L1 (43.18%). However, the difference could not be confirmed statistically. (Nor could the L1–L2 difference be statistically confirmed for those cases where the students failed in the inferencing process.) Nevertheless, the differences between the standard deviations underscore how difficult inferencing can be in a second language for some students. Compare 8.53 (L2) with 5.07 (L1) for successful inferences.
8.93%
8.57%
English test
No info offered about test item
Swedish test
Out of 840 tokens
17.38%
31.43%
3.69% 7.26%
27.02% 18.33%
2.62%
2.86%
Incorrect info given
Correct info given
Correct info given
Incorrect info given
Indicated as not known
Indicated as known
Table 7.6 The distribution of answers on Texts C and D
7.14%
5.12%
Successful
3.69%
4.64%
Not successful
Recognised, but still inferred
9.05%
15.24%
Successful
Items inferred
11.90%
15.12%
Not successful
240 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Inferencing 241
Table 7.7 The students’ inferencing results on Texts C and D Successful
Not successful
%
Mean
SD
Standardised scores
%
Mean
SD
Standardised scores
English test
50.20% (=128/255)
8.53
5.68
Highest: 1.49 Lowest: −1.50
49.80% (=127/255)
8.47
6.71
Highest: 2.47 Lowest: −1.26
Swedish test
43.18% (=76/176)
5.07
4.37
Highest: 2.05 Lowest: −1.16
56.82% (=100/176)
6.67
7.08
Highest: 3.01 Lowest: −0.94
The above means that, in both languages, the informants achieved higher scores on the more unorthodox test than the more traditional one. It thus seems that introducing an inferencing test in which students are forced to read the entire text, making use of all the clues available, may be a more fruitful way of initiating lessons on inferencing in the L2 classroom than implementing a more traditional test. In fact, the differences between the results of the two test types were confirmed statistically at a significance level of 5% in both languages. Furthermore, the results in Table 7.7 can be compared to the results seen in the Albrechtsen et al. (2008) study discussed previously, in which the university students performed considerably better in their L1 (58%) than their L2 (48%). There may be several reasons why the informants in the present investigation did not achieve higher scores in their mother tongue than in their second language. Firstly, as shown by the comparatively high percentage of items indicated to be known but which in reality were not known (7.26%; see Table 7.6), some students may have been overly confident about their L1 knowledge and might therefore instead have benefitted from trying to infer the meanings of some of these words. The reader is also reminded that the informants worked with the whole English test first, and only when they had completed this and handed it in did they move on to the Swedish test. This means that the students were probably rather tired once they started working on the L1 test. After the testing, some of the students also reported that they had worked with all of the texts making up the more unconventional test type first (Texts A, B, E and F), saving the longer texts for last. Students working in this way were probably even more exhausted once they got to the more conventional inferencing part of the Swedish test. Lastly, a few students reported that they had generally put more effort into the L2 test than the one testing their L1 since they, in their own words, ‘were studying English, not Swedish’. Lastly, it needs to be pointed out that, although great effort was made to use L1 and L2 texts that were similar in content and difficulty, these factors could not be controlled fully and may thus also have affected the results. It is also interesting to consider individual achievements on the traditional test part. Incorporating the weaker form of inferencing, Students 1,
242 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
5, 9 and 14 were among the five most successful in both languages. Two of these learners – Students 9 and 14 – were also among those who received top scores in both languages, disregarding whether inferencing occurred or not. It thus appears that, as both learners made extensive use of inferencing (especially Student 14), successful inferencing was the primary reason these two students achieved such good results. Students 2, 3, 4, 8 and 10, on the other hand, made little use of inferencing in either language. For Students 2 and 3, the reason for this dearth of inferences appears to be their already acquired vocabulary knowledge, with Student 2 achieving the second highest score in both her L1 and L2 and Student 3 also being among the top five students in both languages. Furthermore, while Student 8 appears to be an intermediate learner as to already acquired vocabulary knowledge, Students 4 and 10 were seen to be low achievers, and none of them were helped to any great extent by their meagre use of inferencing. The reader is here reminded that in order to create an optimal situation for inferencing, between 95% (Liu & Nation, 1985) and 98% (Nation, 2001) of the words in a text need to be known (see Section 7.1). This may of course not be the case in the present investigation where at least some of the learners probably felt the texts to be too difficult, which in turn means that these students were not really given the opportunity to show their full inferencing potential. Nevertheless, based on the results, it must be concluded that for the learners included in the present investigation making use of contextual clues is an integral part of making a text and its unknown words comprehensible, in an L1 as well as an L2. Lastly, as discussed previously, the learners were asked to describe what it was that had helped them figure out the meanings of the correctly inferred items. In general, even though they were prompted to do so, the students offered few explanations. On the English test, however, there was a slight tendency for the more successful inferencers to attempt to use clues other than those in the immediate vicinity of the unknown word. As discussed in Section 7.2, this was also seen in Albrechtsen et al. (2008). No conclusions can be drawn in this respect concerning the Swedish test items, since the participants here offered even fewer explanations of their inferences. 7.5 Inferencing in Relation to Vocabulary Size as Achieved on Some of the Previous Parallel Tests Used in the Present Study
Based on the students’ self-evaluations, it was observed on several of the eight administered tests implemented in the present study that, in order to be able to offer the correct answers, the learners were relatively often forced to infer the items’ meanings from context. The present section aims to investigate to what extent inferencing helped the students with their achievements on some of these tests, namely vocabulary taught at upper secondary school level, specialised uses of vocabulary, advanced vocabulary, and idioms. In Tables 7.8 (the English tests) and 7.9 (the
No. of infere nces (incor rect infere nces) Corre ctness rate %
66.3 %
52.1 %
54.5 %
57.0 %
RC 199 (67) (=33. 7%)
48.5 %
NKC 334 (172) (=51. 5%)
55.2 %
Total 533 (239) (=44. 8%)
Specialised uses of vocabulary (100 items)
Vocabulary taught at upper secondary school level (50 items) RC NKC TOT 235 121 114 (107) (58) (49) (=43. (=47. (=45. 9%) 0%) 5%)
48.8 %
RC 43 (22) (=51. 2%)
39.0 %
NKC 707 (431) (=61. 0%)
Advanced vocabulary (100 items)
39.6 %
Total 750 (453) (=60. 4%)
66.4 %
RC 140 (47) (=33. 6%)
55.3 %
NKC 479 (214) (=44. 7%)
Idioms (80 items)
57.8 %
Total 619 (261) (=42. 2%)
62.7 %
RC 496 (185) (=37. 3%)
46.7 %
NKC 1,64 1 (875) (=53. 3%)
Total (330 items)
50.4 %
Total 2,13 7 (1060) (=49. 6%)
Table 7.8 T he students’ inferencing with L2 vocabulary taught at upper secondary school level, specialised uses of vocabulary, advanced vocabulary, and idioms (RC = recognise the word but still guessing based on the context, NKC = its meaning is not known but guessing based on the context)
Inferencing 243
No. of inferences (incorrect inferences) Correctness rate %
55.7%
45.0%
53.1%
Vocabulary taught at upper secondary school level (50 items) RC NKC Total 61 (27) 20 (11) 81 (38) (=44.3%) (=55.0%) (=46.9%)
78.1%
RC 192 (42) (=21.9%) 70.4%
NKC 81 (24) (=29.6%) 75.8%
Total 273 (66) (=24.2%)
Specialised uses of vocabulary (100 items)
63.4%
RC 232 (85) (=36.6%) 45.9%
NKC 222 (120) (=54.1%)
Advanced vocabulary (100 items)
54.8%
Total 454 (205) (=45.2%) 79.6%
RC 113 (23) (=20.4%)
77.0%
NKC 74 (17) (=23.0%)
Idioms (80 items)
78.6%
Total 187 (40) (=21.4%)
70.4%
RC 598 (177) (=29.6%)
56.7%
NKC 397 (172) (=43.3%)
Total (330 items)
64.9%
Total 995 (349) (=35.1%)
Table 7.9 T he students’ inferencing with L1 vocabulary taught at upper secondary school level, specialised uses of vocabulary, advanced vocabulary, and idioms (RC = recognise the word but still guessing based on the context, NKC = its meaning is not known but guessing based on the context)
244 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Inferencing 245
Swedish tests), the extent of the learners’ use of inferencing and success rates on these tests are presented. Both the stronger form (‘I am guessing the word’s/expression’s meaning from the context’) and weaker form (‘I recognise this word/expression and I’m guessing its meaning from the context’) of inferencing are considered. With the exception of advanced vocabulary, where the students only display a success rate of 39.6%, the results on the other L2 tests are all above 50%, ranging from 54.5% successful inferencing with vocabulary taught at upper secondary school level to 57.8% with idioms, with specialised uses having a success rate in between of 55.2%. This gives an average success rate of 50.4% for the English test items. This can be compared to the informants’ inferencing in their first language, where the success rates range from 53.1% for vocabulary taught at upper secondary school level to as high as 78.6% for idiomatic expressions, with advanced vocabulary and specialised uses displaying success rates in between 54.8% and 75.8% respectively. For the Swedish test items, this then gives an average success rate of 64.9%, which is 14.5 percentage points higher than in the subjects’ L2. This difference in achievement, confirmed statistically at a significance level of 5%, becomes even more interesting when contrasted with the actual number of inferences made. While in their L2 the students made 2137 inferences, 1077 of which were successful, in their L1 they only made a total number of 995 inferences, 646 of which were successful, the lower number of inferences in their native language of course mainly being due to the learners’ already acquired knowledge of a vast number of L1 items. Comparisons can here be made with Albrechtsen et al.’s (2008) study (see Section 7.2), in which the university students’ results, displaying success rates of 58% and 48% in their L1 and L2 respectively, are remarkably similar to the ones in the present investigation (64.9% and 50.4% respectively). Again, the conclusion to be drawn is that the learners in the present study, just like the students in Albrechtsen et al.’s study, made extensive use of inferencing in order to achieve what they did. This is, as already discussed above, of course more true for some learners than others. Students 5 and 14, for example, successfully inferred a great number of word meanings in their L1 as well as their L2. In fact, in both languages they were among the top four students, thus displaying advanced inferencing skills. (Student 14 was also mentioned in connection with the inferencing tests for the same reason.) As these two students also achieved top results when all the points for all the tests were added up, both being among the top five students in both languages, this means that the inferencing skills possessed by these two students to a great extent helped them achieve their stellar results. Student 3, on the other hand, who made comparatively few inferences in either language but who was still able to score highly in his L2, where he got the fourth best result on all the tests together, seems to have been able to rely almost entirely on already acquired knowledge. (This student was also mentioned in connection with the inferencing tests above for the same reason.)
246 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
7.6 Conclusions and Pedagogical Implications
Based on the results section (Section 7.4), inferencing indeed appears to be a commonly used strategy when learners try to figure out the meanings of previously unknown contextualised words. Since the students included in the present investigation had never been taught how to go about making inferences (information conveyed to the present author directly after the inferencing tests), success rates of 72.50%/43.18% (L1) and 70.92%/50.20% (L2) on the inferencing test on the one hand and 64.9% (L1) and 50.4% (L2) on the other tests discussed in Section 7.5 on the other hand clearly indicate the potential of possessing the skill of making successful inferences. What learners do subconsciously with great success is, however, not made use of in Swedish schools today; that is, to the present author’s knowledge, there is no part of the syllabus that is set aside for teaching students how to make inferences of the meanings of unfamiliar words in context. In order to confirm the validity of what already appears to come naturally to many learners, there is thus an urgent need to introduce this kind of instruction, perhaps by starting with the more untraditional test described in Section 7.3. However, as Nation (2001: 261) states, Guessing from context is a complex activity drawing on a range of skills and types of knowledge. It is worth bearing in mind that it is a subskill of reading and listening and depends heavily on learners’ ability to read and listen with a good level of proficiency. Learning a complex guessing strategy will not adequately compensate for poor reading or listening skills and low proficiency. Developing these reading and listening skills is the first priority.
Since research shows that 95–98% of the words in any text have to be known to a learner in order for that learner to be able to make successful inferences about unfamiliar words, the first measure that needs to be taken is to find spoken and written texts that are of appropriate levels of difficulty on an individual basis. Some of the students in the present investigation, for example, probably found the texts to be too difficult, which in turn means that these students were not really given the opportunity to show their full potential in this respect. Furthermore, since a great deal of research also shows that ‘development in the second language tends to follow and mirror development in comparable areas in the first language’ (Albrechtsen et al., 2008: xv), instruction on inferencing needs to start in the learner’s mother tongue, hence calling for cooperation between teachers in students’ L1s and L2s. This is of course quite an undertaking on the part of instructors, but it is a prerequisite if students, especially low achievers, will ever have a sporting chance of developing and strengthening their L1 inferencing skills and transferring these skills into the L2 they are trying to learn.
8 Summing Up and Directions for Future Research
In the present study, over a three-week period, 15 advanced students were subjected to eight parallel vocabulary tests, each time involving one in their second language (English) and one in their mother tongue (Swedish), thus trying to capture their mastery of L2 and L1 vocabulary in a number of different areas at one specific moment in time. (For comparative reasons, one native speaker, categorised as a low achiever, was also included in the investigation.) The overarching aim was to make a contribution to the understanding of advanced learners’ L1 and L2 mental lexicons so that necessary changes can be made to the L1 and L2 teaching syllabi. As a point of departure, Chapter 2 focused on gaining knowledge of the students’ vocabulary breadth, dealing with three of the eight parallel tests implemented: vocabulary taught at upper secondary school level, specialised uses of vocabulary, and advanced vocabulary, with the test items spanning from high-frequency words, some of which are even found on West’s (1953) general service list, to those that are very infrequently used. Comparatively low L2 scores were observed on all three tests (49.3%, 55.6% and 75% of what the native speaker was able to achieve), despite the fact that the informants are university students of English, hence testifying to them lagging behind in their general L2 developmental vocabulary trajectory. This was especially noticeable on the test focusing on vocabulary taught at upper secondary school level, which contained a productive part, on which, as compared to what they were able to achieve in their L1 and to the native speaker’s results, the learners did particularly poorly. Despite being a long since accepted distinction in the literature, the receptive/productive dichotomy has attracted relatively little research interest (Bogaards & Laufer-Dvorkin, 2004). For advanced learners, it would therefore be interesting, by incorporating even more frequent test items, to explore if there is a point at which there is no gap between receptive and productive knowledge in a second language. If such a point could be found, teachers would be able to select items and teaching techniques 247
248 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
that could close the gap at the next level, and so on. Chapter 2 was also able to show that there were correlations between what the students were able to achieve in their first as compared to their second language, with seven of the 15 informants doing consistently either well or poorly in both languages. In Chapter 3, with a focus on derivational suffixation (hence exploring vocabulary depth rather than breadth), the students’ word formation skills were put to the test. Here the first half of the parallel task tested word pairs in the form of a contextualised gap-fill exercise, while the second half dealt with the learners’ knowledge of word-family members in a decontextualised form. Surprisingly, the students did almost as well as the native speaker on the first part (achieving 92.0% of the native speaker score) and outperformed him on the second, where his ignorance of word classes resulted in a very low score. The most interesting finding, however, was that frequency was shown to play an important role in whether the learners were able to provide a correct derivative form or not. This effect was present both between word families and especially between members of one and the same family, and was particularly obvious on the first part of the test, where the students did really well with high-frequency items and generally showed total ignorance of the lowfrequency items. It was concluded that the most common derivatives (about one third of the test items on Part A) had probably already been ingrained in memory and were thus approached as unanalysed mental units. For those items that were neither comparatively frequent nor comparatively infrequent (also about one third), on the other hand, the students were thought to be in a transitional stage in which the derivatives were still analysed as consisting of stem and affixes, and the success or non-success with which they were able to provide the correct answers was dependent on a convoluted interplay among three factors: (1) the independent frequencies of the stem and derivative, (2) the complexity of the suffix (in terms of its frequency, predictability, productivity and regularity), and (3) the relative frequencies of the stem and the suffix (in which the parsing ratio of the suffix is a useful tool). Thus, if an experiment could be set up to specifically explore these parameters, the validity of this statement could be tested in greater detail. In the long run, if confirmed, it could have positive effects in the L2 classroom, enabling teachers to approach derivative forms in a more pedagogical way. Additionally, similar to other studies, the results showed that there were correlations between the students’ vocabulary size, as seen in Chapter 2, and their suffixation skills, with the learners who knew many words also doing well on the word formation test and vice versa. In Chapter 4, the students’ mastery of idioms and proverbs was explored, with each of the parallel tests incorporating 80 idioms and 20 proverbs, all of which were considered from a transparency and frequency perspective. The results showed that the learners achieved about 62% of
Summing Up and Directions for Future Research 249
what the native speaker did, but that, in contrast to the native speaker, they had to rely heavily on contextual clues to be able to disambiguate the idioms’ meanings. In English they made use of the context in as many as 619 cases, 358 of which resulted in a correct answer, yielding a correctness rate of 57.8%. Tallying well with the model of dual idiom representation (Abel, 2003), the results further showed that the degree of transparency was a determining factor as to whether an L2 idiom was disambiguated correctly or not, highly transparent items being far better understood than semi-transparent and opaque ones. (Unsurprisingly, this factor was of less importance in the informants’ mother tongue.) It was concluded that while the degrees of contextual support and transparency play leading roles in whether an L2 idiom is comprehended or not, frequency, in contrast to what was seen with derivative forms, only plays a smaller, although not insignificant, role. In Karlsson (2019), also focusing on Swedish learners of English as a second language, the major roles played by context and transparency on the one hand and the minor role played by frequency on the other hand were observed irrespective of the age of the testees (ages 13 to 18). It would therefore be interesting to investigate if the strength of the different factors in this interplay remains the same for languages that are more distant from English than Swedish. Focusing on items that have affinities with idioms, Chapter 5 dealt with noun/verb + particle combinations and multi-word verbs, both of which caused the Swedish learners major problems in their L2, with them only achieving 67.1% in the former case and 43.6% in the latter case of what the native speaker was able to attain. This chapter produced two major findings. Firstly, whereas there was a clear frequency effect on the students’ mastery of L2 multi-word verbs, with common items receiving considerably more correct answers than uncommon items, there was no such effect detected with idiomatically used prepositions. Secondly, and linked to the first observation, while the students perceived the multiword verbs as mental units, where the verb most likely prompted the correct particle use for high-frequency items, this was not at all the case with idiomatically used prepositions. As discussed in Chapter 5, the reason for this may have to do with the fact that these bound prepositions are largely ignored in the L2 classroom, very rarely being presented together with their preceding nouns, verbs and adjectives in a systematic way. An experiment based partly on the degree of predictability of these prepositional uses in relation to other types of uses, which proved fruitful in Schmidt (2004), and partly on them being presented as tightly knit units may help determine if this is a viable way to enhance learners’ knowledge of bound prepositions. Digging deeper into the advanced learners’ vocabulary knowledge, Chapter 6 dealt with their mastery of polysemous words (40 test items), lexical fields of near synonyms (another 40 test items), and false friends, all forming small worlds of their own within larger vocabulary networks.
250 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Polysemous words appeared especially difficult, with the learners overestimating their knowledge by not realising the number of meanings a word can actually convey, only achieving 62.9% of what the native speaker attained – though his result was also quite low due to an over-belief in his knowledge. Furthermore, while, in contrast to the results of some other studies, there was no frequency effect based on the polysemous words themselves, which may possibly be due to a comparatively narrow frequency span, such an effect could be observed between the different senses, the most common one generally generating more correct answers than the less common one. Although they also achieved very low success rates with lexical fields of near synonyms, a frequency effect was detected here too in the students’ L2. In this case, it was noticed not only within one and the same lexical field but also between different fields. With false friends, which gave rise to a great many erroneous answers, no such frequencybased observations could be made, so more research is needed in this area. In Chapter 7, finally, the learners’ inferencing skills were investigated, partly by constructing a parallel test for this purpose specifically and partly by exploring their use of context in some of the previous tests (vocabulary taught at upper secondary school level, specialised uses of vocabulary, advanced vocabulary and idioms). The inferencing test consisted of two parts, one being more traditional, in which the students were asked to explain the meaning of underlined words, and the other being more unorthodox in nature, in which the students were asked to find the correct word for a definition offered. On both parts, in both languages, the students showed potential for inferencing, achieving correctness rates of 50.2% (L2) and 43.2% (L1) in the former case and as high as 70.9% (L2) and 72.5% (L1) in the latter case. Tallying with what was observed in Albrechtsen et al.’s (2008) study (L2: 48%; L1: 58%), the students also displayed inferencing potential on the previous tests, where success rates of 50.4% (L2) and 64.9% (L1) were calculated. Based on the above, the conclusion can be drawn that, in order to achieve what they did, the learners made extensive use of contextual clues. As pointed out in Chapter 7, this potential should be taken into consideration when putting together teaching syllabi, recognising the worth in incorporating lessons on making use of contextual clues in general as well as how to do so. As seen in the summary offered above and observed in a wealth of previous research, frequency generally, if not in all cases, plays an important role in whether comprehension of an item of vocabulary is attained or not. Regrettably, as pointed out in Chapter 2, this important fact is not taken to heart when selecting materials for the teaching of English in Swedish schools today (Nordlund & Norberg, 2020). Instead, where the opposite needs to occur, texts appear to be chosen first, and, based on those texts, random words are picked to be made part of the students’ homework.
Summing Up and Directions for Future Research 251
Another overarching finding is that what a student is able to achieve in their first language is, in many cases, a direct reflection of what they will be able to achieve in their second language. Hence teaching of vocabulary needs to start in a learner’s L1, building a solid ground on which they will be able to stand when dealing with vocabulary in a second language, and it needs to continue, irrespective of educational level, in a learner’s L2. While students certainly receive tuition on L1 and L2 vocabulary in Swedish compulsory schooling – albeit not based on frequency, as discussed above – this does not appear to happen at university level to the same extent. When the present author did a search on vocabulary courses in English at tertiary level, there was only one found, in which, at that, vocabulary was only one of several components. Regrettably, the belief held by many seems to be that vocabulary will be acquired implicitly when focusing on other areas. While this may be true in some cases, it is the present author’s strong belief that teaching about vocabulary will enhance learning even further. I therefore hope that this book will spark an interest in doing just that, encouraging people concerned to realise the value of teaching vocabulary on a meta-level.
Appendix 1: Tables Discussed in Chapter 3
Translation of Table 3.4 No.
Word given A
Word to be formed B
4
serious
seriousness
19
bicycle
cyclist
18
forgot (forget)
forgetfulness
24
full (re: food)
satisfaction (re: food)
30
brutal
brutality
43
tolerate
tolerant
2
week
week by week
9
(the) organisation
organise
17
arrange(d)
arrangement
16
journalist
journalism
21
(the) theory
theoretic(al)
25
shame
ashamed
29
shy
shyness
33
boast(s)
boastful
1
suggestion
as a suggestion
37
forlorn
forlornness
3
in the very centre
centre
13
acceptable
accept
20
poor
poverty
11
offer(s) (verb)
offer (noun)
42
simple-minded
simple-mindedness
7
obvious
obviously
39
fall in love
love (affair)
34
idiot
idiocy
31
(the) comfort
comfortable
35
statue
statuette
15
dance
female dancer
252
Appendix 1: Tables Discussed in Chapter 3 253
48
confirm (in church)
confirmand
5
drive (driven)
fit to drive
46
verification
verify
38
theology
theologian
32
tolerate
hardy
14
aware (adj.)
awareness
22
doubt(s) (noun)
of doubtful reputation
45
stand-up
stand-up comedian
27
scholarship
holder of a scholarship
40
fanaticism
fanatic
28
(highest) fever (measured)
free from fever
36
hysteric/al
hysteria
47
instruct
instruction
23
(the) truth
truthful
8
of current interest
bring something to the fore
26
lion
lioness
12
(has) cost
costly
41
(the) quantity
quantitative
10
(peak) season
seasonal
49
fabrication
fabricate
50
gondola
gondolier
44
in an isolated place
recluse
6
(the) agreement
according to agreement
Translation of Table 3.7 WF
Word given
Word(s) to be formed
11
see each other
circle of friends
17
generous
generosity
4
deciding (noun)
decide decisive (adj.)
13
sweet
sweeten
9
dark (keep(ing) dark/, a secret, darkening)
darken
3
end in (ending in)
result
sweetening, sweetness
darkness
aiming at results, giving results, resultative 18
widening (adj.)
widen extent/scope/width, widening (noun)
254 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
12
push (pressing)
narrow/tight, be in a tight spot
8
empty/to be emptied (verb) (emptying)
emptiness, emptying
compare (oneself with) (comparing)
comparison
salt/y (adj.)
salt
crowd(ing)
7
6
empty
comparable
salt, salting saltness 1
giver (giving)
give gift generous
16
lie (lying)
liar lie lying/untruthful
2
glad
delight/happiness/joy/pleasure make someone happy/to be happy gratifying/pleasant
14
to be ashamed
shame/disgrace/dishonour ashamed shameful/disgraceful/dishonourable
5
force/to be forced
imperative/urgent force, necessity, compulsion compulsory, forced, compulsive
20
gossip (gossiping) (noun)
gossip (verb) gossipy gossipy
19
shy
shyness modesty be ashamed of
10
take care of/behave
male nurse nurse care (noun) conscientious
15
fail/let someone down
quitter deceit to be let down deceitful
Appendix 2: Tables Discussed in Chapter 5
Translation of Table 5.7 No.
Noun/verb + preposition (bold)
Bb8 (35)
frysa om (händerna)
(=(your) hands are cold)
Ba3 (17)
höra av sig
(=get in touch)
Ba5 (23)
rätta sig efter
(=conform to)
Bb9 (37)
nicka(de) mot, till, åt
(=nod(ded) at, to)
Bb4 (24)
akta sig för
(=be careful of)
Ba1 (6)
ändra på
(=change)
Bb5 (31)
svära på
(=swear (on))
Bb1 (10)
strunta i
(=ignore)
Aa9 (15)
(det nya) numret av… (tidskriften)
(=(the new) issue of (magazine))
Ba7 (30)
spionera på
(=spy on)
Bb10 (40)
misstänka(er) någon för något
(=suspect someone of something)
Bb3 (20)
förbereda sig (in)för något
(=prepare oneself for something)
Bb6 (32)
vända kappan efter vinden
(=be a turncoat)
Ba4 (22)
lida av
(=suffer from)
Aa3 (7)
(bli) föremål för
(=attract (e.g. interest))
Aa5 (9)
känsla för
(=feeling for)
Aa4 (8)
ansökan om
(=application for)
Ba10 (39)
gripa någon för något
(=arrest someone for something)
Ab5 (19)
av erfarenhet
(=by/from experience)
Ba9 (38)
tala (in)för döva öron
(=talk to deaf ears)
Aa8 (14)
ta initiativ till något
(=take the initiative in something)
Aa7 (13)
under kontroll
(=under control)
Aa2 (5)
orsak till
(=reason for)
Ab3 (3)
ha hopp om
(=entertain/have hopes of)
Ba2 (16)
hyra ut
(=let (e.g. an apartment etc. to someone))
Ab1 (1)
brist på
(=scarcity/shortage of)
Ba8 (33)
duga till något
(=be fine for)
255
256 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Bb7 (34)
svänga(er) om (orkestern, bandet)
(=the (orchestra, band) is rocking)
Ab8 (27)
ur funktion
(=not working/out of order)
Ab10 (36)
synonym till
(=synonym of)
Aa6 (11)
avtal om något
(=agreement about something)
Ab2 (2)
intresse för något
(=interest in something)
Bb2 (18)
bestämma sig för något
(=decide to do something)
Ab9 (29)
ha anlag för (sjösjuka)
(=have a tendency to (get seasick))
Ba6 (26)
lida för något
(=have to pay/suffer for something)
Ab4 (12)
chans till
(=opportunity of)
Aa10 (28)
ha en åsikt i frågan
(=hold an opinion in the matter)
Aa1 (4)
efterfrågan på
(=demand for)
Ab7 (25)
(det finns) tendens(er) till att
(=(there are) tendencies to…)
Ab6 (21)
(se) en fördel med något
(=(see) the advantage of something)
Appendix 3: Tables Discussed in Chapter 6
The correct meanings of the English polysemous words in Table 6.6 and their translations into English No.
Word
Meanings
A36
crooked
(1) krokig (=not straight, twisted)
A18
comfort
A35
nail
(2) ohederlig (=tending to cheat or deceive) (1) tröst (=strength, hope, or sympathy for an unhappy person) (2) bekvämligheter (=something that satisfies one’s physical needs) (1) nagel (=on your finger or toe) (2) spik (=a thin pointed piece of metal for hammering into a piece of wood, usually to fasten the wood to something else) A20
expand
(1) vidga (=to increase in size, number, volume, degree, etc.) (2) utveckla (=to make (a story, argument, etc.) more detailed by addition; enlarge on)
A10
intelligence
(1) intelligens (=(good) ability to learn, reason, and understand) (2) underrättelse (=information, especially about an enemy country)
A26
vocabulary
(1) ordlista (=a list of words, usually in an alphabetical order and with explanations of their meanings) (2) ordförråd (=words known, learned, used, etc.)
A12
journal
(1) tidskrift (=a serious magazine, usually produced by a specialist society) (2) dagbok (=a usually daily record of events)
A22
withdraw
(1) ta tillbaka (=to take back) (2) dra tillbaka (=to take away or take back) (3) ta ut pengar (=to take out money)
A4
weight
(1) vikt (=the heaviness of something) (2) börda (=(something that causes) a feeling of worry or anxiety) (3) brevpress (=a piece of heavy material used to put on different type of paper to make it lie still)
A29
torch
(1) fackla (=a mass of burning material tied to a stick and carried by hand to give light) (2) ficklampa (=a small electric light carried in the hand; flashlight) 257
258 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
A23
support
(1) stötta (=to bear the weight of, especially so as to keep in place or prevent from falling) (2) stödja (=to show the truth or correctness of; substantiate) (3) försörja (=to provide especially for (a person) to live on)
A14
exclusive
(1) förnäm (=that leaves out people considered to be socially unsuitable) (2) exklusive (=not taking into account; without)
A33
expose
(1) utsätta (=leave without protection) (2) avslöja (=to make known (a secretly guilty person or action)) (3) exponera (filmrulle) (=to uncover (film) to the light, when taking a photograph)
A21
pursuit
(1) förföljande (=the act of following) (2) utövande (=an activity to which one gives one’s time, whether for work or for pleasure)
A3
ability
(1) förmåga (=having the power to do something) (2) begåvning (=having the skill to do something)
A13
resignation
(1) avskedsansökan (=(an act or written statement of) quitting one’s job) (2) undergivenhet (=submission to someone)
A34
suppress
(1) slå ned (=to destroy or bring to an end by force) (2) dra in (=to prevent from being printed or made public) (3) hemlighålla (=to prevent from being shown)
A40
brood
(1) ruva (=to sit on eggs) (2) grubbla (=to spend time thinking anxiously or sadly about something; worry or ponder)
A7
credit
(1) tilltro (=belief or trust in the truth or rightness of something) (2) ära (=public approval or praise given to someone because of something they have done) (3) kredit (=a system of buying goods or services and paying for them later) (4) poäng (=a completed unit of a student’s work that forms part of a course, especially at a university)
A6
administration
(1) skötsel (=management) (2) förvaltning (=management of the affairs of a business) (3) regering (=the government) (4) utdelande (=the act of giving)
A32
idle
(1) sysslolös (=not working or operating productively) (2) lat (=disliking or avoiding activity or work) (3) utan mening (=having no particular purpose)
A19
dealer
(1) -handlare (=a person in a stated type of business) (2) börsspekulant (=a person speculating on the stock exchange) (3) kortgivare (=a person handing out cards in a card game) (4) knarklangare (=a person selling drugs)
Appendix 3: Tables Discussed in Chapter 6 259
A38
haul
A11
peak
(1) hala (=to pull with effort or difficulty) (2) transportera (=to carry (goods) in a vehicle, especially a truck) (1) bergstopp (=a sharply pointed mountain top) (2) mösskärm (=the flat, curved part of a cap that sticks out in front above the eyes) (3) höjdpunkt (=the highest point, level, etc., especially of a varying amount, rate, etc.)
A15
wire
(1) ledning (=a piece of metal, usually covered with plastic, used for carrying electricity) (2) målsnöre (=a finishing tape in a race) (3) telegram (=(a piece of paper with) a message sent by telegraph)
A9
keen
(1) skarp (t. ex. kniv) (=sharp (e.g. knife)) (2) genomträngande (t. ex. vind) (=cold (e.g. wind)) (3) skarpsinnig (=(of the mind, the senses, feelings, etc.) good, strong, quick at understanding, etc.) (4) ivrig (=(of a person) having a strong, active interest in something; eager to do something)
A5
domestic
(1) hushålls- (=of or in the house or home) (2) hemkär (=enjoying home duties and pleasures) (3) inrikes (=of or within a particular country) (4) tam (=(animal) not wild but kept in a house or on a farm)
A2
director
(1) direktör (=a person who directs an organisation) (2) regissör (=a person who instructs the actors, cameramen, etc. when making a film) (3) dirigent (=a person who directs an orchestra) (4) studierektor (=a person who is in charge of studies)
A17
suit
(1) kostym (=a set of clothes made of the same material) (2) rättegång (=(an act of) hearing and judging a person, case, or point of law in a court) (3) färg i kortspel (=any of the four sets of cards used in games)
A24
ugly
(1) ful (=unpleasant to look at; extremely unattractive) (2) otrevlig (=very unpleasant or threatening) (3) sur (=very sulky)
A1
board
(1) bräda (=a long thin flat piece of wood) (2) anslagstavla (=a flat piece of hard material used to put or write things on) (3) kost (=(the cost of) meals) (4) styrelse (=an official group that has responsibility for a particular organisation or activity)
A8
sheet
(1) lakan (=a large four-sided piece of usually cotton or nylon cloth used on a bed) (2) platta (=a broad regularly shaped piece of glass or metal) (3) ark (=a piece of paper) (4) tidning (=a newspaper)
260 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
A16
square
(1) kvadrat (=a shape with four straight equal sides forming four right angles) (2) torg (=a broad open area in a town) (3) sjalett (=a piece of cloth to tie around your head) (4) vinkelhake (=a straight-edged often L-shaped tool for drawing and measuring right angles)
A25
grace
(1) charm (=a fine and attractive quality in movement or form, especially when this seems effortless and natural) (2) älskvärdhet (=willingness to behave in a fair and honourable way) (3) frist (=a delay allowed as a favour) (4) bordsbön (=a prayer before or after meals, giving thanks to God)
A27
needle
(1) nål (=a long, pointed metal pin with a hole in one end through which a piece of thread is passed, used in sewing) (2) spruta (=a thin hollow pointed tube that is pushed into someone’s skin to put a liquid (especially medicine) into the body) (3) klippspets (=the top of a mountain) (4) obelisk (=a tall pointed stone pillar built usually in honour of a person or an event)
A28
rude
(1) ohövlig (=(of a person or their behaviour) not at all polite; intentionally bad-mannered, offensive) (2) obildad (=uneducated) (3) våldsam (=sudden and unpleasant) (4) primitiv (=simple and roughly made)
A30
trunk
(1) trädstam (=the thick main stem of a tree) (2) bål (kroppsdel) (=the human body apart from the head and limbs) (3) koffert (=a large heavy case or box in which clothes or belongings are stored or packed for travel) (4) snabel (=the very long cylindrical nose of an elephant)
A31
brow
(1) ögonbryn (=the hair above your eye) (2) panna (=the part of the face above the eyes and below the hair) (3) krön (=the upper part of a slope)
A37
advance
(1) flytta framåt (=to move forward in position) (2) (be)främja (=to help, improve, or bring advantage to) (3) påskynda (=to bring forward to an earlier date or time) (4) lägga fram (=to introduce; suggest) (5) förskottera (=to provide (money) earlier than the proper or usual time)
A39
discharge
(1) lasta av (=to unload) (2) avlossa (=to fire or shoot (a gun, arrow, etc.)) (3) ladda ur (=to lose stored electrical power) (4) avsöndra (=to pour or let out (gas, liquid, etc.)) (5) frige (=to allow or tell (a person) to go)
Appendix 3: Tables Discussed in Chapter 6 261
The correct meanings of the Swedish polysemous words and their translations into English for Table 6.7 No.
Word
Meanings
A19
greppa
(1) ta tag i (=take (a)hold of)
A26
krasch
(2) begripa (=comprehend) (1) skräll (=bang, crash) (2) våldsam sammanstötning (=violent encounter) A24
dragningskraft
A12
platt
A13
samsas
A29
huvudbry
(1) gravitation (=gravitation) (2) charm (=charm) (1) slät (=even, flat) (2) intetsägande (=empty (e.g. conversation), insipid) (1) komma överens (=get on) (2) i sämja dela utrymme (=share space without any trouble) (1) bekymmer (=worry) (2) gåta (=conundrum)
A25
revolutionerande
(1) omstörtande (=subversive) (2) upprorisk (=rebellious) (3) nydanande (=innovative)
A11
atmosfär
A27
begrava
(1) luft (=air) (2) stämning (=atmosphere) (1) jordfästa (=read the funeral service over) (2) täcka (=cover) (3) dra ett streck över (=draw a line across)
A31
frälst
(1) omvänd (=converted) (2) såld på något (=be sold on) (3) räddad (=saved)
A8
remiss
A9
avbryta
(1) överlämnande (=handing over) (2) hänvisning (=referral) (1) hastigt avsluta (=finish abruptly) (2) göra en paus (=take a break) (3) snäsa av (=snub)
A4
tagen
(1) bestört (=perplexed) (2) gripen (=arrested, taken, seized) (3) utmattad (=exhausted)
A21
nicka
(1) hälsa (=nod) (2) skalla (=head) (verb) (3) slumra till (=doze off)
A22
gap
(1) svalg (=throat) (2) öppning (=opening) (3) tomrum (=void)
262 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
A17
fiktion
A37
grina
(1) inbillning (=imagination) (2) skönlitteratur (=fiction) (1) gråta (=cry) (2) grimasera (=grimace) (3) skratta (=laugh)
A15
harmoni
(1) välljud (=euphony) (2) överensstämmelse (=concordance) (3) sinnesro (=tranquillity of mind)
A33
flyktig
(1) avdunstande (=evaporating) (2) obeständig (=transient) (3) ytlig (=superficial) (4) snabb (=swift)
A30
darra
(1) skaka (=tremble) (2) rysa (=shiver, shudder) (3) flimra (=flicker) (4) vibrera (=vibrate)
A1
börja
(1) sätta igång (=get started) (2) ta sin början (=begin) (3) tillträda (=take office) (4) grunda (=establish) (5) ta första steget (=take the first step)
A7
skugga
(1) obelyst plats (=a place in the shadow) (2) blek avbild (=not a very good copy) (3) skepnad (=phantom, figure) (4) ständig följeslagare (=constant companion)
A38
teknikalitet
A2
ersättning
(1) teknisk detalj (=technical detail) (2) formalitet (=formality) (1) kompensation (=compensation) (2) betalning (=payment) (3) gengäld (=in return) (4) reserv (=substitute)
A32
fullfjädrad
(1) flygfärdig (=(fully) fledged) (2) erfaren (=experienced) (3) tvättäkta (=genuine, true)
A35
herravälde
(1) makt (=power) (2) överhand (=the upper hand) (3) dominans (=dominance) (4) kommando (=command)
A20
brunn
(1) vattenhål (=water-hole) (2) fontän (=fountain)
Appendix 3: Tables Discussed in Chapter 6 263
(3) kurort (=health resort) (4) schakt (=shaft) A28
dirigera
(1) vara chef över (=be the manager of) (2) leda uppförandet av (=be in charge of the building, construction, erection, etc. of) (3) ge riktning åt (=direct) (4) skicka (=send)
A34
avig
(1) bakfram (=the wrong way round) (2) ovillig (=unwilling) (3) klumpig (=awkward) (4) befängd (=absurd)
A39
krafsa
(1) skrapa (=scrape) (2) klottra (=doodle) (3) klösa (=scratch)
A6
marginal
(1) kant (=border, edge, verge) (2) spelrum (=scope) (3) reserv (=reserve)
A36
arbetsam
A3
tillfälle
(1) flitig (=hard-working, industrious) (2) tröttsam (=tiresome) (1) händelse (=event, occurrence) (2) tidpunkt (=moment) (3) erbjudande (=offer) (4) möjlighet (=chance)
A5
dryg
(1) överlägsen (=haughty) (2) varaktig (=lasting) (3) riklig (=abundant) (4) tidsödande (=time-consuming)
A10
element
(1) beståndsdel (=component) (2) grundämne (=element (e.g. gold)) (3) värmekälla (=source of heat) (4) väder och vind (=wind and weather) (5) omgivning (=surroundings)
A14
depression
(1) tryckt sinnesstämning (=low-spirited mood) (2) lågkonjunktur (=recession) (3) dalgång (=long(ish) valley) (4) lågtryck (=low pressure (weather))
A16
jäsa
(1) höja sig (=rise) (2) koka inombords (=boil (be very angry without showing it)) (3) kråma sig (=strut)
A18
lukta
(1) sprida doft (=diffuse (scent))
264 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
(2) vädra (=scent) (3) andas in (=breathe in, inhale) (4) ha en anstrykning av (=have a touch of) (5) studera flyktigt eller ytligt (=browse) A23
koloni
(1) jordlott (=allotment) (2) nybygge (=colony) (3) utlandssvenskar (=a group of Swedish citizens living in the same area abroad) (4) sommarläger (=summer camp)
A40
tetig
(1) besvärlig (=difficult) (2) underlig (=strange)
The definitions of the English near synonyms in Table 6.15 Lexical field
Word
Definition
LF2
bright
(of a colour) strong, clear, and easily seen (the opposite is dull)
light
(of the shade of a colour) (the opposite is dark)
fair
(of a colour) blond, used almost always about hair
carpet
heavy woven often woollen material for covering floors or stairs, i.e. the general term for matta
rug
(1) thick, usually woollen, comparatively small; (2) large warm woollen covering to wrap round oneself
mat
a piece of rough strong material for covering part of a floor
solicitor
gives advice, speaks in the lower courts of law, and prepares cases for another person to argue in a higher court (especially in England)
lawyer
the general term for advokat
attorney
a person whose business is to advise people about laws and to represent them in court (especially in the USA)
barrister
has the right of speaking in the higher courts of law (especially in England and Wales)
advocate
(1) a person whose business is to advise people about laws and to represent them in court; (2) a person who speaks for or supports an idea, a way of life, etc.
claim
to declare to be true, especially in the face of opposition
maintain
to (continue to) argue in favour of or declare to be true
state
to say, express, or put into words, especially formally (neutral word)
assert
to declare forcefully
allege
to declare without proof or before finding proof
tremble
to shake uncontrollably, especially from fear, excitement, or weakness
shiver
to shake slightly, especially from cold or fear; the most common word with the adjective cold (i.e. X with cold)
shudder
to shake uncontrollably for a moment, especially from fear, cold, or strong dislike
LF5
LF4
LF6
LF8
Appendix 3: Tables Discussed in Chapter 6 265
LF3
LF7
LF1
quiver
to make a slight movement, especially from fear or excitement
quaver
to say something in a shaky voice
accident
something, especially something unpleasant or damaging, that happens unexpectedly or by chance, i.e. the general term for olycka
disaster
(1) a sudden serious event causing great suffering and damage; (2) a complete failure
catastrophe
a sudden, unexpected and terrible event that causes great suffering, or ruin
misfortune
bad luck, often of a serious kind
calamity
a sudden, unexpected, and terrible event that causes great suffering, or ruin1
mishap
something that goes wrong; not very serious
jump
to push oneself into the air or away from a surface by the force of one’s legs, i.e. the general term for hoppa
leap
to jump up in the air or over something
hop
(of people) to jump on one leg
skip
to move in a light dancing way, with quick steps and jumps
bound
(1) to move along quickly from place to place by jumping movements; (2) to bounce back from a surface
spring
to move quickly and suddenly upwards or forwards as if by jumping
disease
an unhealthy condition caused by infection etc., but not by an accident; often serious
illness
an unhealthy state of the body or mind, i.e. the general term for sjukdom
sickness
(1) the condition of not being well; (2) nausea
infirmity
a weakness of body or mind, used in ‘the Xs of old age’
ailment
something that is not serious, sometimes imaginary
disorder
a failure of part of the body or mind, used in ‘some X of the stomach’
complaint
something that is not serious, used especially for children
Note (1) According to the reference book used, there is no difference in meaning between catastrophe and calamity.
Appendix 4: Tables Discussed in Chapter 7
The English words tested in Texts A, B, E and F No. (frequency order)
English word to be found in context (presented in order of accuracy, regardless of whether the words were inferred or not)
A6 (41)
slacker (14)
A10 (27)
misogynists (14)
E3 (28)
copycats (13)
E1 (35)
hoaxers (13)
B10 (20)
poachers (13)
B1 (39)
teemed with (13)
E4 (23)
crestfallen (12)
A9 (44)
wiggled through (12)
B5 (29)
antlers (12)
B4 (6)
deforestation (12)
A7 (10)
dean (12)
A8 (4)
disparity (11)
A11 (3)
murky (11)
F9 (38)
retirees (10)
B8 (34)
mishmash (10)
B11 (40)
pilfer (10)
E10 (32)
was in stitches (10)
F6 (42)
crime-infested (9)
E6 (43)
matter (noun) (9)
F7 (12)
play down (9)
B3 (21)
epitomizes (9)
B9 (16)
tenacious (8)
B6 (26)
takes its toll on (8)
A3 (33)
lamented (8)
B7 (36)
gumption (8)
266
Appendix 4: Tables Discussed in Chapter 7 267
F4 (2)
gangsters, thugs1 (7)
E5 (5)
spawned (7)
F1 (8)
mayhem (7)
F8 (25)
hotbeds (7)
A5 (37)
pterodactyl (7)
E8 (13)
chagrin (noun) (7)
A4 (22)
parity with (7)
E2 (18)
apparition (6)
E11 (11)
untold (6)
B2 (1)
plight (noun) (6)
A2 (9)
alumni (6)
F3 (17)
succinct (6)
A12 (19)
bigotry (5)
F5 (14)
dereliction (5)
E7 (7)
judicious (5)
A1 (24)
affirmative action (4)
B12 (31)
litigious (3)
E9 (15)
forays (3)
F2 (30)
telltale (2)
The Swedish words tested in Texts A, B, E and F No. (frequency order)
Swedish word to be found in context (presented in order of accuracy, regardless of whether the words were inferred or not)
F9 (36)
tjudra (15)
(=tether) (verb)
F1 (10)
provokativ (15)
(=provocative)
A1 (23)
deserterade (14)
(=deserted)
E2 (15)
initialt (14)
(=initially)
F2 (11)
skärskådade (14)
(=examined)
E1 (32)
förolyckade (14)
(=those who have lost their lives)
E3 (35)
av första rang (14)
(=leading) (adj.)
B1 (12)
likar (14)
(=equals)
B4 (16)
artikulera (13)
(=articulate)
E5 (2)
fragmenten (13)
(=the fragments)
B5 (30)
eggar (13)
(=stimulate)
A2 (38)
umbäranden (13)
(=hardships)
E8 (29)
suggereras (13)
(=is stimulated)
B3 (9)
instruktioner (12)
(=instructions)
E11 (18)
exponerats för (12)
(=was exposed to)
E7 (1)
sedermera (12)
(=later on)
268 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
F4 (7)
animal (12)
(=animal)
E12 (28)
fabricerade (12)
(=fabricated)
F8 (39)
diminutiver (12)
(=diminutives)
A4 (33)
nihilist (12)
(=nihilist)
F3 (22)
penetrera (11)
(=penetrate)
E6 (6)
erinra sig (11)
(=remind themselves)
B7 (4)
intimt (11)
(=intimately)
B2 (27)
fränder (11)
(=kinsmen)
E13 (5)
benägna (11)
(=inclined, willing)
A9 (34)
velocipeden (10)
(=the bicycle)
A5 (21)
kuriositet (10)
(=peculiarity)
E9 (37)
välbelagt (10)
(=well-proven)
B8 (20)
imperativ (9)
(=imperative)
A11 (8)
mylla (9)
(=earth) (noncount)
A3 (3)
amnesti (9)
(=amnesty)
B6 (43)
uttryckshorisont (9)
(=the extent of expression)
E10 (13)
bedrägliga (9)
(=deceptive)
E14 (40)
har bäring på (9)
(=have bearing on)
A12 (24)
illusorisk (9)
(=imaginary)
F6 (44)2
last (8)
(=bad habit, vice)
E15 (14)
härdade (7)
(=patient) (adj.)
F5 (26)
älskog (7)
(=love-making)
A8 (17)
kontroverser (7)
(=controversies)
E4 (19)
i god ordning (6)
(=as intended)
F7 (41)
liderliga (6)
(=lecherous)
A6 (31)
disparata (4)
(=different)
A7 (42)
taxonomi (2)
(=taxonomy)
A10 (25)
patologisk (0)
(=pathological)
The English words tested in Texts C and D No. (frequency order)
English word given in context (presented in order of accuracy, regardless of whether the words were inferred or not)
C5 (12)
raspberries (14)
D16 (13)
ginger (13)
C17 (22)
shears (13)
C10 (48)
leather thong (13)
C4 (28)
currants (12)
C6 (32)
pitcher (to put liquid in) (12)
D20 (30)
minced (12)
Appendix 4: Tables Discussed in Chapter 7 269
C19 (55)
toppled off (12)
D27 (41)
dimples (11)
D3 (6)
limelight (11)
D29 (33)
downtrodden (11)
C2 (3)
bridle (11)
D32 (36)
tinkling (11)
D22 (34)
is not a patch on (11)
D19 (27)
besieged (11)
D2 (8)
constellation (10)
D24 (46)
slippered (10)
D11 (44)
in cold storage (used symbolically) (10)
D14 (15)
bondage (9)
D13 (39)
remonstrated (9)
C12 (26)
stoutly (9)
C16 (11)
gymnasium (8)
D7 (10)
succumb (8)
C9 (17)
denounced (8)
D4 (18)
infallible (8)
D36 (1)
stale (8)
C8 (31)
listlessly (8)
C15 (53)
planed (wood) (8)
D31 (7)
gaunt (8)
D12 (45)
alluringly (7)
C3 (50)
pettish (7)
D28 (54)
sober-suited (6)
D33 (35)
connivance (6)
D5 (42)
charwomen (5)
C13 (23)
fiends (4)
C11 (37)
simpleton (4)
C20 (20)
nettles (noun) (4)
D26 (40)
perfunctorily (4)
D8 (5)
inextricably (4)
D23 (52)
trammelled (4)
C18 (9)
furnace (3)
D9 (4)
sullen (3)
C14 (47)
dismal (3)
D18 (29)
rectitude (3)
C7 (21)
sultry (2)
D35 (14)
grimy (2)
D1 (2)
arid (2)
270 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
D25 (51)
connubial (2)
D17 (19)
dingy (1)
D34 (16)
coy (1)
C1 (43)
buckwheat (1)
D10 (25)
abeyance (1)
D6 (38)
wistfulness (0)
D21 (24)
canvassing (0)
D15 (49)
chafe (0)
D30 (56)
wool-gathered (0)
The Swedish words tested in Texts C and D No. (frequency order)
Swedish word given in context (presented in order of accuracy, regardless of whether the words were inferred or not)
C12 (2)
måhända (15)
(=maybe)
C2 (16)
eskalerade (15)
(=escalated)
D22 (7)
dispyten (15)
(=the argument)
D3 (14)
odören (14)
(=the odour)
D5 (6)
tynande (adj.) (14)
(=fading away)
C1 (4)
stundom (14)
(=sometimes)
D4 (5)
spirande (adj.) (13)
(=growing)
D20 (35)
allmosa (13)
(=alms)
D25 (10)
botanik (13)
(=botany)
D15 (23)
expedieras (13)
(=be sent away)
C18 (18)
falsarier (13)
(=falsifications)
D16 (27)
bastarden (12)
(=a child born out of wedlock)
C16 (44)
understår sig (12)
(=dares)
D32 (53)
tvådde sig (12)
(=washed himself)
D26 (17)
oraklen (12)
(=the oracles)
C11 (43)
luttrats (11)
(=have been tested)
D8 (48)
katakomber (11)
(=catacombs)
C4 (8)
torftig (11)
(=plain and shabby) (=intense)
D6 (22)
infernalisk (11)
C15 (56)3
gravare (10)
(=more damaging)
D27 (37)
amulett (10)
(=amulet)
D19 (42)
genmälde (10)
(=replied)
D31 (55)
vämjdes (10)
(=made him feel sick)
C9 (31)
inkorporerat (9)
(=have incorporated)
C22 (38)
näpst (9)
(=have been rebuked)
D24 (32)
entledigad (9)
(=dismissed)
Appendix 4: Tables Discussed in Chapter 7 271
C3 (21)
bryderiet (8)
(=tricky situation)
C19 (29)
förvillelse (8)
(=aberration)
C13 (19)
fabulerande (8)
(=giving her imagination free rein)
C24 (3)
vanmakt (8)
(=powerlessness)
D2 (33)
inskränkte sig (7)
(=were limited to)
D28 (15)
enfaldigt (7)
(=silly)
C14 (12)
gungfly (7)
(=quagmire)
C5 (11)
i vardande (7)
(=soon becoming…)
D7 (47)
grifterna (6)
(=the graves)
D29 (54)
vederstyggliga (6)
(=abominable)
C10 (24)
relativiserar (6)
(=compare)
C17 (30)
i högönsklig välmåga (6)
(=in the best of health)
D9 (1)
alltjämt (5)
(=still)
C20 (25)
altruistisk (5)
(=altruistic)
C8 (20)
misskund (5)
(=compassion)
C6 (9)
snöpliga (5)
(=disappointing)
D11 (41)
effektuerades (5)
(=were executed)
D30 (51)
prisgav (5)
(=did not reveal)
D13 (45)
undfick (4)
(=received)
C7 (13)
indignerade (4)
(=indignant)
D23 (49)
käxade (4)
(=nagged)
D10 (52)
räntabelt (3)
(=profitable)
C21 (26)
vederbörligen (2)
(=in due form)
C23 (36)
förhärdad (2)
(=callous)
D18 (34)
rundlig (2)
(=a great while)
D17 (28)
fryntlig (2)
(=jovial)
D21 (46)
genstörtighet (2)
(=recalcitrance)
D12 (39)
tillhållna (2)
(=were urged)
D14 (40)
priorn (2)
(=the prior)
D1 (50)
mörkmän (0)
(=obscurantists)
Notes (1) Both words were possible solutions here. (2) This word has a second meaning that is more common than the one used here. Since it was impossible to separate these two meanings in the Swedish corpus, it is not possible to say anything about the frequency of the meaning used here. (3) The frequency of this word could not be calculated due to the fact that, in addition to the meaning included in the present study, it also has a second, more common, meaning.
References
Abdullah, K. and Jackson, H. (1998) Idioms and the language learner: Contrasting English and Syrian Arabic. Language in Contrast 1, 83–107. Abel, B. (2003) English idioms in the first language and second language lexicon: A dual representation approach. Second Language Research 19, 329–358. Adolphs, S. and Schmitt, N. (2003) Lexical coverage of spoken discourse. Applied Linguistics 24, 425–438. Aitchison, J. (1987) Words in the Mind: An Introduction to the Mental Lexicon (1st edn). Oxford: Blackwell. Aitchison, J. (1994) Words in the Mind: An Introduction to the Mental Lexicon (2nd edn). Oxford: Blackwell. Aitchison, J. (2003) Words in the Mind: An Introduction to the Mental Lexicon (3rd edn). Oxford: Blackwell. Albrechtsen, D., Haastrup, K. and Henriksen, B. (2008) Vocabulary and Writing in a First and Second Language: Processes and Development. New York: Palgrave Macmillan. Alderson, J.C. (2005) Diagnosing Foreign Language Proficiency: The Interface between Learning and Assessment. London: Continuum. Alderson, J.C., Clapham, C.M. and Steel, D. (1997) Metalinguistic knowledge, language aptitude, and language proficiency. Language Testing Research 1, 93–121. Al-Duleimi, A.D.D. and Aziz, R.N. (2016) Humour as EFL learning-teaching strategy. Journal of Education and Practice 7 (10), 105–115. Allén, S., Gellerstam, M. and Malmgren, S.-G. (1989) Orden speglar samhället. Stockholm: Allmänna Förlaget. Alm-Arvius, C. (1998) Introduction to Semantics. Lund: Lund University Press. Alm-Arvius, C. (2003) Figures of Speech. Lund: Studentlitteratur. Alstertun, R. (1996a) Ordkunskap 1 (2nd edn). Stockholm: Almqvist & Wiksell. Alstertun, R. (1996b) Ordkunskap 2 (2nd edn). Stockholm: Almqvist & Wiksell. Anderson, J.R. (1983) The Architecture of Cognition. Cambridge, MA: Harvard University Press. Anderson, J.R. (1990) Cognitive Psychology and Its Implications (3rd edn). New York: Lawrence Erlbaum. Andersson, L.-G. (1987) Språktypologi och språksläktskap (3rd edn). Stockholm: Skriptor/Almqvist & Wiksell. Ard, J. and Homburg, T. (1983) Verification of language transfer. In S. Gass and L. Selinker (eds) Language Transfer in Language Learning (pp. 47–70). Rowley, MA: Newbury House. Artley, A.S. (1943) Teaching word-meaning through context. Elementary English Review 20, 68–74. Bahns, J. and Eldaw, M. (1993) Should we teach EFL students collocations? System 21, 101–114. Barrow, J., Nakashimi, Y. and Ishino, H. (1999) Assessing Japanese college students’ vocabulary knowledge with a self-checking familiarity survey. System 27, 223–247. Bauer, L. and Nation, P. (1993) Word families. International Journal of Lexicography 6 (4), 253–279. 272
References 273
Becker, W.C., Dixon, R. and Anderson-Inman, L. (1980) Morphographic and Root Word Analysis of 26,000 High-Frequency Words. University of Oregon, Follow Through Project. Eugene, OR: College of Education. Bell, H. (2009) The messy little details: A longitudinal case study of the emerging lexicon. In T. Fitzpatrick and A. Barfield (eds) Lexical Processing in Second Language Learners: Papers and Perspectives in Honour of Paul Meara (pp. 111–127). Bristol: Multilingual Matters. Bensoussan, M. and Laufer, B. (1984) Lexical guessing in context in EFL reading comprehension. Journal of Research in Reading 7, 15–32. Beréndi, M., Csábi, S. and Kövecses, Z. (2008) Using conceptual metaphors and metonymies in vocabulary. In F. Boers and S. Lindstromberg (eds) Cognitive Linguistic Approaches to Teaching Vocabulary and Phraseology (pp. 65–100). Berlin: Mouton de Gruyter. Berko, J. (1958) The child’s learning of English morphology. Word 14, 150–177. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (1999) Longman Grammar of Spoken and Written Language. Harlow: Longman. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (2021) Grammar of Spoken and Written English. Amsterdam: John Benjamins. Bird, N. (1987) Words, lemmas and frequency lists: Old problems and new challenges. (Parts 1 & 2). Al-Manakh 6, 42–50. Bird, N. (1990) A First Handbook of the Roots of English. Hong Kong: Lapine Education and Language Services. Blum, S. and Levenston, E.A. (1978) Universals of lexical simplification. Language Learning. A Journal of Research in Language Studies 28 (2), 399–415. Bobrow, S. and Bell, S. (1973) On catching on to idiomatic expressions. Memory and Cognition 1, 343–346. Boers, F. (2000) Metaphor, awareness and vocabulary retention. Applied Linguistics 21, 553–571. Boers, F. and Demecheleer, M. (2001) Measuring the impact of cross-cultural differences on learners’ comprehension of imageable idioms. ELT Journal 55, 255–262. Boers, F., Demecheleer, M. and Eyckmans, J. (2004) Etymological elaboration as a strategy for learning idioms. In P. Bogaards and B. Laufer-Dvorkin (eds) Vocabulary in a Second Language: Selection, Acquisition and Testing (pp. 53–78). Philadelphia, PA: John Benjamins. Bogaards, P. and Laufer-Dvorkin, B. (2004) Vocabulary in a Second Language. Amsterdam: John Benjamins. Bolinger, D. (1971) The Phrasal Verb in English. Cambridge, MA: Harvard University Press. Boyd Zimmerman, C. (1997) Historical trends in second language vocabulary instruction. In J. Coady and T. Huckin (eds) Second Language Vocabulary Acquisition: A Rationale for Pedagogy (pp. 5–19). Cambridge: Cambridge University Press. Bradley, D.C. (1979) Lexical representation of derivational relation. In M. Aronoff and M.L. Kean (eds) Juncture (pp. 37–55). Cambridge, MA: MIT Press. Brown, R. and McNeill, D. (1966) The ‘tip of the tongue’ phenomenon. Journal of Verbal Learning and Verbal Behavior 5, 325–337. Bywater, F.V. (1969) A Proficiency Course in English. London: University of London Press. Cacciari, C. and Levorato, M.C. (1989) How children understand idioms in discourse. Journal of Child Language 16, 387–405. Cacciari, C. and Tabossi, P. (1988) The comprehension of idioms. Journal of Memory and Language 27, 668–683. Caramazza, A., Laudanna, A. and Romani, C. (1988) Lexical access and inflectional morphology. Cognition 28, 297–332.
274 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Carlisle, J.F. (2000) Awareness of the structure and meaning of morphologically complex words: Impact on reading. Reading and Writing: An Interdisciplinary Journal 12, 169–190. Carlo, M.S., August, D., McLaughlin, B., Snow, C.E., Dressler, C., Lippman, D.N., Lively, T.J. and White, C.E. (2004) Closing the gap: Addressing the vocabulary needs of English language learners in bilingual and mainstream classrooms. Reading Research Quarterly 39 (2), 188–215. Carnine, D., Kameeniu, E.J. and Coyle, G. (1984) Utilization of contextual information in determining the meaning of unfamiliar words. Reading Research Quarterly 19, 188–204. Cederberg, S. (1993) Språkfärdighet i engelska: Systematiska övningar med lösningar. Gothenburg: New School. Cederholm, P. and Danell, A. (2007) Känslan för ord. Svenska B. Malmö: Gleerup. Celce-Murcia, M. (1978) The simultaneous acquisition of English and French in a twoyear old child. In E.M. Hatch (ed.) Second Language Acquisition (pp. 38–53). Rowley, MA: Newbury House. Celce-Murcia, M. and Larsen-Freeman, D. (1999) The Grammar Book: An ESL/EFL Teacher’s Course (2nd edn). Boston, MA: Heinle/Cengage. Cermak, L. and Craik, F. (1979) Levels of Processing in Human Memory. Hillsdale, NJ: Lawrence Erlbaum. Chacón-Beltrán, R., Abello-Contesse, C. and del Mar Torreblanca-López, M. (2010) Vocabulary teaching and learning: Introduction and overview. In R. Chacón-Beltrán, C. Abello-Contesse and M. del Mar Torreblanca-López (eds) Insights into Nonnative Vocabulary Teaching and Learning (pp. 1–12). Bristol: Multilingual Matters. Channell, J. (1981) Applying semantic theory to vocabulary teaching. ELT Journal 35, 115–122. Chihara, T., Oller, J., Weaver, K. and Chavez-Oller, M.A. (1977) Are cloze items sensitive to discourse constraints? Language Learning 27, 63–73. Church, K.W., Gale, W., Hanks, P., Hindle, R. and Moon, R. (1994) Lexical substitutability. In B.T. Atkins and A. Zampolli (eds) Computational Approaches to the Lexicon (pp.153–177). Oxford: Oxford University Press. Cieślicka, A. (2006) On building castles on the sand, or exploring the issue of transfer in the interpretation and production of L2 fixed expressions. In J. Arabski (ed.) Crosslinguistic Influences in the Second Language Lexicon (pp. 226–245). Clevedon: Multilingual Matters. Cignoni, L. and Coffey, S. (2000) A corpus study of Italian proverbs: Implications for lexicographic description. In U. Heid, S. Evert, E. Lehmann and C. Rohrer (eds) Proceedings of the Ninth EURALEX International Congress, EURALEX 2000 (pp. 549–555). Stuttgart: Institut für Maschinelle Sprachverarbeitung, Universität Stuttgart. Clark, J.M. and Paivio, A. (1991) Dual coding theory and education. Educational Psychology Review 3, 233–262. Coady, J. and Huckin, T. (eds) (1997) Second Language Vocabulary Acquisition. Cambridge: Cambridge University Press. Cohen, A.D. and Aphek, E. (1981) Easifying second language learning. Studies in Second Language Acquisition 3 (2), 221–236. Coleman, L. and Kay, P. (1981) Prototype semantics: The English verb lie. Language 75 (1), 26–44. Condon, N. (2008) How cognitive linguistic motivations influence the learning of phrasal verbs. In F. Boers and S. Lindstromberg (eds) Cognitive Linguistic Approaches to Teaching Vocabulary and Phraseology (pp. 133–158). Berlin: Mouton de Gruyter. Condon, N. and Kelly, P. (2002) Does Cognitive Linguistics have anything to offer English language learners in their efforts to master phrasal verbs? International Journal of Applied Linguistics 137–138, 205–231.
References 275
Cooper, T.C. (1998) Teaching idioms. Foreign Language Annals 31, 255–266. Cooper, T.C. (1999) Process of idioms by L2 learners of English. TESOL Quarterly 33, 233–262. Craik, F. and Lockhart, R.S. (1972) Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior 11, 671–684. Crossley, S., Salsbury, T. and McNamara, D. (2010) The development of polysemy and frequency use in English second language speakers. Language Learning 60 (3), 573–605. Cunningham, P. (1998) The multisyllabic word dilemma: Helping students build meaning, spell, and read ‘big’ words. Reading and Writing Quarterly: Overcoming Learning Difficulties 14, 189–218. Dagut, M.B. and Laufer, B. (1985) Avoidance of phrasal verbs by English learners, speakers of Hebrew: A case for contrastive analysis. Studies in Second Language Acquisition 7, 73–80. Daneman, M. and Green, I. (1986) Individual differences in comprehending and producing words in context. Journal of Memory and Language 25, 1–18. Dupuy, H.J. (1974) The Rationale, Development and Standardization of a Basic Word Vocabulary Test. Washington, DC: U.S. Government Printing Office. Edegran, B.-M. and Lönneborg, A. (2005) Ord med mening: Ordträning i svenska som andraspråk. Stockholm: Bonnier utbildning. Ellis, N.C. (2003) Constructions, chunking, and connectionism: The emergence of second language structure. In C.J. Doughty and M.H. Long (eds) The Handbook of Second Language Acquisition (pp. 63–103). Malden, MA: Blackwell. Ellis, N.C. and Beaton, A. (1993) Factors affecting foreign language vocabulary: Imagery keyword mediators and phonological short-term memory. Quarterly Journal of Experimental Psychology 46A, 533–558. Engström, E. (1995) Svensk-engelsk teknisk ordbok: Swedish-English Technical Dictionary (10th edn). Stockholm: Svensk trävarutidning. Erman, B. and Warren, B. (2000) The idiom principle and the open choice principle. Text – An Interdisciplinary Journal for the Study of Discourse 20 (1), 29–62. Faerch, C. and Kasper, G. (1987) Perspectives on language transfer. Applied Linguistics 8 (2), 111–136. Fernando, C. (1996) Idioms and Idiomaticity. Oxford: Oxford University Press. Fernando, C. and Flavell, R. (1981) On Idiom: Critical Views and Perspectives. Exeter: University of Exeter. Ferrer i Cancho, R. and Solé, R.V. (2001) The small world of human language. Proceedings of the Royal Society of London B 268, 2261–2265. Ferrer i Cancho, R., Solé, R.V. and Köhler, R. (2004) Patterns in syntactic dependency networks. Physical Review E69, 0519151-0519-158. Fillmore, C. (1975) An alternative to checklist theories of meaning. In Proceedings of the First Annual Meeting of the Berkeley Linguistics Society (pp. 123–131). Berkeley, CA: Berkeley Linguistics Society. Flores d’Arcais, G.B. (1993) The comprehension and semantic interpretation of idioms. In C. Cacciari and P. Tabossi (eds) Idioms: Processing, Structure, and Interpretation (pp. 79–99). Hillsdale, NJ: Lawrence Erlbaum. Foss, D. (1970) Some effects of ambiguity upon sentence comprehension. Journal of Verbal Learning and Verbal Behavior 9, 699–706. Francis, W.N. and Kučera, H. (1982) Frequency Analysis of English Usage. Boston, MA: Houghton Mifflin Company. Fraser, B. (1976) The Verb-Particle Construction in English. New York: Academic Press. Frauenfelder, U.H. and Schreuder, R. (1992) Constraining psycholinguistic models of morphological processing and representation: The role of productivity. In G. Booij and J. van Marle (eds) Yearbook of Morphology 1991 (pp. 165–183). Dordrecht: Kluwer.
276 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Freyd, P. and Baron, J. (1982) Individual differences in acquisition of derivational morphology. Journal of Verbal Learning and Verbal Behavior 21, 282–295. Gardner, D. (2007) Validating the construct of word in applied corpus-based vocabulary research: A critical survey. Applied Linguistics 28 (2), 241–265. Gardner, D. and Davies, M. (2007) Pointing out frequent phrasal verbs: A corpus-based analysis. TESOL Quarterly 41 (2), 339–359. Gerganov, E. and Taseva-Rangelova, K. (1982) The impact of the factors ‘associative value’ and ‘number of syllables’ of English lexical items on word memorization in teaching English to Bulgarian students. Contrastive Linguistics 7 (4), 3–12. Gibbs, R.W. (1980) Spilling the beans on understanding and memory for idioms in conversations. Memory and Cognition 8, 148–164. Gibbs, R.W. (1994) The Poetics of Mind: Figurative Thought, Language, and Understanding. Cambridge: Cambridge University Press. Gibbs, R.W., Nayak, N. and Cutting, C. (1989) How to kick the bucket and not decompose: Analyzability and idiom processing. Journal of Memory and Language 28, 576–593. Goulden, R., Nation, P. and Read, J. (1990) How large can a receptive vocabulary be? Applied Linguistics 11 (4), 341–363. Granger, S. (1999) Use of tenses by advanced EFT learners: Evidence from an error-tagged computer corpus. In H. Hasselgård and S. Oksefjell (eds) Out of Corpora: Studies in Honour of Stig Johansson (pp. 191–202). Amsterdam: Rodopi. Granger, S. (2003) The international corpus of learner English: A new resource for foreign language learning and teaching and second language acquisition research. TESOL Quarterly 37 (3), 538–546. Grant, L.E. and Bauer, L. (2004) Criteria for re-defining idioms: Are we barking up the wrong tree? Applied Linguistics 25, 38–61. Greidanus, T. and Nienhuis, L. (2001) Testing the quality of word knowledge in L2 by means of word associations: Types of distracters and types of associations. The Modern Language Journal 85, 567–577. Greidanus, T., Bogaards, P., Van Der Linden, E., Nienhuis, L. and De Wolf, T. (2004) The construction and validation of a deep word knowledge test for advanced learners of French. In P. Bogaards and B. Laufer (eds) Vocabulary in a Second Language (pp. 191–208). Amsterdam: John Benjamins. Greidanus, T., Beks, B. and Wakely, R. (2005) Testing the development of French word knowledge by advanced Dutch- and English-speaking learners and native speakers. The Modern Language Journal 89 (2), 221–238. Griffin, G. and Harley, T.A. (1996) List learning of second language vocabulary. Applied Psycholinguistics 17, 443–460. Gu, Y. and Johnson, R.K. (1996) Vocabulary learning strategies and language learning outcomes. Language Learning 46, 643–679. Gustawsson, E. (2006) Idioms Unlimited: A Study of Non-Canonical Forms of English Verbal Idioms in the British National Corpus. Gothenburg: University of Gothenburg. Gyori, G. (2002) Semantic change and cognition. Cognitive Linguistics 13 (2), 123–166. Haastrup, K. (1991) Lexical Inferencing Procedures or Talking about Words: Receptive Procedures in Foreign Language Learning with Special Reference to English. Tübingen: Gunter Narr. Haastrup, K. and Henriksen, B. (2000) Vocabulary acquisition: Acquiring depth of knowledge through network building. International Journal of Applied Linguistics 10 (2), 221–239. Halliday, M.A.K. (1994) An Introduction to Functional Grammar (2nd edn). London: Edward Arnold. Hallström, A. and Östberg, U. (2001) Svår grammatik och ordbildning i svenska som främmande språk. Lund: Studentlitteratur. Hallström, A. and Östberg, U. (2003) De rätta orden: Svenska betydelsefält. Lund: Studentlitteratur.
References 277
Hancin-Bhatt, B. and Nagy, W. (1994) Lexical transfer and second language morphological development. Applied Psycholinguistics 15, 289–310. Hargevik, S. (1996) Hargeviks engelska grammatik. Stockholm: Almqvist & Wiksell. Hargevik, S. (1998) Hargeviks 5000 ord. Gothenburg: Utbildningsstaden AB. Hargevik, S. and Stevens, M. (1978) English Synonyms and False Friends. Lund: Liber Läromedel. Harley, B. and Hart, D. (2000) Vocabulary learning in the content-oriented second-language classroom: Student perceptions and proficiency. Language Awareness 9 (2), 78–96. Harwood, F.W. and Wright, A.M. (1956) Statistical study of English word formation. Language 32, 260–273. Haspelmath, M. and Sims, A.D. (2010) Understanding Morphology (2nd edn). London: Hodder Education. Hatch, E.M. (1983) Psycholinguistics: A Second Language Perspective. Rowley, MA: Newbury House. Hay, J. and Baayen, R.H. (2002) Parsing and productivity. In G. Booij and J. van Marle (eds) Yearbook of Morphology 2001 (pp. 218–236). Dordrecht: Kluwer. Hazenberg, S. and Hulstijn, J. (1996) Defining a minimal receptive second-language vocabulary for non-native university students: An empirical investigation. Applied Linguistics 17, 145–163. Hebbe, A. (1999) Bygg upp ert ordförråd med roliga övningar och enkla prov. Stockholm: Natur & Kultur. Hebbe, A. and Ek, K. (2005) Ordglädje: Bygg vidare på ditt ordförråd. Stockholm: Natur & Kultur. Herman, P.A., Anderson, R.C., Pearson, P.D. and Nagy, W.E. (1987) Incidental acquisition of word meaning from expositions with varied text features. Reading Research Quarterly 22, 263–284. Herskovits, A. (1986) Language and Spatial Cognition: An Interdisciplinary Study of the Prepositions in English. Cambridge: Cambridge University Press. Higa, M. (1965) The psycholinguistic concept of ‘difficulty’ and the teaching of foreign language vocabulary. Language Learning 15, 167–179. Hill, R.J. (1982) A Dictionary of False Friends. London: Macmillan. Hinkel, E. (2012, February) Language teaching and construction grammar. Applied Linguistics Forum: ALIS Newsletter. Holmegaard, M. (2000) Språkmedvetenhet och ordinlärning – Lärare och inlärare reflekterar kring en betydelsefältsövning i svenska som andraspråk. In H. Åhl (ed.) Svenskan i tiden: Verklighet och visioner (pp. 109–120). Stockholm: Stockholms Universitets Förlag. Holmes, P. and Hinchliffe, I. (1994) Swedish: A Comprehensive Grammar. London: Routledge. Honeck, R.P. and Temple, J.G. (1994) Proverbs: The extended conceptual base and great chain metaphor theories. Metaphor and Symbolic Activity 9 (2), 85–112. Horst, M., Cobb, T. and Meara, P. (1998) Beyond A Clockwork Orange: Acquiring second language vocabulary through reading. Reading in a Foreign Language 11, 207–223. Huddleston, R. and Pullum, G.K. (2002) The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. Hultfors, P. (1986) Reactions to Non-Native English: Native English-Speakers’ Assessments of Errors in the Use of English Made by Non-Native Users of the Language. Part 1. Acceptability and Intelligibility. Stockholm: Almqvist & Wiksell. Irujo, S. (1986) Don’t put your leg in your mouth: Transfer in the acquisition of idioms in a second language. TESOL Quarterly 20, 287–304. Jansson, U. (2006) Svenska ordspråk, uttryck och talesätt. Stockholm: Liber. Jenkins, J.R. and Dixon, R. (1983) Vocabulary learning. Contemporary Educational Psychology 8, 237–260.
278 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Jiang, N. (2004) Semantic transfer and development in adult L2 vocabulary acquisition. In P. Bogaards and B. Laufer (eds) Vocabulary in a Second Language (pp. 161–126). Amsterdam: John Benjamins. Johnson, M. (1987) The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason. Chicago, IL: University of Chicago Press. Jordens, P. (1977) Rules, grammatical intuitions and strategies in foreign language learning. Interlanguage Studies Bulletin 2 (2), 5–76. Karlsson, Å. (1987) Ordlista till Basic Technical English. Hörby: Bokdistribution. Karlsson, M. (2002) Progression and Regression: Aspects of Advanced Swedish Students’ Competence in English Grammar. Lund: Lund University. Karlsson, M. (2019) Idiomatic Mastery in a First and Second Language. Bristol: Multilingual Matters. Kellerman, E. (1977) Towards a characterization of the strategy of transfer in second language learning. Interlanguage Studies Bulletin 2, 58–145. Kim, H. (2003) Vocabulary Comprehension of Advanced ESL Learners in Academic Reading: A Collective Case Study. MA thesis, University of Ottawa. Koda, K. (2005) Insights Into Second Language Reading: A Cross-linguistic Approach. Cambridge: Cambridge University Press. Kövecses, Z. and Szabó, P. (1996) Idioms: A view from cognitive semantics. Applied Linguistics 17, 326–335. Kurtyka, A. (2001) Teaching English phrasal verbs: A cognitive approach. In M. Pütz, S. Niemeier and R. Dirven (eds) Applied Cognitive Linguistics (Volume II): Language Pedagogy (pp. 29–54). Berlin: Mouton de Gruyter. Lackner, J.R. and Garrett, M.F. (1972) Resolving ambiguity: Effects of biasing context in the unattended ear. Cognition 1, 359–372. Lakoff, G. (1987) Women, Fire and Dangerous Things: What Categories Reveal about the Mind. Chicago, IL: University of Chicago Press. Lakoff, G. (1990) The invariance hypothesis: Is abstract reasoning based on image schemas? Cognitive Linguistics 1, 39–74. Lakoff, G. and Johnson, M. (1980) Metaphors We Live By. Chicago, IL: University of Chicago Press. Larsen-Freeman, D. (2012, February) The emergence of lexicogrammatical patterns from use. Applied Linguistics Forum: ALIS Newsletter. Laufer, B. (1989a) What percentage of text-lexis is essential for comprehension? In C. Laurén and M. Nordmann (eds) Special Language: From Humans Thinking to Thinking Machines (pp. 316–323). Clevedon: Multilingual Matters. Laufer, B. (1989b) A factor of difficulty in vocabulary learning: Deceptive transparency. AILA Review 6, 10–20. Laufer, B. (1991) Some properties of the foreign language learner’s lexicon as evidenced by lexical confusions. International Review of Applied Linguistics in Language 29 (4), 317–330. Laufer, B. (1997a) The lexical plight in second language reading: Words you don’t know, words you think you know, and words you can’t guess. In J. Coady and T. Huckin (eds) Second Language Vocabulary Acquisition (pp. 20–34). Cambridge: Cambridge University Press. Laufer, B. (1997b) What’s in a word that makes it hard or easy: Some intralexical factors that affect the learning of words. In N. Schmitt and M. McCarthy (eds) Vocabulary: Description, Acquisition and Pedagogy (pp. 140–155). Cambridge: Cambridge University Press. Laufer, B. (1998) The development of passive and active vocabulary in a second language: Same or different? Applied Linguistics 19, 255–271. Laufer, B. (2010) Form-focused instruction in second language vocabulary learning. In R. Chacón-Beltrán, C. Abello-Contesse and M. del Mar Torreblanca-López (eds)
References 279
Insights into Non-native Vocabulary Teaching and Learning (pp. 15–27). Bristol: Multilingual Matters. Laufer, B. and Eliasson, S. (1993) What causes avoidance in L2 learning: L1-L2 difference, L1-L2 similarity, or L2 complexity? Studies in Second Language Acquisition (SSLA) 15 (1), 35–48. Laufer, B. and Sim, D. (1985) Taking the easy way out: Non-use and misuse of clues in EFL reading. English Teaching Forum 23, 7–10, 20. Lemmouh, Z. (2010) The Relationship among Vocabulary Knowledge, Academic Achievement and the Lexical Richness in Writing in Swedish University Students of English. Stockholm: Stockholm University. Leontjev, D., Huhta, A. and Mäntylä, K. (2016) Word derivational knowledge and writing proficiency: How do they link? System 59, 73–89. Liljestrand, B. (1993) Så bildas orden: Handbok i ordbildning. Lund: Studentlitteratur. Lindstromberg, S. (1997) English Prepositions Explained. Amsterdam: John Benjamins. Liontas, J. (2002) Context and idiom understanding in second languages. In EUROSLA Yearbook 2 (pp. 155–185). University of Notre Dame: John Benjamins. Liu, D. (2008) Idioms: Description, Comprehension, Acquisition, and Pedagogy. New York: Routledge. Liu, D. (2010) Is it chief, main, major, primary, or principal concern? A corpus-based behavioral profile study of the near synonyms. International Journal of Corpus Linguistics 15, 56–87. Liu, D. (2012, February) Using corpus and cognitive linguistic analysis in teaching lexis and grammar. Applied Linguistics Forum: ALIS Newsletter. Liu, D. (2013) Salience and construal in the use of synonymy: A study of two sets of near synonymous nouns. Cognitive Linguistics 24, 67–113. Liu, D. and Espino, M. (2012) Actually, genuinely, really, and truly: A corpus-based behavioral profile study of the near-synonymous adverbs. International Journal of Corpus Linguistics 17, 198–228. Liu, N. and Nation, I.S.P. (1985) Factors affecting guessing vocabulary in context. RELC Journal 16, 33–42. Liu, D. and Zhong, S. (2014) L2 vs. L1 use of synonymy: An empirical study of synonymy use/acquisition. Applied Linguistics 37 (2), 239–261. Ljung, M. (2003) Making Words in English. Lund: Studentlitteratur. Lowie, W.M. and Verspoor, M.H. (2001) Het leren van polyseme woorden vanuit de kernbetekenis. Toegepaste Taalwetenschap in Artikelen 65 (1), 9–21. Lowie, W.M. and Verspoor, M. (2004) Input versus transfer? The role of frequency and similarity in the acquisition of L2 prepositions. In M. Achard and S. Niemeier (eds) Cognitive Linguistics, Second Language Acquisition, and Foreign Language Teaching (pp. 77–94). Berlin: Mouton de Gruyter. Luthman, H. (2002) Svenska idiom: 3500 vardagsuttryck. Lund: Folkuniversitetets förlag. Luthman, H. (2006) Svenska idiom. Övningsbok (2nd edn). Lund: Folkuniversitetets förlag. Mackay, D. (1966) To end ambiguous sentences. Perception and Psychophysics 1, 426–436. MacKay, S. (1980) Teaching the syntactic, semantic, and pragmatic dimensions of verbs. TESOL Quarterly 14 (1), 17–26. Mahony, D., Singson, M. and Mann, V. (2000) Reading ability and sensitivity to morphological relations. Reading and Writing: An Interdisciplinary Journal 12, 191–218. Makkai, A. (1972) Idiom Structure in English. The Hague: Mouton de Gruyter. Malvern, D., Richards, B., Meara, P. and Milton, J. (2008) Introduction: Special issue on knowledge and use of the lexicon in French as a second language. French Language Studies 18, 269–276. Marslen-Wilson, W., Tyler, L., Waksler, R. and Older, L. (1994) Morphology and meaning in the English mental lexicon. Psychological Review 101, 3–33.
280 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Martin, M. (1984) Advanced vocabulary teaching: The problem of synonyms. The Modern Language Journal 68 (2), 130–137. Martinsson, Y., Blank, Y. and Johansson, I. (eds) (2004) Stora ordspråksboken: Svenska, engelska, tyska, franska och spanska ordspråk och talesätt. Stockholm: Norstedts ordbok. Marton, W. (1977) Foreign vocabulary learning as problem number one of foreign language teaching at the advanced level. Interlanguage Studies Bulletin 2 (1), 33–47. McCarthy, M.J. (1992) English idioms in use. Revista Canaria de Estudios Ingleses 25, 55–65. McCarthy, M.J. and Carter, R.A. (1994) Language as Discourse: Perspectives for Language Teaching. London: Longman. McElree, B. and Nordie, J. (1999) Literal and figurative interpretations are computed in equal time. Psychonomic Bulletin and Review 6, 486–494. Meara, P. (1978) Learners’ word associations in French. Interlanguage Studies Bulletin Utrecht 3, 192–211. Meara, P. (1982) Vocabulary acquisition: A neglected aspect of language learning. In V. Kinsella (ed.) Cambridge Language Teaching Surveys (pp. 221–246). Cambridge: Cambridge University Press. Meara, P. (1990) A note on passive vocabulary. Second Language Research 6, 150–154. Meara, P. (1995) Single subject studies of lexical acquisition. Second Language Research 11, i–iii. Mecartty, F. (2000) Lexical and grammatical knowledge in reading and listening comprehension by foreign language learners of Spanish. Applied Language Learning 11, 323–348. Miller, G.A. and Charles, W.G. (1991) Contextual correlates of semantic similarity. Language and Cognitive Processes 6, 1–28. Miller, G. and Fellbaum, C. (1991) Semantic networks of English. Cognition 41, 197–229. Milton, J. (2009) Measuring Second Language Vocabulary Acquisition. Bristol: Multilingual Matters. Milton, J. and Meara, P.M. (1995) How periods abroad affect vocabulary growth in a foreign language. International Journal of Applied Linguistics 107–108, 17–34. Miralpeix, I. (2007) Testing receptive vocabulary size: X_Lex and Y_Lex. Paper presented at the 29th Language Testing Research Colloquium, Barcelona. Mirhassani, A. and Toosi, A. (2000) The impact of word-formation knowledge on reading comprehension. International Review of Applied Linguistics in Language Teaching 38, 301–311. Moberg, L. (1996) Om engelskans stundom välgörande inflytande på svenskan. In Språket lever!: Festskrift till Margareta Westman den 27 mars 1996. Stockholm: Svenska språknämnden. Mochizuki, M. (1998) Understanding English affixes by Japanese learners. Reitaku Review 4, 100–120. Mochizuki, M. and Aizawa, K. (2000) An affix order of acquisition for EFL learners: An explanatory study. System 28, 291–304. Möijer, K. (1998) Svensk ordlära: Så bildas orden. Solna: Ekelund. Mondor, M. (2008) Figuring it Out: A Corpus-Based Comparison of the Verb-Particle Construction in Argumentative Writing by Swedish Advanced Learners and Native Speakers of English. Gothenburg: University of Gothenburg. Montan, P. and Rosenqvist, H. (1992) Prepositionsboken (4th edn). Stockholm: Almqvist & Wiksell. Montan, P. and Rosenqvist, H. (1993) Prepositionsövningar 2 (2nd edn). Stockholm: Skriptor. 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.
References 281
Moon, R. (1998) Fixed Expressions and Idioms in English: A Corpus-Based Approach. Oxford: Clarendon Press. Moseley, D. (1994) From theory to practice: Errors and trials. In G.D.A. Brown and N.C. Ellis (eds) Handbook of Spelling (pp. 459–479). Chichester: John Wiley and Sons. Mueller, C.M. (2011) English learners’ knowledge of prepositions: Collocational knowledge or knowledge based on meaning? System 39, 480–490. Murphy, M.L. (2003) Semantic Relations and the Lexicon: Antonymy, Synonymy and Other Paradigms. Cambridge: Cambridge University Press. Murphy, G.L. (2004) The Big Book of Concepts. Cambridge, MA: MIT Press. Nagy, W.E. (1989) Teaching Vocabulary to Improve Reading Comprehension. Newark, NJ: International Reading Association. Nagy, W.E. and Anderson, R.C. (1984) How many words are there in printed school English? Reading Research Quarterly 19, 304–330. Nagy, W.E., Anderson, R.C. and Herman, P.A. (1987) Learning word meanings from context during normal reading. American Educational Research Journal 24, 237–270. Nagy, W.E., Anderson, R.C., Schommer, M., Scott, J.A. and Stallman, A. (1989) Morphological families in the internal lexicon. Reading Research Quarterly 24, 263–282. Nagy, W.E., Diakodoy, I.N. and Anderson, R.C. (1993) The acquisition of morphology: Learning the contribution of suffixes to the meanings of derivatives. Journal of Reading Behaviour 25, 155–169. Namei, S. (2002) The bilingual lexicon from a developmental perspective: A word association study of Persian-Swedish bilinguals. Stockholm: Centre for Research on Bilingualism, Stockholm University. Namei, S. (2004) Bilingual lexical development: A Persian-Swedish word association study. International Journal of Applied Linguistics 14 (3), 363–388. Nassaji, H. (2004) The relationship between depth of vocabulary knowledge and L2 learners’ lexical inferencing strategy use and success. Canadian Modern Language Review 61 (1), 107–134. Nation, I.S.P. (1990) Teaching and Learning Vocabulary. Boston, MA: Heinle & Heinle. Nation, I.S.P. (1996) Vocabulary Lists. English Language Institute Occasional Publication No. 17. Wellington: Victoria University of Wellington. Nation, I.S.P. (2001) Learning Vocabulary in Another Language. Cambridge: Cambridge University Press. Nation, I.S.P. (2004) A study of the most frequent word families in the British National Corpus. In P. Bogaards and B. Laufer (eds) Vocabulary in a Second Language: Selection, Acquisition and Testing (pp. 3–13). Philadelphia, PA: John Benjamins. Nation, I.S.P. (2006) How large a vocabulary is needed for reading and listening? The Canadian Modern Language Review 63, 59–82. Nation, I.S.P. (2008) Teaching Vocabulary: Strategies and Techniques. Boston, MA: Heinle & Heinle. Nation, I.S.P. and Waring, R. (1997) Vocabulary size text coverage and word lists. In N. Schmitt and M. McCarthy (eds) Vocabulary: Acquisition and Pedagogy. Cambridge: Cambridge University Press. Nattinger, J. and DeCarrico, J. (1992) Lexical Phrases and Language Teaching. Oxford: Oxford University Press. New, W.H. (ed.) (1997) Canadian Short Fiction. Scarborough, Ontario: Prentice Hall. Nordlund, M. and Norberg, C. (2020) Vocabulary in EFL teaching materials for young learners. International Journal of Language Studies 14 (1), 89–116. Nurweni, A. and Read, J. (1999) The English vocabulary knowledge of Indonesian university students. English for Specific Purposes 18, 161–175. Odefalk, A. (2004) A study of Swedish students’ mastery of words and their meaning(s). Unpublished undergraduate study, Halmstad University.
282 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
O’Dell, F. (1997) Incorporating vocabulary into the syllabus. In N. Schmitt and M. McCarthy, (eds) Vocabulary: Description, Acquisition and Pedagogy (pp. 258–278). Cambridge: Cambridge University Press. Odlin, T. (2003) Cross-linguistic influence. In C.J .Doughty and M.H. Long (eds) The Handbook of Second Language Acquisition. Malden, MA: Blackwell. Olejniczak, P. (2006) Phrasal verb idioms and the normative concept of the interlanguage hypothesis. In J. Arabski (ed.) Cross-linguistic Influences in the Second Language Lexicon (pp. 259–272). Clevedon: Multilingual Matters. Ortony, A., Schallert, D.L., Reynolds, R.E. and Antos, S. (1978) Interpreting metaphors and idioms: Some effects of context on comprehension. Journal of Verbal Learning and Verbal Behavior 7, 465–477. Övergaard, G., Lundén, R. and Spånberg, S.-J. (1992) Short Fictions: Stories from the English-Speaking World of Today (2nd edn). Stockholm: Universaltryck Grafiska. Paivio, A. (1986) Mental Representations. Oxford: Oxford University Press. Palmberg, R. (1988) On lexical inferencing and language distance. Journal of Pragmatics 12, 207–214. Paribakht, T.S. (2005) The influence of first language lexicalisation on second language lexical inferencing: A study of Farsi-speaking learners of English as a foreign language. Language Learning 55, 701–748. Paribakht, T.S. and Tréville, M.-C. (2007) Lexical inference in French native-speakers and Persian speakers during reading English texts: Effects of lexicalization in the first language. Canadian Modern Language Review 63 (3), 399–428. Paribakht, T.S. and Wesche, M. (2006) Lexical inferencing in L1 and L2: Implications for learning and instruction at advanced levels. In H. Byrnes, H. Weger-Guntharp and K.A. Sprang (eds) Educating for Advanced Foreign Language Capacities: Constructs, Curriculum, Instruction, Assessment (pp. 118–135). Washington, DC: Georgetown University Press. Pawley, A. and Syder, F. (1983) Two puzzles for linguistic theory: Nativelike selection and nativelike fluency. In J. Richards and R. Schmidt (eds) Language and Communication (pp.191–226). London: Longman. Piaget, J. and Inhelder, B. (1967) The Child’s Conception of Space (trans. F.J Langdon and J.L. Lunzer). New York: W.W. Norton and Company. Piasecka, L. (2006) ‘Don’t lose your head’ or how Polish learners of English cope with L2 idiomatic expressions. In J. Arabski (ed.) Cross-linguistic Influences in the Second Language Lexicon (pp. 246–258). Clevedon: Multilingual Matters. Piper, T.H. and Leicester, P.F. (1980) Word association behavior as an indicator of English language proficiency. Educational Resources Information Center (ERIC) Documents, ED227651. Plag, I. (1999) Morphological Productivity: Structural Constraints in English Derivation. Berlin: Mouton de Gruyter. Pollio, H.R., Barlow, J.M., Fine, H.J. and Pollio, M.R. (1977) Psychology and the Poetics of Growth. Hillsdale, NJ: Lawrence Erlbaum. Qian, D. (2004) Second language lexical inferencing: Preferences, perceptions, and practices. In P. Bogaards and B. Laufer (eds) Vocabulary in a Second Language: Selection, Acquisition, and Testing (pp. 155–169). Amsterdam: John Benjamins. Quian, D.D. (1999) Assessing the roles of depth and breadth of vocabulary knowledge in reading comprehension. Canadian Modern Language Review 56, 282–307. Quian, D.D. (2002) Investigating the relationship between knowledge and academic reading performance: An assessment perspective. Language Learning 52, 513–536. Quirk, R., Greenbaum, S., Leech, G. and Svartvik, J. (1985) A Comprehensive Grammar of the English Language. London: Longman. Raffelsiefen, R. (1996) Gaps in word-formation. In U. Kleinhenz (ed.) Interfaces in Phonology 1996 (pp. 194–209). Berlin: Akademie Verlag.
References 283
Raffelsiefen, R. (1999) Phonological constraints on English word formation. In G. Booij and J. van Marle (eds) Yearbook of Morphology 1998 (pp. 225–287). Dordrecht: Kluwer. Read, J. (1993) The development of a new measure of L2 vocabulary knowledge. Language Testing 10, 355–371. Richards, J. (1976) The role of vocabulary teaching. TESOL Quarterly 10 (1), 77–89. Rijpma, M.T. (1999) The effectiveness of the learning strategy of core meaning on guessing and retention of second language polysemous words. Unpublished MA thesis, University of Groningen. Ringbom, H. (1987) The Role of the First Language in Foreign Language Learning. Clevedon: Multilingual Matters. Rodgers, T.S. (1969) On measuring vocabulary difficulty: An analysis of item variables in learning Russian-English vocabulary pairs. International Review of Applied Linguistics 7, 327–343. Rosch, E. and Mervis, C. (1975) Family resemblances: Studies in the internal structures of categories. Cognitive Psychology 7, 573–605. Rudzka, B., Channel, J., Pudseys, Y. and Ostyn, P. (1981) The Words You Need. London: Macmillan. Rudzka-Ostyn, B. (2003) Word Power: Phrasal Verbs and Compounds. Berlin: Mouton de Gruyter. Rye, J. (1985) Are cloze items sensitive to constraints across sentences? A review. Journal of Research in Reading (UKRA) 8, 94–105. Sager, O. and Ormelius, S.-E. (1983) Engelsk ordbildning. Solna: Esselte studium. Sandra, D. (1990) On the representation and processing of compound words: Automatic access to constituent morphemes does not occur. Quarterly Journal of Experimental Psychology 42A, 529–567. Schmidt, L. (2004) A study of upper secondary students’ mastery of English prepositions. Unpublished undergraduate study, Halmstad University. Schmitt, N. (1997) Vocabulary learning strategies. In N. Schmitt and M. McCarthy (eds) Vocabulary: Description, Acquisition and Pedagogy (pp. 199–227). Cambridge: Cambridge University Press. Schmitt, N. (1998) Tracking the incremental acquisition of second language vocabulary: A longitudinal study. Language Learning 48 (2), 281–317. Schmitt, N. (1999) The relationship between TOEFL vocabulary items and meaning, association, collocation, and word class knowledge. Language Testing 16, 189–216. Schmitt, N. (2008) Instructed second language vocabulary learning. Language Teaching Research 12 (3), 329–363. Schmitt, N. (2010a) Key issues in teaching and learning vocabulary. In R. Chacón-Beltrán, C. Abello-Contesse and M. del Mar Torreblanca-López (eds) Insights into Non-native Vocabulary Teaching and Learning (pp. 28–40). Bristol: Multilingual Matters. Schmitt, N. (2010b) Researching Vocabulary: A Vocabulary Research Manual. Basingstoke: Palgrave Macmillan. Schmitt, N. (2010c) Vocabulary in Language Teaching. Cambridge: Cambridge University Press. Schmitt, N. and Dunham, B. (1999) Exploring native and non-native intuitions of word frequency. Second Language Research 15 (4), 389–411. Schmitt, N. and Marsden, R. (2006) Why Is English Like That? Historical Answers to Hard ELT Questions. Ann Arbor, MI: University of Michigan Press. Schmitt, N. and McCarthy, M. (eds) (1997) Vocabulary: Description, Acquisition and Pedagogy. Cambridge: Cambridge University Press. Schmitt, N. and Meara, P. (1997) Research vocabulary through a word knowledge framework – Word associations and verbal suffixes. Studies in Second Language Acquisition 19, 17–36.
284 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Schmitt, N. and Zimmerman, C.B. (2002) Derivative word forms: What do learners know? TESOL Quarterly 36 (2), 145–171. Schneider, V.I., Healy, A.F. and Bourne Jr., L.E. (2002) What is learned under difficult conditions is hard to forget: Contextual interference effects in foreign vocabulary acquisition, retention and transfer. Journal of Memory and Language 46, 419–440. Schuyten, M.C. (1906) Experimentelles zum studien der gebräuchlichsten methoden im fremdsprachlichen unterricht. Experimentelle Pädagogik 3, 199–210. Shefelbine, J.L. (1990) Student factors related to variability in learning word meanings from context. Journal of Reading Behavior 22, 71–97. Shillaw, J. (1995) Using a word list as a focus for vocabulary learning. The Language Teacher 19, 58–59. Shioji, J. (1982) Affirmative approaches to vocabulary development. TESOL Convention, Honolulu, HI. Sinclair, J. (1991) Corpus, Concordance and Collocation. Oxford: Oxford University Press. Singleton, D. (1999) Exploring the Second Language Mental Lexicon. Cambridge: Cambridge University Press. Singson, M., Mahony, D. and Mann, V. (2000) The relation between reading ability and morphological skills: Evidence from derivational suffixes. Reading and Writing: An Interdisciplinary Journal 12, 219–252. Sjöholm, K. (1995) The Influence of Cross-Linguistic, Semantic and Input Factors on the Acquisition of English Phrasal Verbs: A Comparison between Finnish and Swedish Learners at an Intermediate and Advanced Level. Åbo: Åbo Akademi University Press. Skoufaki, S. (2008) Conceptual metaphoric meaning clues in two idiom presentation methods. In F. Boers and S. Lindstromberg (eds) Cognitive Linguistic Approaches to Teaching Vocabulary and Phraseology (pp. 101–132). Berlin: Mouton de Gruyter. Smedts, W. (1988) De beheersing van de nederlandse woordvorming tussen 7 en 17. In F. van Besien (ed.) First Language Acquisition (pp. 103–127). ABLA Papers 12. Antwerp: Association Belge de Linguistique Appliquée/Universitaire Instelling Antwerpen. Stahl, S.A. and Fairbanks, M.M. (1986) The effects of vocabulary instruction: A modelbased meta-analysis. Review of Educational Research 56, 72–110. Stahl, S.A. and Shiel, T.G. (1992) Teaching meaning vocabulary: Productive approaches for poor readers. Reading and Writing Quarterly: Overcoming Learning Difficulties 8, 223–241. Stein, B.S., Bransford, J.D., Franks, J.J., Owings, R.A., Vye, N.J. and McGraw, W. (1982) Differences in the precision of self-generated elaborations. Journal of Experimental Psychology: General 111, 399–405. Sternberg, R.J. (1987) Most vocabulary is learned from context. In M.G. McKeown and M.E. Curtis (eds) The Nature of Vocabulary Acquisition (pp. 89–105). Hillside, NJ: Lawrence Erlbaum. Stock, R.D. (1976) Some factors affecting the acquisition of foreign language lexicon in the classroom. Unpublished PhD thesis, University of Illinois. Stoddard, G.D. (1929) An experiment in verbal learning. Journal of Educational Psychology 20, 452–457. Strömqvist, S. (2009) Språkets öga: Om vägarna mellan tankar och ord. Lund: Studentlitteratur. Süskind, P. (1987) Parfymen. Berättelsen om en mördare (trans. U. Wallenström). Stockholm: Wahlström & Widstrand. Sutarsyah, C., Nation, P. and Kennedy, G. (1994) How useful is EAP vocabulary for ESP? A corpus-based study. RELC Journal 25, 34–50.
References 285
Svartvik, J. and Sager, O. (1988) Modern engelsk grammatik (3rd edn). Uppsala: Almqvist & Wiksell. Svartvik, J. and Sager, O. (1996) Engelsk universitetsgrammatik (2nd edn). Stockholm: Almqvist & Wiksell. Swan, M. and Smith, B. (eds) (2001) Learner English: A Teacher’s Guide to Interference and other Problems (2nd edn). Cambridge: Cambridge University Press. Swanborn, E. and de Glopper, K. (1999) Incidental word learning while reading: A metaanalysis. Review of Educational Research 69, 261–285. Swinney, D. (1979) Lexical access during sentence comprehension: (Re)consideration of context effects. Journal of Verbal Learning and Verbal Behavior 18, 645–659. Swinney, D.A. and Cutler, A. (1979) The access and processing of idiomatic expressions. Journal of Verbal Learning and Verbal Behavior 18, 523–534. Söderman, T. (1993) Word associations of foreign language learners and native speakers: The phenomenon of a shift in response type and its relevance for lexical development. In H. Ringbom (ed.) Near Native Proficiency in English (pp. 91–182). Åbo: Åbo Akademi University Press. Söderström, F., Thorén, J. and Karlsson, M. (2022) Pun intended: The use of word play to teach polysemous words in English as a second language. International Journal of Language Studies 16 (2), 95–116. Tabossi, P. and Zardon, F. (1993) The activation of idiomatic meaning in spoken language comprehension. In C. Cacciari and P. Tabossi (eds) Idioms: Processing, Structure, and Interpretation (pp. 145–162). Hillsdale, NJ: Lawrence Erlbaum. Tabossi, P. and Zardon, F. (1995) The activation of idiomatic meaning. In E. Everaert, E.J. van der Linden, A. Schenk and R. Schreuder (eds) Idioms: Structural and Psychological Perspective (pp. 273–282). Hillsdale, NJ: Lawrence Erlbaum. Thagg Fischer, U. (1985) The Sweet Sound of Concord: A Study of Swedish Learners’ Concord Problems in English. Malmö: Liber. Thorndike, E.L. (1941) The Teaching of English Suffixes. New York: Columbia University Teachers College. Titone, D.A. and Connine, C.M. (1994) Descriptive norms for 171 idiomatic expressions: Familiarity, compositionality, predictability, and literality. Metaphor and Symbolic Activity 9, 247–270. Titone, D.A. and Connine, C.M. (1999) On the compositional and noncompositional nature of idiomatic expressions. Journal of Pragmatics 31, 1655–1674. Trulsson, A. (2007) Ursäkta, men skulle du kunna göra mig en björntjänst? Om idioms betydelser och deras fortlevnad i svenskan. Unpublished undergraduate study, Uppsala University. Tyler, A. and Nagy, W. (1989) The acquisition of English derivational morphology. Journal of Memory and Language 28, 649–667. Tyler, A. and Nagy, W. (1990) Use of derivational morphology during reading. Cognition 36, 17–34. Verspoor, M.H. (1997) True blue: A cognitive linguistic approach to vocabulary acquisition. In B. Smieja and M. Tasch (eds) Human Contact through Language and Linguistics (pp. 203–217). Frankfurt: Peter Lang. Verspoor, M.H. and Lowie, W. (2003) Making sense of polysemous words. Language Learning 53 (3), 547–586. Viberg, Å. (1993) Andraspråksinlärning i olika åldrar. In E. Cerú (ed.) Svenska som andraspråk: Lärarbok. Mera om språket och inlärningen (pp. 13–83). Stockholm: Utbildningsradion/Natur & Kultur. Viberg, Å. (2000) Svenskt ordnät – lexikon på dator som model för ordförrådet i hjärnan hosinfödda talare och andraspråkstalare. In H. Åhl (ed.) Svenskan i tiden: Verklighet och visioner (pp. 287–305). Stockholm: Stockholms Universitets Förlag. Wallace Robinett, B. (1978) Teaching English to Speakers of Other Languages. New York: McGraw-Hill.
286 Advanced Students’ Knowledge of Vocabulary in a First and Second Language
Wallin, G. and Jaginder, J. (2021) Swedish EFL students’ awareness of connotations: A quantitative and qualitative study on students’ awareness of connotations. Halmstad: Halmstad University. Waring, R. (1997) A study of receptive and productive vocabulary learning from word cards. Studies in Foreign Languages and Literature (Notre Dame Seishin University) 21 (1), 94–114. Webb, S. (2008) The effects of context on incidental vocabulary learning. Reading in a Foreign Language 20, 232–245. Weinius, K. and Whitlam, J. (1997) Advanced Vocabulary Check: Aktiv ordkunskap för gymnasiets kurs B studieförberedande program. Strängnäs: Appia Publishing. Wesche, M.B. and Paribakht, T.S. (2009) Lexical Inferencing in a First and Second Language: Cross-linguistic Dimensions. Bristol: Multilingual Matters. West, M. (1953) A General Service List of English Words. London: Longman. White, T.G., Power, M.A. and White, S. (1989) Morphological analysis: Implications for teaching and understanding vocabulary growth. Reading Research Quarterly 24, 283–304. Wilks, C. and Meara, P. (2002) Untangling word webs: Graph theory and the notion of density in second language word association networks. Second Language Research 18, 303–324. Wolter, B. (2001) Comparing the L1 and L2 mental lexicon: A depth of individual word knowledge model. Studies in Second Language Acquisition 23, 41–69. Wolter, B. (2006) Lexical network structures and L2 vocabulary acquisition: The role of L1 lexical/conceptual knowledge. Applied Linguistics 27, 741–747. Wray, A. (2002) Formulaic Language and the Lexicon. Cambridge: Cambridge University Press. Wysocki, K. and Jenkins, J. (1987) Deriving word meanings through morphological generalizations. Reading Research Quarterly 22, 66–81. Xue, G. and Nation, P. (1984) A university word list. Language Learning and Communication 3, 215–229. Yule, G. (1996) The Study of Language. Cambridge: Cambridge University Press. Zechmeister, E.B., Chronis, A.M. Cull, W.L., D’Anna, C.A. and Healy, N.A. (1995) Growth of a functionally important lexicon. Journal of Reading Behavior 27, 201–212. Zhang, D. (2017) Derivational morphology in reading comprehension of Chinesespeaking learners of English: A longitudinal structural equation modeling study. Applied Linguistics 38 (6), 871–895. Zimmerman, R. (2006) Metaphorical transferability. In J. Arabski (ed.) Cross-linguistic Influences in the Second Language Lexicon (pp. 193–209). Clevedon: Multilingual Matters. Zipf, G.K. (1945) The meaning-frequency relationship of words. Journal of General Psychology 33, 251–256. Zwitserlood, P. (1994) The role of semantic transparency in the processing and representation of Dutch compounds. Language and Cognitive Processes 9, 341–368.
Dictionaries Cambridge International Dictionary of English (1995) Cambridge: Cambridge University Press. Collins Cobuild Idioms Dictionary (2nd edn) (2002) London: HarperCollins. The Concise Oxford Dictionary (1985) Oxford: Oxford University Press. Longman Dictionary of Contemporary English (1987) Harlow: Longman. The New Oxford Dictionary of English (1998) Oxford: Clarendon. Norstedts engelska ordbok: Engelsk-svensk/svensk-engelsk. CD-Rom version. Stockholm: Norstedts Förlag AB.
References 287
Norstedts stora engelsk-svenska ordbok (1993) Stockholm: Norstedts Förlag AB. Oxford Dictionary of Phrasal Verbs (1993) Oxford: Oxford University Press. Oxford English Dictionary, OED Online (2000–) http://dictionary.oed.com. Oxford: Oxford University Press. Strömberg, A. (1979) Stora synonymordboken. Stockholm: Strömberg. Strömberg, A. (1998) Stora synonymordboken (2nd edn). Stockholm: Strömberg. Webster’s Third New International Dictionary of the English Language (1964) Springfield, MA: G. and C. Merriam Co.
Newspaper and Magazine Articles Gibbs, N. (2008) Affirmative action for boys. Time, 3 April. Huus, K. (1993) Call of the wild. Newsweek, 17 August. Jaroff, L. (1991) It happens in the best circles. Time, 23 September. Unknown author (1993) Mayhem in Miami. Newsweek, 19 April.
Websites British National Corpus, University of Oxford, http://www.natcorp.ox.ac.uk. Svenska Akademiens ordlista (SAOL) Svenska Akademien, http://www.svenskaakademien.se/svenska_spraket/svenska_akademiens_ordlista/saol_pa_natet/ordlista. Språkbanken, University of Gothenburg, http://spraakbanken.gu.se. Utbildningsdepartementet, Regeringskansliet, http://www.regeringen.se/utbildning.
Author Index
Abdullah and Jackson (1998), 108 Abel (2003), 105, 106, 107, 249 Adolphs and Schmitt (2003), 14 Aitchison (1987), 12, 141, 178, 179, 180 Aitchison (1994), 208 Aitchison (2003), 181, 182 Albrechtsen, Haastrup and Henriksen (2008), 10, 11, 12, 231, 232, 233, 241, 242, 245, 246, 250 Alderson (2005), 2 Alderson, Clapham and Steel (1997), 78 Al-Duleimi and Aziz (2016), 189 Allén, Gellerstam and Malmgren (1989), 13 Alm-Arvius (1998), 181 Alm-Arvius (2003), 179 Alstertun (1996a), 17 Alstertun (1996b), 17 Anderson (1983), 189 Anderson (1990), 189 Andersson (1987), 72, 98 Ard and Homburg (1983), 216 Artley (1943), 228 Bahns and Eldaw (1993), 10 Barrow, Nakashimi and Ishino (1999), 14 Bauer and Nation (1993), 64, 65, 66, 67, 68, 69, 84, 86, 87, 89 Becker, Dixon and Anderson-Inman (1980), 58 Bell (2009), 2 Bensoussan and Laufer (1984), 43, 183, 227 Beréndi, Csábi and Kövecses (2008), 113 Berko (1958), 58, 64 Biber, Johansson, Leech, Conrad and Finegan (1999), 99 Biber, Johansson, Leech, Conrad and Finegan (2021), 141 Bird (1987), 57 Bird (1990), 57
Blum and Levenston (1978), 43 Bobrow and Bell (1973), 102 Boers (2000), 152 Boers and Demecheleer (2001), 113 Boers, Demecheleer and Eyckmans (2004), 11, 108, 113, 114, 115, 123 Bogaards and Laufer-Dvorkin (2004), 15, 247 Bolinger (1971), 141 Boyd Zimmerman (1997), 1 Bradley (1979), 60 Brown and McNeill (1966), 180 Bywater (1969), 138 Cacciari and Levorato (1989), 105 Cacciari and Tabossi (1988), 104 Caramazza, Laudanna and Romani (1988), 59, 105 Carlisle (2000), 57, 96 Carlo, August, McLaughlin, Snow, Dressler, Lippman, Lively and White (2004), 96 Carnine, Kameeniu and Coyle (1984), 228 Cederberg (1993), 73, 74 Cederholm and Danell (2007), 234, 236 Celce-Murcia (1978), 42 Celce-Murcia and Larsen-Freeman (1999), 2, 139 Cermak and Craik (1979), 112 Chacón-Beltrán, Abello-Contesse and del Mar Torreblanca-López (2010), 1, 15 Channell (1981), 225 Chihara, Oller, Weaver and Chavez-Oller (1977), 228 Church, Gale, Hanks, Hindle and Moon (1994), 204 Cieślicka (2006), 109, 110, 111, 123, 128 Cignoni and Coffey (2000), 102 Clark and Paivio (1991), 112 Coady and Huckin (1997), 2 288
Author Index 289
Cohen and Aphek (1981), 10, 181 Coleman and Kay (1981), 139 Condon (2008), 152, 153, 154, 155, 158 Condon and Kelly (2002), 152 Cooper (1998), 99 Cooper (1999), 107 Craik and Lockhart (1972), 112 Crossley, Salsbury and McNamara (2010), 179, 182, 184, 185 Cunningham (1998), 57, 96 Dagut and Laufer (1985), 43, 138 Daneman and Green (1986), 228 Dupuy (1974), 13 Edegran and Lönneborg (2995), 118 Ellis (2003), 144 Ellis and Beaton (1993), 42 Engström (1995), 30 Erman and Warren (2000), 140 Faerch and Kasper (1987), 217 Fernando (1996), 141 Fernando and Flavell (1981), 101 Ferrer i Cancho and Solé (2001), 179 Ferrer i Cancho, Solé and Köhler (2004), 179 Fillmore (1975), 139 Flores d’Arcais (1993), 104, 105 Foss (1970), 179 Francis and Kučera (1982), 29 Fraser (1976), 142, 143 Frauenfelder and Schreuder (1992), 106 Freyd and Baron (1982), 67 Gardner (2007), 59, 67, 97, 119, 159 Gardner and Davies (2007), 139 Gerganov and Taseva-Rangelova (1982), 42 Gibbs (1980), 102, 103 Gibbs (1994), 112 Gibbs, Nayak and Cutting (1989), 104, 105, 112 Goulden, Nation and Read (1990), 13 Granger (1999), 139 Granger (2003), 139 Grant and Bauer (2004), 141 Greidanus and Nienhuis (2001), 12 Greidanus, Bogaards, Van Der Linden, Nienhuis and De Wolf (2004), 12 Greidanus, Beks, and Wakely (2005), 12
Griffin and Harley (1996), 15, 16 Gu and Johnson (1996), 230 Gustawsson (2006), 101 Gyori (2002), 43, 181 Haastrup (1991), 1, 216, 217, 218, 227 Haastrup and Henriksen (2000), 179, 206, 208, 214 Halliday (1994), 2 Hallström and Östberg (2001), 74, 156 Hallström and Östberg (2003), 208 Hancin-Bhatt and Nagy (1994), 89 Hargevik (1996), 176 Hargevik (1998), 14, 185 Hargevik and Stevens (1978), 208, 218 Harley and Hart (2002), 229 Harwood and Wright (1956), 58 Haspelmath and Sims (2010), 63, 70, 71, 72, 98 Hatch (1983), 217 Hay and Baayen (2002), 63 Hazenberg and Hultstijn (1996), 15 Hebbe (1999), 30, 44 Hebbe and Ek (2005), 44 Herman, Anderson, Pearson and Nagy (1987), 233 Herskovits (1986), 139 Higa (1965), 42 Hill (1982), 216 Hinkel (2012), 175 Holmegaard (2000), 225 Holmes and Hinchcliffe (1994), 156 Honeck and Temple (1994), 105 Horst, Cobb and Meara (1998), 227 Huddleston and Pullum (2002), 142 Hultfors (1986), 137, 143, 145 Irujo (1986), 108 Jansson (2006), 117 Jenkins and Dixon (1983), 227, 228 Jiang (2004), 180 Johnson (1987), 112 Jordens (1977), 217 Karlsson (1987), 30 Karlsson (2002), 12, 136, 137, 139, 140, 141, 143, 145, 149, 160, 176 Karlsson (2019), 5, 100, 109, 112, 116, 129, 130, 133, 134, 249 Kellerman (1977), 217
290 Language Learning and Forced Migration
Kim (2003), 227 Koda (2005), 231 Kövecses and Szabó (1996), 152, 154 Kurtyka (2001), 152 Lackner and Garrett (1972), 179 Lakoff (1987), 113, 182 Lakoff (1990), 113 Lakoff and Johnson (1980), 106 Larsen-Freeman (2012), 1, 2 Laufer (1989a), 227 Laufer (1989b), 42 Laufer (1991), 42, 43 Laufer (1997a), 42 Laufer (1997b), 41, 43, 233 Laufer (1998), 14 Laufer (2010), 13 Laufer and Eliasson (1993), 138 Laufer and Sim (1985), 227 Lemmouh (2010), 2 Leontjev, Huhta and Mäntylä (2016), 96 Liljestrand (1993), 70, 71, 73, 97 Lindstromberg (1997), 139, 140, 175 Liontas (2002), 109 Liu (2008), 99, 100, 101, 104, 107, 108, 119, 137, 141, 142 Liu (2010), 204 Liu (2012), 2 Liu (2013), 204 Liu and Espino (2012), 204 Liu and Nation (1985), 227, 242 Liu and Zhong (2014), 206, 214 Ljung (2003), 64, 65, 70, 71, 86, 88 Lowie and Verspoor (2001), 190 Lowie and Verspoor (2004), 144 Luthman (2002), 117, 157 Luthman (2006), 117 Mackay (1966), 179 Mackay (1980), 225 Mahony, Singson and Mann (2000), 96 Makkai (1972), 141 Malvern, Richards, Meara and Milton (2008), 2 Marslen-Wilson, Tyler, Waksler and Older (1994), 59, 84 Martin (1984), 1, 118, 168, 178, 180, 203, 204, 225 Martinsson, Blank and Johansson (2004), 118 Marton (1977), 43
McCarthy (1992), 100 McCarthy and Carter (1994), 100 McElree and Nordie (1999), 103 Meara (1978), 10, 181, 217 Meara (1982), 181 Meara (1990), 12 Meara (1995), 2 Mecartty (2000), 179 Miller and Charles (1991), 204 Miller and Fellbaum (1991), 206 Milton (2009), 2, 8, 14, 205 Milton and Meara (1995), 14 Miralpeix (2007), 14 Mirhassani and Toosi (2000), 96 Moberg (1996), 112 Mochizuki (1998), 68 Mochizuki and Aizawa (2000), 68, 69, 95 Möijer (1998), 73 Mondor (2008), 137, 138, 142, 146, 147, 148, 169, 176 Montan and Rosenqvist (1992), 156 Montan and Rosenqvist (1993), 156 Moon (1997), 99, 100, 101, 139, 141, 147, 153 Moon (1998), 100, 101, 135, 141 Moseley (1994), 42 Mueller (2011), 143, 144, 145, 162, 168 Murphy (2003), 11 Murphy (2004), 43, 181 Nagy (1989), 225 Nagy and Anderson (1984), 57 Nagy, Anderson and Herman (1987), 228 Nagy, Anderson, Schommer, Scott, and Stallman (1989), 58, 59, 60, 62, 94 Nagy, Diakodoy and Anderson (1993), 57, 58, 96 Namei (2002), 10, 11 Namei (2004), 10, 11 Nassaji (2004), 231 Nation (1990), 9, 13, 183 Nation (1996), 68, 69 Nation (2001), 8, 9, 13, 14, 29, 30, 33, 43, 57, 58, 65, 168, 189, 194, 205, 227, 228, 229, 231, 242, 246 Nation (2004), 8 Nation (2006), 14, 15, 205 Nation (2008), 43 Nation and Waring (1997), 8, 13, 14 Nattinger and DeCarrico (1992), 2 New (1997), 236
Author Index 291
Nordlund and Norberg (2020), 5, 55, 250 Norstedts engelska ordbok: Engelsksvensk/svensk-engelsk. CD-Rom version. 191, 192 Norstedts stora engelsk-svenska ordbok (1993), 192 Nurweni and Read (1999), 14 Odefalk (2004), 185, 186, 187, 188, 191, 182, 194, 195, 196, 201 O’Dell (1997), 1 Odlin (2003), 231 Olejniczak (2006), 146, 147, 149, 161 Ortony, Schallert, Reynolds and Antos (1978), 103 Övergaard, Lundén and Spånberg (1992), 236 Paivio (1986), 112 Palmberg (1988), 227 Paribakht (2005), 231, 233 Paribakht and Tréville (2007), 228 Paribakht and Wesche (2006), 228 Pawley and Syder (1983), 2 Piaget and Inhelder (1967), 140 Piasecka (2006), 95, 111 Piper and Leicester (1980), 10, 181 Plag (1999), 71 Pollio, Barlow, Fine, and Pollio (1977), 99 Qian (2004), 228, 229 Quian (1999), 179 Quian (2002), 179 Quirk, Greenbaum, Leech, and Svartvik (1985), 140, 176 Raffelsiefen (1996), 71 Raffelsiefen (1999), 71 Read (1993), 10 Richards (1976), 1 Rijpma (1999), 189 Ringbom (1987), 216, 231 Rodgers (1969), 42 Rosch and Mervis (1975), 139 Rudzka, Channel, Pudseys and Ostyn (1981), 204 Rudzka-Ostyn (2003), 152 Rye (1985), 228 Sager and Ormelius (1983), 70, 73, 74 Sandra (1990), 105
Schmidt (2004), 149, 151, 160, 161, 175, 249 Schmitt (1997), 229, 230 Schmitt (1998), 2, 9, 62, 78, 183, 184, 194 Schmitt (1999), 62, 78 Schmitt (2008), 14, 205 Schmitt (2010a), 9 Schmitt (2010b), 183 Schmitt (2010c), 3, 201 Schmitt and Dunham (1999), 205, 206, 214, 224 Schmitt and Marsden (2006), 13 Schmitt and McCarthy (1997), 58, 227 Schmitt and Meara (1997), 62, 67, 68, 69, 78, 95 Schmitt and Zimmerman (2002), 56, 57, 60, 61, 62, 76, 79, 82, 94, 215 Schneider, Healy and Bourne (2002), 16 Schuyten (1906), 16 Shefelbine (1990), 228, 233 Shillaw (1995), 14 Shioji (1982), 225 Sinclair (1991), 2 Singleton (1999), 10, 11, 181 Singson, Mahony, and Mann (2000), 96 Sjöholm (1995), 138 Skoufaki (2008), 113 Smedts (1988), 57 Stahl and Fairbanks (1986), 227 Stahl and Shiel (1992), 96 Stein, Bransford, Franks, Owings, Vye and McGraw (1982), 189 Sternberg (1987), 227, 228 Stock (1976), 42 Stoddard (1929), 16 Strömberg (1979), 191 Strömberg (1998), 191 Strömqvist (2009), 14, 204 Süskind (1987), 236 Sutarsyah, Nation and Kennedy (1994), 30 Svartvik and Sager (1988), 176 Svartvik and Sager (1996), 140 Swan and Smith (2001), 161, 216 Swanborn and de Glopper (1999), 227 Swinney (1979), 179 Swinney and Cutler (1979), 103 Söderman (1993), 10, 181 Söderström, Thorén and Karlsson (2022), 188, 224
292 Language Learning and Forced Migration
Tabossi and Zardon (1993), 104 Tabossi and Zardon (1995), 104 Thagg Fischer (1985), 12 Thorndike (1941), 65 Titone and Connine (1994), 105, 106 Titone and Connine (1999), 105 Trulsson (2007), 112, 121, 135 Tyler and Nagy (1989), 57 Tyler and Nagy (1990), 57 Verspoor (1997), 189 Verspoor and Lowie (2003), 179, 182, 189, 190, 191 Viberg (1993), 28 Viberg (2000), 178, 179, 180 Wallace Robinett (1978), 225 Wallin and Jaginder (2021), 208, 210 Waring (1997), 16, 225 Webb (2008), 227 Weinius and Whitlam (1997), 17
Wesche and Paribakht (2009), 230 West (1953), 8, 20, 24, 26, 79, 82, 84, 87, 163, 195, 198, 214, 235, 247 White, Power, and White (1989), 57, 58 Wilks and Meara (2002), 11 Wolter (2001), 10, 11, 12, 181 Wolter (2006), 180, 215, 216 Wray (2002), 144 Wysocki and Jenkins (1987), 67 Xue and Nation (1984), 183 Yule (1986), 181 Zechmeister, Chronis, Cull, D’Anna and Healy (1995), 13 Zhang (2017), 96 Zimmerman (2006), 113 Zipf (1945), 182, 184 Zwitserlood (1994), 105
Dictionaries Cambridge International Dictionary of English (1995), 216 Cobuild Idioms Dictionary (2nd edn) (2002), 117, 130, 133 The Concise Oxford Dictionary (1985), 100 Longman Dictionary of Contemporary English (1987), 193
The New Oxford Dictionary of English (1998), 198, 225 Oxford Dictionary of Phrasal Verbs (1993), 142 Oxford English Dictionary, OED Online (2000–), 100, 102, 287 Webster’s Third New International Dictionary of the English Language (1964), 13
Newspaper and magazine articles Gibbs (2008), 234 Huus (1993), 233
Jaroff (1991), 234 Unknown author (1993), 233
Websites British National Corpus (also BNC), 8, 17, 18, 20, 26, 28, 30, 31, 33, 37, 44, 45, 52, 54, 73, 76, 78, 82, 87, 93, 117, 119, 121, 130, 138, 139, 144, 156, 159, 168, 194, 195, 205, 210, 219, 235, 236 Svenska Akademiens ordlista (SAOL) Svenska Akademien, 192
Språkbanken, 8, 17, 18, 19, 20, 26, 30, 31, 33, 44, 45, 52, 54, 76, 78, 83, 93, 119, 121, 130, 156, 159, 194, 219, 235, 236 Utbildningsdepartementet, Regeringskansliet, http://www. regeringen.se/utbildning, 30
Subject Index
allomorphy, 63 association test, 10, 11, 181 basic use, definition of (preposition), 139, 140, 141, 143, 145, 149, 150, 151 clang response, 10, 217 cognitive linguistics (multi-word verbs), 152, 153, 154 cognitive-semantic theory (idiom), 112 compositional (framework), 9, 12 connectionist (framework), 10, 12, 20, 179, 181 core meaning/sense (polysemous word), 182, 185, 188, 189, 190, 191, 198, 202, 224, 225 decomposable (idiom), 104, 105, 106, 107, 108, 120, 127, 128, 129, 130 decomposition (derivative form), 42, 58, 59, 62, 63, 64, 76, 84, 86 dual coding theory (idiom), 112 figurative first hypothesis (idiom), 102, 103 idiom decomposition hypothesis, 105, 105, 106, 107 idiom key hypothesis, 104 layering (polysemous word), 181, 182, 189 levels of processing theory (idiom), 112 literal first hypothesis (idiom), 102, 103 morphological restriction (affix), 71, 86 network (building), 10, 12, 145, 176, 178, 179, 180, 181, 184, 189, 195, 206, 208, 211, 215, 217, 224, 231, 249 paradigmatic response, 11, 206 parsing ratio, 63, 64, 84, 86, 89
peripheral meaning/sense (polysemous word), 182, 189 phonological restriction (affix), 71, 97 pragmatic restriction (affix), 72 predictability (affix), 5, 64, 84, 89, 97, 248 predictability (preposition), 141, 149, 175, 249 productivity (affix), 65, 66, 70, 86, 89 prototypical response, 11 productive (knowledge), 8, 10, 12, 15, 16, 17, 18, 19, 20, 24, 25, 26, 27, 28, 29, 37, 55, 57, 61, 67, 68, 95, 96, 115, 116, 146, 153, 155, 161, 167, 183, 223, 247 receptive (knowledge), 8, 10, 11, 12, 15, 16, 17, 18, 19, 20, 22, 23, 26, 28, 29, 30, 31, 44, 61, 67, 68, 95, 96, 114, 116, 153, 161, 183, 184, 296, 223, 247 regularity (affix), 5, 59, 60, 64, 65, 84, 89, 97, 248 semantic restriction (affix), 71, 72, 88 simultaneous processing hypothesis (idiom), 103, 104 source domain (idiom and proverb), 111, 112, 113, 114, 115, 116, 122, 123, 134 synform, 42, 43, 187, 201 syntagmatic response, 11 systematic use, definition of (preposition), 139, 140, 141, 143, 145, 149, 151, 176 technicalness, 29, 33 unseparated wholes (derivative form), 58, 63, 64, 84, 90, 106 word family, 13, 59, 60, 61, 62, 74, 76, 90, 91, 92, 93, 94, 96, 97
293