Metaphor in Foreign Language Instruction

Applications of
Cognitive Linguistics

Editors
Gitte Kristiansen
Francisco J. Ruiz de Mendoza Ibáñez
Honorary editor
René Dirven

Volume 42
Metaphor in
Foreign Language
Instruction
Edited by
Ana María Piquer-Píriz
Rafael Alejo-González
ISBN 978-3-11-062673-5
e-ISBN (PDF) 978-3-11-063036-7
e-ISBN (EPUB) 978-3-11-062687-2
ISSN 1861-4078

Library of Congress Control Number: 2019947565

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available on the Internet at http://dnb.dnb.de.

© 2020 Walter de Gruyter GmbH, Berlin/Boston

Typesetting: Integra Software Services Pvt. Ltd.
Printing and binding: CPI books GmbH, Leck

www.degruyter.com
Acknowledgments
First and foremost, this book is a message of gratitude to Fiona MacArthur for
her substantial contribution to the study of metaphor in second language learn-
ing and for her generous work as an academic and a university lecturer.
Without her inspiration, the field would not be what it is now. We would also
like to thank the authors of the different chapters included in this volume for
their excellent articles and for having attended our call to contribute to the vol-
ume while at the same time complying with the deadlines as well as accepting
to go through a serious review process. We are also grateful to all the reviewers
of the chapters and especially to Frank Boers, who commented on the pre-final
manuscript thus contributing to improve the final result. Finally, we are thank-
ful to the series editors and to the editors of Mouton de Gruyter, who have
made the whole process of publication easier.

https://doi.org/10.1515/9783110630367-202
Contents
Acknowledgments V

Introduction

Ana M. Piquer-Píriz and Rafael Alejo-González

Metaphor in L2 Instruction: Key Issues 3

Part I. Theoretical Considerations: Reviews

and Perspectives

Raymond W. Gibbs, Jr.

The Particularities of Metaphorical Experience: An Appreciation of Fiona
MacArthur’s Metaphor Scholarship 17

Graham Low
Taking Stock after Three Decades: “On Teaching Metaphor”
Revisited 37

Ana M. Piquer-Píriz
Figurative Language and Young L2 Learners 57

Laura Suárez-Campos, Alberto Hijazo-Gascón

and Iraide Ibarretxe-Antuñano
Metaphor and Spanish as a Foreign Language 79

Part II. New Empirical Studies

Part II. 1. Learners’ Use and Knowledge of L2 Figurative

language

Jeannette Littlemore, Paula Pérez-Sobrino, Nina Julich and Danny Leung

Is Comfort Purple or Green? Word-colour Associations in the First and
Second Language 103
VIII Contents

Gill Philip
Metaphorical Reasoning in Comprehension and Translation: An Analysis
of Metaphor in Multiple Translations 131

Tina Krennmayr
Metaphorical Alignment in Cross-Cultural Office Hours’
Consultations 149

Susan Nacey
Development of L2 Metaphorical Production 173

Irene Castellano-Risco and Ana M. Piquer-Píriz

Measuring Secondary-School L2 Learners Vocabulary Knowledge:
Metaphorical Competence as Part of General Lexical Competence 199

Part II. 2. Fostering Knowledge of L2 Figurative Language

Xinqing Wang, Frank Boers and Paul Warren

Using Literal Underpinnings to Help Learners Remember Figurative
Idioms: Does the Connection Need to Be Crystal Clear? 221

Rafael Alejo-González and Verónica García-Bermejo

“The Manage of Two Kingdoms Must”: An Analysis of Metaphor in Two
CLIL Textbooks 241

Rawan A. Saaty
An Enactment-Based Approach to the Teaching of Metaphoric
Expressions: A Case of Saudi EFL Learners 263

Subject Index 287

Introduction
Ana M. Piquer-Píriz and Rafael Alejo-González
Metaphor in L2 Instruction: Key Issues
1 Rationale
Since the publication of Metaphors We Live By (Lakoff and Johnson 1980) and the
subsequent development and consolidation of Cognitive Linguistics (CL) as a well-
established theory in Linguistics, exploring the role that metaphor, in particular,
and figurative language, in general, play in our linguistic and conceptual systems
has attracted a great deal of attention from scholars of different disciplines.
One of the areas in which applications of the theoretical tenets of CL have
been particularly fruitful is foreign language instruction (Bielak 2011; Boers and
Lindstromberg 2006; De Knop, Boers and De Rycker 2010; Littlemore and Low
2006; Piquer-Píriz and Alejo-González 2016; Piquer-Píriz and Boers 2019). A
ground-breaking contribution to how to approach metaphor when teaching an
L2 was Low’s seminal article published in Applied Linguistics in 1988 (which is
revisited by the author in Chapter 2 of this volume). This was followed in the
1990s by several journal articles introducing CL-oriented teaching proposals
(Boers and Demecheleer 1998; Deignan, Gabryś and Solska 1997; Kövecses and
Szabó 1996; Lazar 1996, Lindstromberg 1996; MacLennan 1994; Ponterotto
1994). In the 2000s, several monographs and collective volumes further defined
what has become known as Applied Cognitive Linguistics (ACL) (Achard and
Niemeier 2004; Boers and Lindstromberg 2008; De Knop, Boers and De Rycker
2010; Holme 2009; Littlemore 2009, 2017; Littlemore and Juchem-Grundmann
2010; Littlemore and Low 2006; Nacey 2017; Pütz, Niemeier and Dirven 2001a,
2001b, Pütz and Sicola 2010; Robinson and Ellis 2008). While some of these vol-
umes included mostly contributions of a rather theoretical or exploratory na-
ture, they have been complemented by an ever growing number of empirical
studies demonstrating that the application of CL tenets can have a positive influ-
ence on instructed L2 acquisition (Alejo 2010; Alejo, Piquer Píriz and Reveriego
2010; Boers 2000, 2004; Boers, Eyckmans and Stengers 2007; Boers, Piquer-Píriz,
Stengers and Eyckmans 2009; Condon 2008; Dirven 2001; Lindstromberg and
Boers 2005; Littlemore and MacArthur 2007, 2008; MacArthur and Piquer-Píriz
2007; MacArthur and Littlemore 2008; Piquer-Píriz 2008, 2010; Saaty 2016; Tyler,

Ana M. Piquer-Píriz, English Philology (Faculty of Education), University of Extremadura,

Badajoz, Spain, e-mail: [email protected]
Rafael Alejo-González, English Philology (Faculty of Education), University of Extremadura,
Badajoz, Spain, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-001
4 Ana M. Piquer-Píriz and Rafael Alejo-González

Mueller and Ho 2010). Most of these studies have focused on the notion of moti-
vated polysemy as opposed to semantic arbitrariness and have shown how en-
hancing the notion of linguistic motivation among L2 learners by means of
different techniques (such as identifying conceptual metaphors, raising metaphor
awareness in the L1 or using images, etymology, cross-cultural comparisons, or
physical enactment to reconnect figurative meanings to their literal underpin-
nings) may facilitate L2 learners’ comprehension and recall of figurative vocabu-
lary. Boers (2011, 2013) reviews CL-inspired studies published between 1996 and
2010 that put the effectiveness of this approach to the test, pointing out the main
contributions of these studies to the field but also some of their methodological
limitations. In his paper on assessing and integrating Cognitive Linguistic ap-
proaches to teaching vocabulary, he concludes that:

It must be acknowledged, however, that many of these are small-scale, some show only
small effect sizes, and some are hard to interpret due to confounding variables. Taken collec-
tively, the reported experiments are nevertheless beginning to constitute a body of evidence
in favour of CL-informed instruction that is hard to dismiss, so there is reason to believe that
this kind of instruction deserves a niche in second language programmes. (Boers 2013: 208)

However, despite the apparent important insights generated by this body of re-
search in the past three decades, little impact seems to have been made on main-
stream teaching materials (Lazar 2003, Rudzka-Ostyn 2003 and Lindstromberg and
Boers 2008 are some exceptions) and L2 classroom practice. For example, as
MacArthur (2017) states, although it has been shown (Littlemore and Low 2006)
that metaphor plays an important role in all four dimensions of communicative
competence (linguistic, sociolinguistic, discourse and strategic), it is conspicuously
absent in the descriptors established for the Common European Framework of
Reference for Languages (CEFRL) and “is still regarded as somewhat marginal in
the materials published for English as S/FL.” (MacArthur 2017: 418). Thus, despite
the continued and fruitful effort to give metaphor its well-deserved place in foreign
language instruction, there are still many gaps to be filled.
Exploring the role of figurative language in L2 instruction and cross-cultural
communication have been the two main foci of Dr. Fiona MacArthur, the scholar
to whom we dedicate the present collective volume (e.g., MacArthur 2005, 2010,
2016a, 2016b, 2016c, 2017; MacArthur et al. 2012; MacArthur, Krennmayr and
Littlemore 2015; Musolff, MacArthur and Pagani 2014). The opening chapter of this
volume, by Gibbs, reviews a selection of MacArthur’s publications and emphasises
how “by recognizing the limitations of our methods and the constant need for
broader contextualization of how we study real-life metaphor use [. . .] she offers
us a distinctive, ideal model on how to best study metaphor.” (Gibbs, this volume:
35). Throughout her academic career, she has not only shown an endless interest
 Metaphor in L2 Instruction: Key Issues 5

in metaphor but she has been, and is still, able to instil this enthusiasm also in
her students and colleagues. A group of close colleagues, some of whom are also
eminent scholars in the field, have got together in this volume to offer an over-
view of the study of metaphor in L2 instruction, including identification of areas
that need further research, as well as to illustrate new developments in empirical
Applied Cognitive Linguistics research using diverse methods (such as analyses
of learner corpora and classroom interventions). The findings demonstrate the
relevance of the basic premises of CL for L2 acquisition research and provide in-
sights into how CL can be applied in real instructional contexts.
This volume consists of two main parts. Part I comprises four review ar-
ticles that touch upon issues of continuing relevance to the discipline (e.g. why
metaphor is relevant for L2 learners and how it can be taught) and identify
areas in need of further research (such as metaphor in the instruction of lan-
guages other than English or how metaphorical competence can be fostered
with young L2 learners). These conceptual reviews of specific bodies of research
will hopefully help readers to interpret and appreciate the value of the empiri-
cal studies reported in the second part of the book. Part II consists of eight new
empirical studies that illustrate both methodological challenges and promising
practices, lighting various paths that can be followed when analysing the role
that metaphorical language plays in L2 instruction in real contexts. These eight
original studies belong to two different broad areas: The first five explore L2
learners’ knowledge of L2 figurative language and how they use it while the
last three chapters evaluate different possibilities for enhancing knowledge of
non-literal language in the foreign language classroom.

2 Part I. Theoretical Considerations:

Reviews and Perspectives
Gibbs opens the first part of the volume with a review of what he considers the
most influential contributions by Fiona MacArthur to the study of metaphor. He
shows how her work on metaphor in classroom discourse, metaphor in interac-
tion between native and non-native speakers of a language, and metaphor use
in various socio-cultural contexts is directly relevant to some of the general
topics that have been raised in metaphor studies such as: (1) the understanding
of primary metaphors, (2) the role of metaphorical reasoning in an L2, (3) how
not only linguistic but also historical, cultural and social factors underlie meta-
phorical language, (4) how speakers of more than one language ‘mix’ meta-
phors and (5) metaphor identification. According to Gibbs, one of MacArthur’s
6 Ana M. Piquer-Píriz and Rafael Alejo-González

greatest contributions to our discipline is that she keeps reminding us of the

complexities of studying metaphor, the interconnection between language, cul-
ture and thought and the particularities of metaphor use in situated discourse,
where metaphor should be viewed as a resource to meet individual and collec-
tive adaptive needs.
The chapter by Low is a review of one of the most influential (and most cited)
publications on metaphor and foreign language instruction, written by the author
over thirty years ago. Low identifies the issues that remain relevant in relation to
the importance of metaphor for L2 learners such as his ‘treatment’ definition of
metaphor, its ‘core’ functions and the importance of metaphorical competence,
and he also summarises a number of notable new developments in the field in
the past thirty years, including the outcomes from corpus research, studies into
learner language and learning strategies, and empirical studies of teaching meta-
phor as part of foreign language training.
The last two chapters in this first part tackle under-explored issues in the
field of metaphor and foreign language instruction. Piquer-Píriz addresses the
topic of metaphor and young L2 learners, and connects the findings from child
language studies into children’s understanding and production of figurative
language in their L1 with the insights from metaphor research studies on the
pedagogical potential of enhancing learners’ appreciation of motivated poly-
semy already with L2 learners at a young age.
Finally, Suárez-Campos, Hijazo-Gascón and Ibarretxe-Antuñano focus on met-
aphor in L2 instruction in a language other than English, in this case, Spanish.
They point out that although English has been the predominant language to
which CL-oriented pedagogies have been applied, other languages, such as
Chinese or Spanish have also been explored. Focusing mainly on the notions of
‘conceptual metaphor’ and ‘metaphorical competence’ and their importance for L2
instruction, they review the available studies that use metaphor in the acquisition
of Spanish as an L2, thus illustrating how the pedagogical applications of meta-
phor theory as a tool are also applicable to Spanish as a second language.

3 Part II. New Empirical Studies

Part II of the book offers a selection of new empirical studies which expand and
continue the research carried out following an Applied Cognitive Linguistic per-
spective, as cited above. More specifically, the chapters included here examine
broad areas organised in two different sections. In the first section, five chapters
deal with L2 learners’ knowledge and use of metaphorical language while in
 Metaphor in L2 Instruction: Key Issues 7

the second part, three chapters evaluate pedagogic procedures for fostering L2
metaphorical competence in diverse instructional contexts.
The first chapter included in section 1 is authored by Littlemore, Pérez-
Sobrino, Julich and Leung, who present the results of a large-scale question-
naire-based project on word-colour associations. Respondents were speakers of
two of the most widely used languages in the world: English and Cantonese.
One of the aims of the study was to determine to what extent the level of em-
bodiment perceived in word-colour associations influences the likelihood that
L2 learners will gradually adopt typical L2 associations in their interlanguage.
The presence of metaphor in Higher Education is dealt with in the following
chapter, where Philip analyses the problems posed by the metaphors that appear
in a translation course (English-Italian) at an Italian university. The more basic
challenge has to do with the recognition of metaphors in the text-task at which
most students are successful but also with choosing the translation that best fits
the context and is compatible with the lexis of the source domain of the meta-
phor in the target language. Sometimes, students are unable to find the appropri-
ate term but the fact that in most cases they are willing to risk a metaphorically
less optimal translation in the target language is already an indication that they
are becoming aware of its importance in translation. Philip also finds that lower-
proficiency students tend to be less successful at translating metaphors and are
more prone to misinterpreting metaphorical language as literal. So, again, the
profile of the learner matters, even within a population of students sharing the
same interest and aspiration (in this case, becoming professional translators).
Krennmayr, who has already published an article with Fiona MacArthur on
a similar topic before (MacArthur et al. 2015), pays attention to yet another edu-
cational context where metaphor is common: office hours’ consultations by
Spanish Erasmus students on their year abroad. These consultations are carried
out in English and provide a good opportunity for exploring the use of meta-
phor in academic dialogue between novice and expert users, when the typical
context for oral academic communication, lectures, is mostly monologic and
unidirectional. In this way, it provides the perfect testing ground to analyse
how interlocutors use and reuse the same conceptual metaphor to achieve
alignment, a feature of oral interaction that allows the conversation to flow. In
her article, she focuses on SIGHT metaphors and explores the use of the
WMatrix semantic annotation tools to carry out her analysis.
The following chapter, authored by Nacey, also deals with metaphorical
competence, although this time the connection is established with general profi-
ciency. Reminiscent of earlier work by Littlemore et al. (2014), a corpus of written
English (TraWL) serves to examine metaphor use by Norwegian school learners
belonging to different age groups. The greater metaphor density found in the
8 Ana M. Piquer-Píriz and Rafael Alejo-González

more advanced school grades –grade being used as a proxy of proficiency- is in-
terpreted as a natural progression in L2 acquisition, not only because of the de-
velopment in metaphorical competence by the learners but also because of the
more abstract nature of the topics they are asked to write about. This progression
is not only shown quantitatively but also qualitatively in the selection of func-
tions that metaphors perform.
The relationship between lexical and metaphorical competence in L2 acquisi-
tion, a connection seldom dealt with in the literature, is explored by Castellano-
Risco and Piquer-Píriz. To do so, they analyse the extent to which the learners’
knowledge of the metaphors identified in two standard vocabulary tests (VLT
and PVLT) correlate with their global scores on these tests, research comple-
mented by an analysis of the success rate of the individual metaphor items in
comparison to the other lexical items in the tests. The results are preliminary but
point to the possibility that including metaphorical items in the tests may imply
an added degree of difficulty not captured by word frequency in a corpus (i.e.,
the main selection principle behind these vocabulary tests).
The second section of part II begins with a chapter by Wang, Boers and
Warren, who give continuation to the research emphasizing the importance of
providing meaning motivation in the teaching of L2 vocabulary. In this case,
the authors study to what extent students at different levels of proficiency ben-
efit from the explanations provided about the origin or motivation of idiomatic
expressions. The results show that more advanced students do not need the
motivation to be convincing to remember the idioms they have been taught,
while less advanced students did not profit as much from explanations that
they found far-fetched. In this way, the authors highlight the importance of the
other part of the equation in teaching contexts: the learners. Not all methods or
approaches are created equal when it comes to metaphor teaching.
The need to raise metaphor awareness in another educational context, CLIL
(Content and Language Integrated Learning), is emphasized in the chapter by
Alejo-González and García-Bermejo, who analyse the metaphor density of two text-
books (Social and Natural Science) used to teach these subjects in bilingual
schools in Spain. The study shows that, in spite of being designed for non-native
speakers of English, these books contain a high number of open class metaphors,
although not as many as in ‘real’ academic prose. At the same time, the analysis
shows that metaphor density is not evenly spread. For one, the lessons belonging
to Natural Science show a greater use of metaphorical language. For another,
within each lesson unit, some sections show a greater metaphor density than
others, and this can be accounted for by the pedagogic functions they fulfil.
Finally, the chapter by Saaty explores the importance of metaphor awareness
in vocabulary teaching and performs a comparative analysis of three different
 Metaphor in L2 Instruction: Key Issues 9

techniques used in the teaching of vocabulary to L2 learners: 1) the use of

bodily enactment as a way of connecting figurative uses of expressions to their
literal meanings, 2) the use of verbal discussions to help students appreciate
the conceptual metaphor underlying the figurative expressions, and 3) activi-
ties involving the same figurative expressions but without explicit attention to
the connections with their literal meanings. The results show an advantage of
the first technique, which not only enhances comprehension but also longer-
term retention of the target expressions.

4 Conclusion and Outlook

As pointed out above, despite the continued and fruitful efforts to give metaphor
its well-deserved place in foreign language instruction for the past thirty years,
there are still many gaps to be filled.
In the first place, it has been largely acknowledged that the concept (and
measures) of ‘metaphorical competence’ needs to be further developed. Five of the
chapters included in this volume (those by Low; Piquer-Píriz; Suárez-Campos,
Hijazo-Gascón and Ibarretxe-Antuñano; Nacey; and Castellano-Risco and Piquer-
Píriz) approach this issue from different perspectives.
Secondly, very few attempts have been made to develop CL-oriented specific
teaching materials and the main findings from this line of research do not seem
to have reached mainstream textbooks and classroom practice (Boers, 2014,
MacArthur 2017). The chapters by Saaty; Alejo-González and García-Bermejo; and
Wang, Boers and Warren offer some suggestions for either specific teaching tech-
niques or general ways to enhance metaphor awareness in the English as an L2
classroom.
Thirdly, there is a clear need to extend the applications of CL to L2 instruction
in languages other than English. Suárez-Campos, Hijazo-Gascón and Ibarretxe-
Antuñano’s chapter offers relevant insights for the case of Spanish as an L2.
And fourthly, there is also an evident need to be sensitive to the specific pro-
files of different learners groups (and individual learners within seemingly homog-
enous groups). Four of the chapters illustrate this issue, by looking at young L2
learners (in the chapters by Nacey and Piquer-Píriz), learners specialising in a par-
ticular area (students of Translation, in the chapter by Philip), and learners with
different levels of L2 achievement (in the chapters by Nacey and Wang, Boers and
Warren).
Both the more theoretical accounts and the new empirical studies pre-
sented in these pages also indicate promising avenues for future research and
10 Ana M. Piquer-Píriz and Rafael Alejo-González

possible pedagogical applications of metaphor in L2 classrooms. We indeed

hope that the information presented in the different chapters of this volume
will inspire further research into the fascinating, multifaceted issue of meta-
phor in L2 instruction.

References
Achard, Michel & Susanne Niemeier. 2004. Cognitive linguistics, second language acquisition,
and foreign language teaching (Vol. 18). Berlin/New York: Mouton de Gruyter.
Alejo, Rafael. 2010. Where does the money go? An analysis of the container metaphor in
economics: The market and the economy. Journal of Pragmatics 42(4).1137–1150.
Alejo, Rafael, Ana M. Piquer-Píriz & Guadalupe Reveriego. 2010. Phrasal verbs in EFL
coursebooks. In Sabine De Knop, Frank Boers & Antoon De Rycker (eds.), Fostering
language teaching efficiency through cognitive linguistics, 59–78. Berlin: Mouton de
Gruyter.
Bielak, Jakub. 2011. Cognitive linguistics and foreign anguage pedagogy: An overview of
recent trends and developments. In Miroslaw Pawlak (eds.), Extending the boundaries of
research on second language learning and teaching, 241–262. Berlin, Heidelberg:
Springer Berlin Heidelberg.
Boers, Frank. 2000. Metaphor awareness and vocabulary retention. Applied Linguistics 21.
553–571.
Boers, Frank. 2004. Expanding learners’ vocabulary through metaphor awareness: What
expansion, what learners, what vocabulary? In Susanne Niemeier & Michel Achard (eds.),
Cognitive linguistics, second language acquisition, and foreign language teaching,
211–232. Berlin: Mouton de Gruyter.
Boers, Frank. 2011. Cognitive semantic ways of teaching figurative phrases: An assessment.
Review of Cognitive Linguistics 9. 227–261.
Boers, Frank. 2013. Cognitive linguistic approaches to teaching vocabulary: Assessment and
integration. Language Teaching 46. 208–224.
Boers, Frank. 2014. Cognitive linguistics and language pedagogy: Finding ways forward. Paper
presented at the 9th AELCO International conference, Badajoz, Spain.
Boers, Frank & Murielle Demecheleer. 1998. A few metaphorical models in (western) economic
discourse. In Wolf A. Liebert, Gisela Redeker & Linda R. Waugh (eds.), Discourse and
perspective in cognitive linguistics, 115–131. Amsterdam: John Benjamins.
Boers, Frank, June Eyckmans & Hélène Stengers. 2007. Presenting figurative idioms with a touch
of etymology: More than mere mnemonics? Language Teaching Research 11(1). 43–62.
Boers, Frank & Seth Lindstromberg. 2006. Cognitive linguistic approaches to second or
foreign language instruction: Rationale, proposals and evaluation. In Gitte Kristaensen,
René Dirven, Michel Achard & Francisco J. Ruiz-Mendoza (eds.), Cognitive linguistics:
Current applications and future perspectives, 305–358. Berlin: Mouton de Gruyter.
Boers, Frank & Seth Lindstromberg (eds.). 2008. Cognitive linguistic approach to teaching
vocabulary and phraseology. Berlin: Mouton de Gruyter.
 Metaphor in L2 Instruction: Key Issues 11

Boers, Frank & Seth Lindstromberg. 2008. How cognitive linguistics can foster effective
vocabulary teaching. In Frank Boers & Seth Lindstromberg, Cognitive linguistic approach
to teaching vocabulary and phraseology, 1–65. Berlin: Mouton de Gruyter.
Boers, Frank & Seth Lindstromberg. 2009. Optimizing a lexical approach to instructed second
language Acquisition. Basingstoke: Palgrave Macmillan.
Boers, Frank Ana M. Piquer-Píriz, Hélène Stengers & June Eyckmans. 2009. Does pictorial
elucidation foster recollection of idioms? Language Teaching Research 13. 267–382.
Condon, Nora. 2008. How cognitive linguistic motivations influence the learning of phrasal
verbs. In Frank Boers & Seth Lindstromberg (eds.), Cognitive linguistic approaches to
teaching vocabulary and phraseology. 133–158. Berlin: Mouton de Gruyter.
De Knop, Sabine, Frank Boers & Antoon De Rycker. 2010. Fostering language teaching
efficiency through cognitive linguistics (Vol. 17). Berlin: Mouton de Gruyter.
Deignan, Alice, Danuta Gabryś & Agnieszka Solska. 1997. Teaching English metaphors using
cross-linguistic awareness-raising activities. ELT journal 51(4). 352–360.
Dirven, René. 2001. English phrasal verbs: Theory and didactic application. In Martin Pütz,
Susanne Niemeier & René Dirven (Eds.), Applied cognitive linguistics II: Language
pedagogy, 3–27. Berlin: Mouton de Gruyter.
Holme, Randal. 2009. Cognitive linguistics and language teaching. Berlin: Springer Berlin
Heidelberg.
Kövecses, Zoltán & Péter Szabó. 1996. Idioms: A view from cognitive semantics. Applied
Linguistics, 17(3). 326–355.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago
Press.
Lazar, Gillian. 1996. Using figurative language to expand students’ vocabulary. ELT Journal
50(1). 43–51.
Lazar, Gillian. 2003. Meaning and Metaphors: Activities to practise figurative language.
Cambridge: Cambridge University Press.
Lindstromberg, Seth. 1996. Prepositions: Meaning and method. ELT Journal 50(3). 225–236.
Lindstromberg, Seth & Frank Boers. 2005. From movement to metaphor with manner-of-
movement verbs. Applied Linguistics 26(2). 241–26.
Lindstromberg, Seth & Frank Boers. 2008. Teaching chunks of language: From noticing to
remembering. London: Helbling Languages.
Littlemore, Jeannette. 2009. Applying cognitive linguistics to second language learning and
teaching. Berlin Heidelberg: Springer Berlin Heidelberg.
Littlemore, Jeannette. 2017. Metaphor use in educational contexts: Functions and variations.
In Elena Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and
language, 283–295. London: Routledge.
Littlemore, Jeannette & Constanze Juchem-Grundmann. 2010. Introduction to the interplay
between cognitive linguistics and second language learning and teaching. AILA Review
23(1). 1–6.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sara Turner. 2014. An investigation into
metaphor use at different levels of second language writing. Applied Linguistics 35(2).
117–144.
Littlemore, Jeannette & Graham D. Low. 2006. Figurative thinking and foreign language
learning. Basingstoke: Palgrave Macmillan.
Littlemore, Jeannette & Fiona MacArthur. 2007. What do learners need to know about the
figurative extensions of target langauge words? A contrastive corpus-based analysis of
12 Ana M. Piquer-Píriz and Rafael Alejo-González

thread, hilar, wing and aletear. In Ignasi Navarro i Fernando, José L. Otal Campo &
Antonio J. Silvestre López (eds.), Metaphor and discourse, a special edition of culture,
language and representation: Cultural studies journal of universitat Jaume I 5. 131–149.
Littlemore, Jeannette & Fiona MacArthur. 2012. Figurative extensions of word meaning: How
do corpus data and intuition match up? In Dagmar Divjak & Stefan T. Gries (eds.),
Frequency effects in language representation, 195–233. Berlin: Mouton de Gruyter.
Low, Graham D. 1988. On teaching metaphor. Applied Linguistics 9(2). 125–147.
MacArthur, Fiona. 2005. The competent horseman in a horseless world: Observations on a
conventional metaphor in Spanish and English. Metaphor and Symbol 20. 71–94.
MacArthur, Fiona. 2010. Metaphorical competence in EFL. Where are we and where should be
going? A view from the language classroom. AILA Review 23 (1). 155–173.
MacArthur, Fiona. 2016a. Beyond engaged listenership: Assessing Spanish undergraduates’
active participation in academic mentoring sessions in English as academic lingua
franca. In Jesús Romero-Trillo (ed.), Yearbook of Corpus Linguistics and Pragmatics 2016.
Global Implications for Society and Education in the Networked Age, 153–178. Cham:
Springer.
MacArthur, Fiona. 2016b. Overt and covert uses of metaphor in the academic mentoring in
English of Spanish undergraduate students at five European universities. Review of
Cognitive Linguistics 41(1). 23–50.
MacArthur, Fiona. 2016c. When languages and cultures meet: Mixed metaphors in the
discourse of Spanish speakers of English. In Raymond W. Gibbs (Ed.), Mixing metaphor,
133–154. Amsterdam: John Benjamins.
MacArthur, Fiona. 2017. Using metaphor in the teaching of second/foreign languages. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
413–424. London: Routledge.
MacArthur, Fiona, Tina Krennmayr & Jeannette Littlemore. 2015. How basic is UNDERSTANDING
IS SEEING when reasoning about knowledge? Asymmetric uses of SIGHT metaphors in
office hours’ consultations in English as academic lingua franca. Metaphor and Symbol
30(3). 184–217.
MacArthur, Fiona & Jeannette Littlemore. 2008. A discovery approach using corpora in the
foreign language classroom. In Frank Boers & Seth Lindstromberg (eds.), Cognitive
linguistic approaches to teaching vocabulary and phraseology, 159–188. Berlin: Mouton
de Gruyter.
MacArthur, Fiona, José L. Oncins-Martínez, Manuel Sánchez-García & Ana M. Piquer-Píriz. 2012.
Metaphor in use: Context, culture and communication. Amsterdam: John Benjamins.
MacArthur, Fiona & Ana M. Piquer-Píriz. 2007. Staging the introduction of figurative extensions
of familiar vocabulary items in EFL: Some preliminary considerations. Ilha de Desterro: a
Journal of English Language, Literatures in English and Cultural Studies 53. 123–134.
MacLennan, Carol H. 1994. Metaphors and prototypes in the learning teaching of grammar
and vocabulary. IRAL-International Review of Applied Linguistics in Language Teaching
32(2). 97–110.
Mussolf, Andreas, Fiona MacArthur & Giulio Pagani (eds.). 2014. Metaphor and intercultural
communication. London: Bloomsbury Academic.
Nacey, Susan. 2017. Metaphor comprehension and production in a second language. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
503–515. London: Routledge.
 Metaphor in L2 Instruction: Key Issues 13

Piquer-Píriz, Ana M. 2008. Reasoning figuratively in early EFL: Some implications for the
development of vocabulary. In Frank Boers & Seth Lindstromberg (eds.), Cognitive
linguistics approaches to teaching vocabulary and phraseology, 219–240. Berlin: Mouton
de Gruyter.
Piquer-Píriz, Ana M. 2010. Can people be cold and warm? Developing understanding of
figurative meanings of temperature terms in early EFL. In Graham D. Low, Zazie Todd,
Alice Deignan & Lynne Cameron (eds.), Researching and applying metaphor in the real
world, 21–33. Amsterdam: John Benjamins.
Piquer-Píriz, Ana M. & Rafael Alejo-González. 2016. Applying Cognitive Linguistics: Identifying
some current research foci (figurative language in use, constructions and typology).
Review of Cognitive Linguistics 14(1). 1–20.
Piquer-Píriz, Ana M. & Frank Boers. 2019. La lingüística cognitiva y sus aplicaciones a la
enseñanza de lenguas extranjeras. In Iraide Ibarretxe-Antuñano, Teresa Cadierno &
Alejandro Casteñeda-Castro (eds.), Lingüística cognitiva y español LE/L2. London:
Routledge.
Ponteretto, Diane. 1994. Metaphors we can learn by: How insights from cognitive linguistic
research can improve the teaching/learning of figurative language. English Teaching
Forum 32(3). 2–7.
Pütz, Martin, René Dirven & Susanne Niemeier. 2001a. Applied cognitive linguistics I: Theory
and Language Acquisition. Berlin: Mouton de Gruyter.
Pütz, Martin, René Dirven & Susanne Niemeier. 2001b. Applied cognitive linguistics II:
Language pedagogy. Berlin: Mouton de Gruyter.
Pütz, Martin & Laura Sicola (eds.). 2010. Cognitive processing in second language acquisition:
Inside the learner's mind. Amsterdam: John Benjamins.
Robinson, Peter & Nick C. Ellis. 2008. Handbook of cognitive linguistics and second language
acquisition. New York: Routledge.
Rudzka-Ostyn, Brygida. 2003. Word Power: Phrasal verbs and compounds. A cognitive
approach. Berlin: Mouton de Gruyter.
Saaty, Rawan. 2016. Teaching L2 Metaphor through awareness-raising activities: Experimental
studies with Saudi EFL learners. Unpublished Doctoral Dissertation. University of
Birmingham.
Tyler, Andrea, Charles Mueller & Vu Ho. 2010. Applying cognitive linguistics to instructed L2
learning: The English modals. AILA Review 23(1). 30–49.
Part I. Theoretical Considerations:
Reviews and Perspectives
Raymond W. Gibbs, Jr.
The Particularities of Metaphorical
Experience: An Appreciation of Fiona
MacArthur’s Metaphor Scholarship
1 The Particularities of Metaphorical Experience
Being a metaphor scholar is really hard work. I recently talked with an old friend,
a neurosurgeon, about my own life as a metaphor scholar. He asked pointed ques-
tions about what metaphor was, how I studied it as a cognitive scientist, what the-
ories explained metaphor in language and thought, and, finally, why on earth I
decided to devote my life to the topic of metaphor in the first place. After hearing
my different responses to these questions, my friend commented that metaphor
seemed vague, even ethereal, and wondered if any progress in the scientific study
of metaphor could ever occur. My old friend even said, and this is an exact quote,
“It sounds to me that operating on people’s brains is much easier than studying
something like metaphor”. I laughed at this response, but could not argue with its
validity as studying metaphor can often be terribly challenging (even if also quite
rewarding).
I admire most of the people in the metaphor community for their courage in
the work they do, and have special appreciation for Fiona MacArthur and her
scholarship over many years. In my view, she distinguishes herself as a metaphor
scholar, par excellence, for approaching metaphor with a keen eye toward the par-
ticularities of metaphorical experience in all of its glorious, messy details. Quite a
few of us in the multidisciplinary world of metaphor research examine metaphor
from a God’s eye perspective and aim to capture some of the important generalities
about metaphor in broad theoretical terms. Indeed, some of the most famous theo-
ries of metaphor have evolved from the consideration of only a few, often literary,
examples of metaphor, or from instances noted in dictionaries, or from isolated ex-
pressions that arise from our own minds or memories. Others of us look at meta-
phor as it exists in real discourse, but then scrape the metaphors out of these
contexts to draw larger generalizations about metaphor and the presumed people
who produce and understand these selected examples. Still others, including my-
self, study people’s understandings of verbal metaphor in scientific experiments,
but here too quickly draw inferences from the small group of participants and

Raymond W. Gibbs, Jr., Soquel, CA, USA, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-002
18 Raymond W. Gibbs, Jr.

specific verbal metaphors examined to offer grand conclusions about human

minds and unconscious, fast-acting processes of metaphor acquisition and inter-
pretation. Finally, other scholars investigate metaphorical discourse in order to
draw larger generalizations about how metaphor always emerges from cultural
ideas and beliefs. These differing perspectives each offer potentially important in-
sights into the ways metaphor operates in thought, language, and expressive com-
munication. But it is still often the case that these varying approaches to metaphor
are seen as competing with one another rather than pointing toward a more har-
monious and detailed picture of the adaptive value of metaphor in human life.
Fiona MacArthur has long resisted the temptation to draw facile conclusions
about the ways metaphor works in human interaction. She looks at the particular-
ities of how metaphor emerges in discourse and offers a cautionary voice about
not ignoring some of the complexities of metaphor use simply because so many of
us are primarily interested in grand theories of metaphor. Reading her many publi-
cations over the years offers a mosaic, a tapestry of observations about many of
the detailed realities of metaphor that are too often ignored in many of the ongoing
debates about how people, especially non-native speakers, learn and communicate
with metaphorical language. The enduring impression I get from reading each one
of her articles is the warning to fellow metaphor research of “Not so fast! Slow
down and see how the complexities of metaphorical language use must be a fun-
damental part of any attempt to theorize about metaphor”.
I now review a select few of her scholarly works to both remind us of their
important empirical findings and to understand some of the deeper themes that
motivate her research as a metaphor scholar. A key focus of my appreciation of her
research is that metaphorical language use is always inherently situated in, and
emerges from, very particular social, cultural, and environmental circumstances,
which make up the “ecology of metaphor”. Thinking of metaphor as a resource to
meet individual and collective adaptive needs entails certain methodological com-
mitments and attention to the dynamics of how metaphor unfolds.

2 A Case Study of the Primary Metaphor

UNDERSTANDING IS SEEING
The introduction of “primary metaphors” within cognitive linguistics added a
new dimension for characterizing the relations between bodily experience with
metaphorical thought and language (Grady 1997). Primary metaphors reflect
strong correlations in everyday embodied experiences, such as the following:
 The Particularities of Metaphorical Experience 19

INTIMACY IS CLOSENESS
“We have a close relationship”.

DIFFICULTIES ARE BURDENS

“She’s weighed down by responsibilities”.

AFFECTION IS WARMTH
“They greeted me warmly”.

IMPORTANT IS BIG
“Tomorrow is a big day”.

MORE IS UP
“Prices are high”.

SIMILARITY IS CLOSENESS
“Those colors aren’t the same, but they’re close”.

CHANGE IS MOTION
“My health has gone from bad to worse”.

PURPOSES ARE DESTINATIONS

“He’ll be successful, but isn’t there yet”.

CAUSES ARE PHYSICAL FORCES

“They pushed the bill through Congress”.

KNOWING IS SEEING
“I see what you mean”.

UNDERSTANDING IS GRASPING
“I’ve never been able to grasp complex math”.

These metaphorical correlations arise out of our embodied functioning in the

world. In each case, the source domain of the metaphor comes from the body’s
sensorimotor system. A primary metaphor is a metaphorical mapping for which
there is an independent and direct experiential basis that can be expressed within
language. Although primary metaphors are often viewed as having a universal
character given their roots in embodied correlations, not all languages exhibit the
same primary metaphors, such as for example UNDERSTANDING IS SEEING
(Ibarretxe-Antuñano 2013).
How do second-language learners of English use and understand a primary
metaphor such as UNDERSTANDING IS SEEING? MacArthur, Krennmayr and
Littlemore (2015) explored the relevance of UNDERSTANDING IS SEEING in a se-
ries of 27 conversations (EuroCoAT corpus) between university instructors of
English and their international students at five European universities (i.e.,
Ireland, England, The Netherlands, Spain, and Sweden). These conversations
were all in English while the students were all native Spanish speakers. In gen-
eral, the UNDERSTANDING IS SEEING metaphor appeared frequently in these
20 Raymond W. Gibbs, Jr.

conversations, notably because the topics of learning and knowledge were the
primary foci of these talk exchanges. Still, there were important variations in the
ways that the English instructors and the Spanish students referred to
UNDERSTANDING IS SEEING.
The students were asked, prior to the meetings with their instructor, to pro-
pose several questions on topics related to the course materials and assignments.
The resulting conversations lasted between 7 and 22 minutes, yielding a corpus of
over six hours of discourse. About 2/3 of the words were produced by the instruc-
tors and 1/3 by the students. An analysis of the talk using corpus analytic tools
(Wmatrix) and hand analyses showed that the instructors used terms related to
“sight” far more so than did the students. A further analysis using the “metaphor
identification procedure” (MIP) (Pragglejaz Group 2007) was attempted to deter-
mine which instances of the “sight” terms possibly conveyed metaphorical mean-
ing related to understanding. However, some examples appeared to indicate both
metaphorical and non-metaphorical meanings at the same time (e.g., an instructor
said, “I have some not so good responses to some questions and you will see what
people have done in the past” as he showed the student some examples of what
other students had done in the past). For this reason, the analysis revealed three
types of “sight” examples (i.e., metaphorical, non-metaphorical, and conflations
of both metaphorical and non-metaphorical).1
The instructors, furthermore, employed a larger range of “sight” terms (e.g.,
“see”, “look”, “focus”, “view”, “reflect”, “observe”, “notice”, “spot”, “visible”)
than did the students (e.g., “”focus”, “look”, “see”, “watch”). Overall, 40% of all
metaphorically used “sight” terms employed the lexical item “see,” followed by
“look” (33%) and “focus” (12%). Most interestingly, the students used “see” in dif-
ferent ways than did the lecturers. For example, students often used “see” in the
passive voice (e.g., “saw” and “have seen”), while the instructors used “see” in the
active voice. The instructors also did not typically use “see” to refer to moments of
understanding in the past (e.g., “I could see that . . . .”), but preferred to use “see”
metaphorically when talking about current or future actions (e.g., “You’re not
going to be able to see that they will be analyzed . . . ”). Many of the instructors’
uses of “see” were conflations (e.g., “Okay, yeah that fine you can also . . . as was
seen in the – in the film Fresca . . . ”), which mostly reflects cases of understanding
that involved visual materials such as films, books, exams, etc. Instructors also sig-
naled their understanding of students’ comments through the metaphorical use of
“see” (e.g., “right right I see . . . so will will you use grammar books?”), but the

1 The issue of “conflations” is a significant one in metaphor research, but much more atten-
tion is needed on this topic.
 The Particularities of Metaphorical Experience 21

students rarely employed “see” in this manner. In fact, the one instructor, who
was a native speaker of Spanish, also did not employ “see” as a way of signaling
understanding of students’ questions or comments. This finding suggests that vi-
sion may not be a typical way of referring to learning in Spanish, especially in the
context of Spanish academics talking with their students.
Students sometimes responded to instructors’ uses of sight metaphors by
also displaying their learning through similar metaphors (e.g., “I have to be
more clear” and “This introduction has to be more focused”). But a further
analysis of the discourse contexts in which UNDERSTANDING IS SEEING ap-
peared indicated that English and Spanish speakers think of this metaphor in
different ways. In many cases, students failed to pick up on the instructors’ use
of sight metaphors even when the instructors gestured in metaphorical ways
about UNDERSTANDING IS SEEING (e.g., fingers splayed outward as the hand
moved downward toward the desk to note what must be focused on).
MacArthur, Krennmayr and Littlemore (2015) offered several conclusions
from their analysis of the instructor-student conversations:

To a certain extent, the mismatch between the lecturers’ and the students’ metaphorical
use of the sight terms can be attributed to the different roles played by the participants in
these mentoring sessions. The lecturers see their role as supporting the students’ efforts
to learn through giving specific advice about how this can be achieved, and as we have
seen, sight terms, when used metaphorically, appear to comprise an important means of
communicating this advice (MacArthur, Krennmayr and Littlemore 2015: 214)

Most generally, they concluded the following:

Speakers of Spanish and English do not use the metaphorical expressions motivated by
the “UNDERSTANDING IS SEEING” mapping in the same ways: we have noted some over-
lap but also substantial divergence in terms of the frequency, forms, and meanings of
sight metaphors in this corpus” (MacArthur et al. 2015: 214). And, later on, “The experien-
tial base for the metaphor is thus probably very similar in Spain, England, the
Netherlands, or Sweden. However, this “primary” metaphor does not surface in the same
way among speakers of these different languages, as our data confirm (MacArthur,
Krennmayr and Littlemore 2015: 215).

MacArthur, Krennmayr and Littlemore’s (2015) examination of UNDERSTANDING

IS SEEING illustrates the tremendous context-sensitivity in the way primary
metaphors operate in both thinking and speaking about knowing and under-
standing processes. Even if different primary metaphors have strong embodied
roots via correlations between source and target domains, this fact alone does
not necessarily imply that all languages will instantiate particular primary meta-
phors in the same way in regard to people’s expressive speaking and writing.
22 Raymond W. Gibbs, Jr.

One may dismiss local variance in the ways primary metaphors appear in differ-
ent languages, or in different facets of discourse, as performance errors on the
part of speakers, or claim that primary metaphors are not all that important in a
theory of metaphorical thinking and language use. But these variations are criti-
cal to understanding the complex ways in which different languages express pri-
mary metaphors such that metaphor behavior is always shaped by specific
discourse practices, as seen in the particularities of how university instructors in-
teract with the second-language learning students in talk of their writing assign-
ments. The more general lesson here is that important regularities exist between
correlations in bodily experience and metaphorical language use. Nonetheless,
these regularities do not completely determine how real speakers necessarily link
primary metaphors to very specific instances of metaphorical language use in very
local discourse contexts. The particularities of the context and discourse practices
associated with a topic always matter in terms of when and how speakers use met-
aphorical language.

3 Learning the Meanings of Body-Part Metaphors

Attention to the specific details of second-language learners’ understanding of
metaphor is also seen in a study of young learners’ of English use and understand-
ing of body part terms (e.g., head, mouth, hand), when these conveyed figurative
meanings (e.g., “head of the stairs”) (MacArthur and Piquer-Píriz 2007). Although
students from ages 5 to 11 understand body part metaphors, they do so using dif-
ferent strategies at different ages. This general finding reveals another particularity
in second-language learners’ understanding of verbal metaphor.
The children were shown different photographs of scenes depicting a hammer,
a flight of stairs, a line of cars, and a bed, and were asked to circle the part of each
photo that was identified as the “head”. After doing so, the students provided ver-
bal explanations for their choices.
Students were best at identifying the head of a hammer and found it most
difficult to circle the head of a line of cars. A closer analysis of the students’ ex-
planation showed differences in why they made their “head” identifications. For
example, the five- and seven-year-olds clearly associated the metaphorical use of
“head” with the top or highest part of an object or scene. One five-year-old cir-
cled the top part of the stairs and then explained, “Because it’s the highest part
and it’s got to be the head because it’s at the top, the head is at the top of the
body.” Eleven-year-olds, on the other hand, described “head” as referring to the
general idea of importance. Thus, one 11-year-old circled the middle of the stairs
 The Particularities of Metaphorical Experience 23

as being the “head” and explained his choice in the following way, “The central
part because- I don’t know- because the head is- because it was the most impor-
tant thing, so- the staircase- it’s the central part”.
A closer look at the children’s choices and explanations revealed more subtle
differences in their metaphorical reasoning. For example, one child explained his
circling of the first car in a line of cars by saying, “The first one because the head
is at the top”. Yet some students, more precisely and also incorrectly, circled only
the roofs of the first and second cars as the “head,” and they reasoned, “Because
it’s on top of the first car” or “cos, the head is on top of the human body”. Thus,
these children relied on the common idea that the head is above the rest of the
body. In a different manner, some children explained their choices of the head in a
line of cars as being like the head of a snake or worm (e.g., “Because it’s like a
snake and snakes have their heads at the beginning”). One nine-year-old correctly
circled the first car and explained “Because it’s the first car and in a line the first
car is the one that gives orders- and the head- and our head is the boss”.
Overall, the younger children were more successful in identifying the head
of a staircase (82%) than the older children (56%). But the older children were
better at identifying the head of a line of cars (83%), and explained these
choices by referring to the idea that the head has a governing role, compared to
only 59% of the seven-year-olds and 42% of the five-year-olds. Even young chil-
dren exhibited some ability to explain metaphorical extensions of body part
terms, but children overall offered different explanations for why they under-
stood these metaphorical uses of words.
One could assume that the children’s different metaphorical reasoning
about the figurative meaning of body part terms is mostly a matter of their first
adopting incorrect understandings of what these terms imply in context. Later
on, these students presumably learned the correct reasons for body-part meta-
phors and performed similar to adults when using these metaphorical terms in
discourse. But MacArthur and Piquer-Píriz (2007) recognized important insights
in children’s incorrect reasoning about why body part terms can express meta-
phorical meanings. Similar to the writings of the great developmental psycholo-
gist Piaget (1923, 1936), who created a theory of cognitive development from the
analysis of children’s errors in different physical reasoning tasks, MacArthur
and Piquer-Píriz concluded that children’s so-called “errors” in their explana-
tion of metaphorically used body part terms reveal distinctive metaphorical
thinking patterns. Thus, metaphorical thinking comes in many different forms,
and adult correct usage of, in this case, body part terms, should not be inter-
preted as the only form of metaphorical competence. Second-language learners
may still be thinking in metaphorical ways, and do so in adaptive ways in cer-
tain contexts. The idea that younger second-language learners may think
24 Raymond W. Gibbs, Jr.

metaphorically when using body part metaphors, even if they do so differently

than older children and adults, is important for another reason. Younger chil-
dren are often thought to use metaphor entirely on the basis of perceptual simi-
larity. As they get older, they presumably acquire an ability to understand
relational connections between diverse domains, not just connections based on
perceptual resemblances. Yet even younger children learning a new language
can relatively quickly infer relational metaphorical mappings. Only by examin-
ing the particularities of metaphor experience can scholars recognize the adap-
tive, communicative value that metaphor has in diverse discourse contexts.
This is exactly what MacArthur and Piquer-Píriz (2007) were able to do.

4 A Case Study of Horse Riding Metaphors

In a different study, MacArthur (2005) explored how certain metaphorical themes
are used to regulate human action, especially in the context of helping people con-
trol potentially damaging events, as seen in the metaphorical idea CONTROL OF
AN UNPREDICTABLE/UNDERSIRABLE FORCE IS A RIDER’S CONTROL OF A
HORSE. Both Spanish and English have many conventional expressions in which
horse and rider serve as the metaphorical source or vehicle. Horses are sometimes
difficult to control and, as such, are often identified with other unpredictable or
undesirable events that are also difficult to control (e.g., human emotions, sexual
passion, disease). Riders exert control over horses through different tack or apparel
such as the bridle and reins, the saddle, stirrups, and spurs. These artifacts enable
riders to “curb” (i.e., to use the reins to pull the horse’s neck in one direction),
which can be readily mapped onto distinct target domains as seen in the following
metaphorical expressions in both English and Spanish (MacArthur 2005: 79–80):

“The publishers carefully curbed their comments on the charges”.

[SPEECH]
“Eating them more often might curb the appetite for fatty foods or confectionery”.
[APPETITE]
“A promise she had made to curb her temper and her excesses and maintain
a low profile”. [UNDESIRABLE EMOTION-BEHAVIOR]
“Good diet curbs cancer risk”. [DISEASE-DANGER]
“The government had an income policy of sorts and had curbed the growth
of public expenditure”. [UNDESIRABLE ECONOMIC SITUATION]
“Children whose instincts are to rebel often get curbed by their teachers”.
[PEOPLE-BEHAVIOR]
“Poseído por un impulso desbocado, se puso en pie”. [UNDESIRABLE
EMOTION-BEHAVIOR]
 The Particularities of Metaphorical Experience 25

“Su política expansionista de déficit público desbocado” [UNDESIRABLE

ECONOMIC SITUATION]
“Se siente fuera de control, desbocado”. [PEOPLE-UNDESIRABLE
EMOTION-BEHAVIOR]
“Sudoroso y satisfecho, víctima del magnetismo febril de sus jóvenes fans,
entre los que se encontraba algún que otro adulto desbocado, puso punto
final al recital”. [PEOPLE-BEHAVIOR]
“El mito del progreso desbocado que pusieron en boga los ilustrados”
[GROWTH-DEVELOPMENT]

These examples demonstrate how horse riding behaviors can be extended to

understand the control (or lack of control in the examples in Spanish) of dif-
ferent, sometimes abstract, events (e.g., DISEASE-DANGER and GROWTH-
DEVELOPMENT). Note also that the above English and Spanish conventional
expressions with “curb” convey a specific evaluative stance regarding the target
domain, namely concern with uncontrolled, “runaway”, progress/development
or disease. Some of the processes in need of control refer to external forces (e.g.,
people, economic situations, and disease), while others relate to the control of
internal forces (e.g., appetite, desires, and passions within an individual).
Other horse riding metaphors express how the rider controls the horse by
being “in the saddle” or “riding high in the saddle” or having “firm control of
the reins” (e.g., a boss in control of a company). Many conventional horse rid-
ing metaphors describe the undesirable or difficult to control aspects of some
force, again in terms of controlling a horse, as seen in the following examples
(MacArthur 2005: 82):

“A man must decide to either curb his appetites or to surrender to them”

“I am holding my thoughts on a tight rein and refusing to think ahead”
“I took myself in hand about a year ago and lost weight”.
“A partir de entonces, ella tuvo que sujetar las riendas de su hogar”.

If one is unable to control a horse, usually through the failure to use proper
riding apparatus or a rider’s improper use of that equipment, this can result in
several negative metaphorical events, such as seen in the following statements
(MacArthur 2005: 82–83):
“It is sheer, unbridled greed,”
“A return to the pagan ways of unbridled pleasure and carnal gratification”.
“They let their emotions run away with them”.
“He lets his imagination run away with him”.
“Kenneth’s aggressive nature has got a bit out of hand”.
“The Kremlin felt things were rapidly getting out of hand”.
“Es verdad: he perdido los estribos y he dicho tonterías”.
“¿Quieres tranquilizarte, por favor? ¡Estás perdiendo los estribos!”
“Se le fue la lengua”.
26 Raymond W. Gibbs, Jr.

“Soltando las riendas al temperamento racial”.

“En aquel desierto de sentimentalismo en el que vivía desde hacía meses,
la carta de Jean de Bornecort desbocó su corazón y su fantasia”.

At the same time, in the expressions seen below, horse riding apparatus can be
employed to bring about more forceful and highly desired actions as seen in
the following examples (MacArthur 2005: 83):

“The shortage of labor acts as a powerful spur to more economical methods of

production”.
“Her approval spurred him to enter the poetry contest”.
“You need someone to spur you on”.
“Federico Mayor Zaragoza espoleó ayer a la comunidad universitaria”.
“La NASA está espoleada por el hallazgo de indicios de vida en Marte”
“Espolea su imaginación”.

Out of control horses, nonetheless, still need to be brought under the rider’s
control, suggesting that some forces may be difficult to control through human
action (MacArthur 2005: 83):

“Dying of galloping consumption”

“The galloping inflation of the previous two years seemed to have been brought under
control”.
“In just six months, the country’s runaway inflation has been brought under control”.
“Una anciana con artritis galopante”.
“Tisis/cáncer galopante”.
“Paralizados en el seno de una verbosidad galopante”.
“Parecía imposible atajar una inflación galopante”.
“La insipidez del Barça se convirtió en apatía galopante”.

In this study, the entrenchment of many conventional metaphors regarding horse

riding results in aspects of the target domains being more easily understood than
aspects of the source domain (something that seems contrary to the common belief
that source domains are typically more familiar and concretely understood than
are target domains. Thus, some dictionaries offer no purely literal definitions of
words such as “unbridled” in English, or detail the literal meaning of “spur” fur-
ther down on the list of the word’s senses. Many figurative senses related to horse
riding may have been part of English, for example, for such a long time that these
have become fossilized.2

2 Contrary to MacArthur, I would maintain that these fossilizations are not necessarily “dead”
metaphors given how productive these horse riding metaphors still appear to be in many dis-
course contexts.
 The Particularities of Metaphorical Experience 27

Another interesting observation is that horse riding may still be relevant to

aspects of contemporary life in Spain and Britain, yet it is not highly salient in
everyday experience anymore. Deignan (2003), in fact, showed that cars are far
more salient for English speakers than are horses, but horse riding appeared to
motivate the creation, and continued use, of many metaphors, far more so than
is the case with cars. This imbalance suggests to Deignan that the link between
metaphor and culture is only indirect as many metaphors “allude to knowledge
shared as part of our cultural repository, but is no longer directly experienced”
(Deignan 2003: 220). MacArthur (2005: 88) further suggested, along this same
line of argument, that “Physical experience or embodiment cannot, thus, ac-
count for its general acceptance as an apt way of reasoning about control either
now or in earlier times. Rather, the widespread use of this (horse riding) meta-
phor in language implies that, for some reason or other”.
MacArthur speculated that the persistent salience of horse riding meta-
phors in Spanish and English may be partly due to the fact that these inherit
part of the more general conceptual metaphor LIFE IS A JOURNEY even if peo-
ple mostly do not have direct experience with horses or horse riding. Thus,
“commonsense knowledge of some domain cannot always, in these cases, be
taken to arise from first-hand experience, but must be a product of cultural
mechanisms of some kind” (MacArthur 2005: 89). Indeed, language itself ena-
bles speakers to be aware of different conceptual representations rather than
conceptual representations emerging from bottom-up embodied experiences.
But the language that persisted over time through “high-density social net-
works” may come from social groups with prestige or power, such as the horse
riding aristocracy within certain European countries. “Seen in this light, social
processes, rather than shared experience may have been responsible for the
spread of not only this (horse riding) but also other entrenched metaphorical
themes, such as those often mentioned in the literature on conceptual meta-
phor” (MacArthur 2005: 91).
This work on horse riding metaphors offers a valuable contemporary lesson
for scholars whenever they aim to discern the motivation for any metaphorical
concept or theme. As much as many scholars often immediately seek out embod-
ied experiences (e.g., horse riding) as the motivating foundations for conven-
tional metaphorical language, care must be taken to explore the importance of
historical, cultural, social and linguistic influences on why people speak in the
particular metaphorical ways they do. More generally, the motivation for why
many metaphorical words and phrases have the meanings they do is a complex,
interacting set of constraints, ranging from historical and cultural forces down to
faster-acting cognitive and linguistic ones (Gibbs 2017). Once again, explaining
the particularities of metaphorical language means that scholars should resist
28 Raymond W. Gibbs, Jr.

the temptation to correlate patterns, including errors in metaphor usage, with a

single level of analysis (e.g., embodiment alone).

5 Mixing Metaphors in Instructor-Second

Language Learner Discourse
Mixing metaphor is something speakers do all the time, mostly without any
awareness that they are doing so. As argued in MacArthur (2016a), people who
speak more than one language also mix metaphors, especially when speaking
to individuals who are also bilingual. Yet this mixing of metaphor is not at all
an indication of mental confusion or misunderstanding about what is to be
said. Instead, mixed metaphors often emerge from speech and writing given
various cognitive constraints and rhetorical considerations that individuals
face in different situations. Thus, mixed metaphors are not production errors
and are actually created for several possible reasons.
As MacArthur (2016a) emphasized, English is now an international lan-
guage or lingua franca, which means that the idea that native speakers adopt
“native-speaker norms” is now on very shaky grounds. Similarly, thinking of
any metaphor production as an error (i.e., the mixing of metaphors) is also mis-
taken given the ubiquity of metaphor in language as an incredibly adaptive rhe-
torical device. On top of the emerging belief that much of everyday abstract
thought is metaphorical, it makes little sense to see metaphor variations as vio-
lations of cognitive or communicative maxims. After all, the context in which
variations of verbal metaphors appear also makes metaphorical language use
of all sorts generally easy to produce and interpret (Gibbs 1994).
MacArthur (2016a) examined several instances of metaphors being employed
by speakers using English who were all native speakers of other languages, in-
cluding Spanish. She has argued from this work that metaphor must always be
examined as a discourse phenomenon and not just as a cognitive realization or a
simple linguistic instantiation of underlying conventional metaphorical thinking.
A number of scholars have argued that effective communication in a second
language greatly depends on adapting the specific linguistic and conceptual
practices of native speakers of the target language. However, MacArthur (2016a)
argued that variations in metaphor use are both critical to native and non-native
speakers’ use of language. The fact that native speakers often create novel meta-
phorical forms, or even mix their metaphors together, offers evidence against
the common belief that standardization and conventional use of metaphor are
the rule in both speech and writing. Being metaphorically fluent requires much
 The Particularities of Metaphorical Experience 29

more than simple acquisition of only the conventional forms of metaphor. Once
again, context enables speakers to express metaphorical ideas in a variety of
ways. Variation is particularly likely to arise in cases in which non-native speakers
are far removed from the places and situations in which a specific target language
first originated (MacArthur 2016a).
MacArthur (2016a) further explored these ideas through an examination of
the ways non-native speakers of English talk with one another in English. A
good example of this is seen in a conversation between three non-native speak-
ers of English, all of whom were business students. The conversation focused
on one student (S1), a male from Venezuela, talking about his reluctance to go
out on a date given a recent haircut he got. Two other speakers (S2, a Dutch
woman, and S3, a man from Indonesia) commented on S1’s reluctance, partly
through discussion of the metaphor employed to describe how S1 believed he
now appeared (MacArthur 2016a: 139):

S: And just have a good night’s sleep go on your DATE or whatever you
want to do tonight=
S: =no: i’m not going to cos i gotta ma:n i’ve shaved i look like a puppy (.)
S: <imitating> so maybe she likes pu:ppies </imitating≥
S =oh man no but this <><fast> you’re a woman </fast> </>
S <> is that </> what’s bothering you? <> <@> your shaving </@> @
</> @@@
S <> @@@@</>
S i don’t MIND a puppy look from time to time?
S <soft> from time to time </soft>
S it’s a ritual <@> man @ @ man’s shaving </@> (.)
S: yeah ()
S: i haven’t shaved in the longest time () like ALL of it. like it’s all gone ()
S: puppy ()
S: and they’re probably thinking something else <to S> no it’s my fa:ce
we’re talking about </to S> ()

S1 describes his facial appearance as being “like a puppy” because he was re-
cently shaved. S2 repeats “puppy” three times (e.g., “maybe she’d like pup-
pies,” “puppy like,” and “puppy”). Note, however, that “I look like a puppy” in
this case is unusual given S1’s negative appraisal of this possibility. English
speakers do not typically refer to “puppy” or looking “like a puppy” in this neg-
ative way. These non-native speakers of English have created a very local un-
derstanding of “look like a puppy,” which reflects what others have referred to
as a “conceptual pact” within conversation. Brennan and Clark (1996) describe
conceptual pacts as cases where speakers frequently refer to some referent in a
30 Raymond W. Gibbs, Jr.

novel manner for temporary purposes of that conversation. Once these concep-
tual pacts are established, they can quickly, yet temporarily, become conven-
tionalized with a specific talk exchange such that “I look like a puppy” later on
in the brief conversation becomes “puppy like” and later just “puppy”. In gen-
eral, these non-native speakers of English were able to quite efficiently use a met-
aphor to communicate their ideas even when the metaphor was not employed
with its most conventional meaning. As MacArthur (2016a: 150) claimed,

the choice of a specific metaphor in a particular communicative context is often the result
of local factors (for example, the topic under discussion) and arises precisely because in-
dividuals and groups of people have different experiences of all kinds –of physical envi-
ronments, of prior experience of discourse events, of concerns and interests, among
many others – that affect the kind of metaphors they employ in different circumstances.

She also suggested that this evidence of contextual variation in second language
learners’ metaphor use has pedagogical implications. Second language learners
should not merely aim to acquire native-speaker norms, such as in the use of con-
ventional English metaphors. Instead, novel or hybrid metaphors enable second-
language learners from different cultural and linguistic backgrounds to effectively
communicate in ways that simply parroting conventional metaphor may be unable
to do.

6 Determining the Basic Meaning of Words

My last appreciation of Fiona MacArthur’s work here focuses on the specifics
of metaphor identification. Several methods have been proposed for identify-
ing metaphorically used words in discourse such as MIP (Pragglejaz Group
2007) and MIPVU (Steen et al. 2010), These methods first ask analysts to deter-
mine the basic meanings of words before drawing inferences about whether
their contextual meanings differed from their basic meanings in terms of simi-
larity. Dictionaries are often employed to determine the basic mean meanings
of words. As a member of Pragglejaz, I easily recall the many debates the
group had over which dictionaries may be most appropriate to use when run-
ning MIP, as we observed significant differences in the ways dictionaries
listed and articulated their varying definitions, or senses, of lexical items.
MacArthur more fully considered the utility of “Macmillan English Dictionary
for Advanced Learners” (MEDAL) for identifying basic senses of potential met-
aphorical words in a corpus of English conversations between university in-
structors and students who were non-native speakers of English (MacArthur 2015).
 The Particularities of Metaphorical Experience 31

These 27 conversations focused on general academic topics, including students’

assignments and difficulties in understanding the course contents. Presented
below is one segment from the dialogues where the instructor was speaking. Note
the different “say” verbs (these are presented in boldface):

FL er and there (.) we want you to introduce each paragraph <fast> with
what we call a topic sentence </fast> and that topic sentence should
then (.) say what the main idea (.) in the paragraph is about (.) erm (.)
Oscar Wilde contradicts himself yet again when he says da da da (.)
then you say that (.) this is (.) that that first sentence tells us that (.) this
is what you will talk about in the paragraph (.) which means that a topic
sentence can never be (.) just a statement (.) Oscar Wilde (.) came from
Ireland (.) that is a bad (.) topic sentence because it’s just a statement
you you cannot (.) prove it or disprove it it’s just a fact (.) what you do is
that you pick out something that you want to (.) argue for or against or
(.) prove or (.) develop (.) that is the that is the topic sentence (MacArthur 2015: 125).

As Fiona summarized, “In this extract, it is clear that the lecturer is metaphorically
framing the topic of successful communication in an academic essay, by inviting
the student to ‘see’ the writing process as one of guiding the reader through the
text”. For example, the instructor noted that the student essay was “difficult to fol-
low” and that the student should “pick/lift out” the main idea to “make it clear,”
in order to “lead the reader into the world of [the] essay”. Thus, the instructor
guided the students by commenting, “we want to see you, we want to see Oscar
Wilde, and we want to see other critics also”. (MacArthur 2015: 129).
Many of the “saying” verbs in the above extract appear to express metaphori-
cal meaning by framing the students’ take on a journey which leads to the writer
guiding readers to a particular place, or understanding of Oscar Wilde. Overall,
the participants used the verb “saying” 16 times, “contradict” 9 times, “talk”
three times, “tell” once, and “argue” once. Yet not all of these related instances
of “say” are necessarily used metaphorically and so Fiona looked up their mean-
ings in MEDAL to see if she could identify a basic sense for each one that may
vary in some way from how the words were used in context. This proved to be a
difficult task.
For instance, what is the basic sense of the verb “say”? The first entry for
“say” stated that the verb means “to express something in words”. This makes
some intuitive sense, perhaps, but what is really meant by “express” or even
“words”? “Express” is defined as “to tell someone about a feeling, opinion by
speaking or writing about it”. “Word” is “a single unit of written or spoken lan-
guage”. Taken together, the meanings of most uses of “say” in the instructor and
student conversation did not appear to be metaphorical.
32 Raymond W. Gibbs, Jr.

When Fiona focused on “contradict” she found it defined as “to say the op-
posite of what someone has said is true”. Herein lies the famous circularity
problem in dictionaries, namely that one has to first define “say” in order to
understand the basic sense of “contradict”.
To “tell” someone something is defined as “to give information to some-
one,” which suggests a physical transfer of something about (information) to
another person. This contrasts with the real physical giving of a material object
to another person, which suggests that it is used metaphorically in the context
of the instructor and student conversations. However, note that the dictionary
definition of “tell” is ambiguous because much depends on what is understood
by “give”. The dictionary definition seems to assume that one can “give” some-
thing to another without any real physical transfer. But “give” may be vague
enough to allow nonphysical transfer in which case “tell” is not really meta-
phorical in the instructor and student exchange.
The meaning of “talk” was also found to be problematic given its dictio-
nary definition of “to communicate using words,” which may or may not indi-
cate the use of words to communicate in written language, not just oral
discourse. Nonetheless, the examples also “Can the baby talk yet?” and “Am I
talking too much?” were also presented by the dictionary. For these reasons,
the use of “talk” in the instructor-student conversations which refer to the stu-
dent’s written communication appears to contrast with the more basic oral
model of communicating sense of “talk”. Thus, the three uses of “talk” in the
student feedback conversations should, perhaps, be marked as potentially
conveying metaphorical meanings.
The meaning of “argue” is defined in two senses that are somewhat differ-
ent: “if people argue, they speak to each other in a challenging way because
they disagree,” and “to give reason why you believe that something is right or
true”. Dictionaries often give multiple senses for words and it is not always
clear which of these is assumed to be most “basic,” if there is even any assump-
tion that some senses are more basic than others in the construction of dictio-
naries. In this case, the first definition explicitly refers to spoken language,
while the second, once more, uses the ambiguous “give,” which has a meta-
phorical flavor. Should definitions that are metaphorical be seen as actually
providing a word’s “basic” sense?
As MacArthur (2015) pointed out, both MIP and MIPVU state the basic
meanings are typically “more concrete” and are “related to body action”. But
these additional specifying criteria do not really help matters much given that
both definitions of “argue” involve human agents engaging in body actions.
We are left, then, being uncertain as to how to classify “argue” in the context of
the instructor and student conversations.
 The Particularities of Metaphorical Experience 33

Finally, “speak” is defined as “to talk to someone about something,” and

secondarily as “to use your voice to talk”. If these senses were “basic,” then the
student’s statement “but when I speak,” in reference to something she actu-
ally wrote, appears to convey metaphorical meaning. Fiona is skeptical of this
conclusion.
Overall, her analysis resulted in two words (e.g., “say” and “contradict”)
being judged as not metaphorical, two words (e.g., “talk” and “argue”) being
marked as potentially metaphorical, while other words (e.g., “speak”) were
metaphorical framings of the writer’s task. If these results were valid, the in-
structor “was – confusingly – continually – switching from the literal to the fig-
urative in his talk about good academic writing. Or at least, this is what the
dictionary tells us he was doing” (MacArthur 2015: 134). But she noted that she
“perceives no such incoherence in the discourse” to really assert that the in-
structor was confused in this way (MacArthur 2015: 134).
In the end, she suggested that “one might simply conclude that going
through the painstaking procedure of identifying the basic meaning of each and
every word in some discourse with a dictionary in hand is simply silly (and, inci-
dentally, very time consuming)”. On the other hand, MacArthur seems to believe
that, “it might simply be that our basic resource (the dictionary recommended) is
to blame, because the definitions it provides do not help us as researchers decide
with any confidence if the contextual meaning of the words being used can be
understood in comparison or contrast with the basic senses of those words – for
the simple reason that in some cases we are not provided with any ‘basic’ sense
of those words, but rather with metaphorically-expressed, pedagogically-oriented
definitions, aimed at nonnative speakers of English” (MacArthur 2015: 134).
This detailed analysis of using a dictionary reveals several difficulties in ap-
plying standard identification procedures to identify metaphor in context. I do
not believe MacArthur aims to discredit attempts to systematize metaphor identi-
fication for different research purposes. Nonetheless, there is perhaps too much
of an emphasis in metaphor studies in applying identification procedures to
large corpora, where analysts then quickly draw generalizations about how meta-
phors appear in some discourse topic or genre. Doing so may overlook many of
the real complexities of metaphor use in very local contexts. Once again, Fiona’s
work offers a cautionary warning to scholars to always attend to the particular-
ities of metaphor, which gives appropriate consideration to the real speakers
(e.g., non-native speakers of English) in specific discourse situations.
34 Raymond W. Gibbs, Jr.

7 Conclusion
This appreciation of Fiona MacArthur’s metaphor scholarship has examined
some of the important empirical findings and theoretical conclusions seen in
several of her research articles. My lasting impression of her research is the de-
tailed manner in which she explores metaphor use in context, paying close at-
tention to some of the particularities of metaphor experience. This attention to
the details of exactly how specific groups of people (e.g., second-language
learners’ of English) learn to use metaphor in particular discourse contexts
(e.g., educational discourse) reveals a complex portrait of the interplay between
generalities and variations in the ecology of metaphor.
Metaphor scholars, as seen across many academic disciplines, often strug-
gle in balancing the desire of offering broad theories of metaphor, possibly rele-
vant to all people in all contexts, with the need to be completely accurate in
spelling out the nuances of metaphorical experience. Many of us in the meta-
phor business choose to embrace one side of this generality and variation di-
vide. However, Fiona MacArthur’s work is constantly striving to see generalities
in the detailed particularities of real people using metaphor in different circum-
stances. She is not just a metaphor scholar, but more importantly, someone
with a deep interest in how people communicate across cultures and languages
while using metaphor for adaptive purposes. For example, she has a keen eye
for detecting how metaphor is a tool for intercultural communication even
when the metaphors employed are novel or vary from what is often thought to
be accepted use. Yet this emphasis is not merely a matter of characterizing the
students’ learning of metaphor in terms of schematic systems of metaphor (e.g.,
primary metaphors, conventional patterns of metaphor within a target lan-
guage). Instead, the larger lesson here is that all metaphor use is inherently
contextualized, situated, and particular to individuals and specific discursive
practices. In this manner, the particularities of metaphor experience help us
understand the true generalities, and richness, of why metaphor is such an im-
portant conceptual and linguistic tool.
Her scholarship makes us stop and think more closely about what we are
doing in our scientific and applied studies of metaphor. She pushes us to see
the interconnectedness between language, culture, and minds and to have a
greater appreciation for the humanity of metaphor as a fully human endeavor,
not just an abstract topic that is examined from different disciplinary perspec-
tives. Reading her many articles, book chapters, and edited collections also
give us a strong sense of her personal and scholarly commitment to “getting
metaphor right” by recognizing the limitations of our methods and the constant
need for broader contextualization of how we study real-life metaphor use.
 The Particularities of Metaphorical Experience 35

I salute Fiona for her diverse scholarly achievements and for what she has
taught many of us about the complex realities in studying metaphor in all of its
glorious, messy, and all so beautiful details. She offers us a distinctive, ideal
model on how to best study metaphor.

References
Brennan, Susan and Herbert Clark, 1996. Lexical choice and conceptual pacts in conversation.
Journal of Experimental Psychology: Learning, Memory, and Cognition 22, 1482–1493.
Deignan, Alice. 2003. Metaphorical expressions and culture: An indirect link. Metaphor and
Symbol 18. 255–271.
Gibbs, Raymond W. 1994. The poetics of mind: Figurative thought, language, and
understanding. New York: Cambridge University Press.
Gibbs, Raymond W. 2017. Metaphor wars: Conceptual metaphor in human life. New York:
Cambridge University Press.
Grady, Joseph. 1997. Theories are buildings revisited. Cognitive Linguistics 8. 267–290.
Ibarretxe-Antuñano, Iraide. 2013. The relationship between conceptual metaphor and culture.
Intercultural Pragmatics 10. 315–339.
MacArthur, Fiona. 2005. The competent horseman in a horseless world: Observations on a
conventional metaphor in Spanish and English. Metaphor and Symbol 20. 71–94.
MacArthur, Fiona. 2015. On using a dictionary to identify the basic senses of words. Metaphor
and the Social World 5. 124–136.
MacArthur, Fiona. 2016a. When languages and cultures meet: Mixed metaphors in the
discourse of Spanish speakers of English. In Raymond W. Gibbs (ed.), Mixing metaphor,
133–154. Amsterdam: John Benjamins.
MacArthur, Fiona. 2016b. Overt and covert uses of metaphor in academic mentoring in English
and Spanish undergraduate students at five European universities. Review of Cognitive
Linguistics 14. 23–50.
MacArthur, Fiona, Tina Krenmayr & Jeannette Littlemore. 2015. How basic is UNDERSTANDING
IS SEEING when reasoning about knowledge? Asymmetric uses of sight metaphor in
office hours’s consultations in English as academic lingua franca. Metaphor and Symbol
30. 184–217.
MacArthur, Fiona & Ana M. Piquer-Píriz. 2007. Staging the introduction of figurative
extensions of familiar vocabulary items in EFL: Some preliminary considerations. Ilha do
desterro: A Journal of English Language, Literatures in English and Cultural Studies 53.
123–135.
Pragglejaz Group. 2007. MIP: A method for identifying metaphorically-used words in
discourse. Metaphor and Symbol 22. 1–40.
Steen, Gerard J., Aletta G. Dorst, Berenike Herrmann, Anna A. Kaal, Tina Krennmayr & Trijntje
Pasma. 2010. A Method for Linguistic Metaphor Identification: From MIP to MIPVU.
Amsterdam: John Benjamins.
Graham Low
Taking Stock after Three Decades:
“On Teaching Metaphor” Revisited
1 Introduction
In the early 1980s, I read R. J. Alexander’s seminal (1983) paper on the impor-
tance of helping English as a Foreign Language (EFL) learners come to terms
with metaphors, connotations and allusions, as well as early work on conceptual
metaphor, like Lakoff and Johnson’s (1980) Metaphors we live by, Lakoff’s (1982)
Categories and cognitive models, and Lakoff and Kövecses’s (1983) The cognitive
model of anger inherent in American English. This, plus some research I was
doing at the time on metaphorical expressions of anger in English, led me to the
realisations (a) that metaphor was a more important component of English use
and lexis than EFL theories, courses and tests were at that time acknowledging,
(b) that metaphor was surprisingly complex, (c) that most applied linguists
seemed to have little idea, or interest in, what to do about it, and (d) that some
ideas from conceptual metaphor theory seemed potentially very relevant to for-
eign language teaching, but the theory did not seem to explain all metaphoric
use, and even in its early stages conceptual metaphor theory and terminology
seemed to need simplifying or reinterpreting for use in the language classroom.
I therefore decided to have a first stab at formulating a number of points that
I felt were important in foreign language teaching and learning. These were di-
vided into four topic areas: What is metaphor? What functions does metaphor
have? What do we want a skilled L21 user to be able to do with metaphor? How
might L2 metaphor usefully be taught (and to a lesser degree, tested)? The result
was published in 1988 in the paper On teaching metaphor. It has been much cited
in the intervening years, and I reviewed my position on metaphor and education
generally in 2008, but until recently (e.g., O’Reilly 2017), many of the ideas in it
were not empirically tested or developed. Thirty something years on (2019), thus
seems like a good point to look back at the 1988 paper and ask whether I still
agree with it, and to try to establish if there were any major gaps or gaffes. I shall

1 I shall use the acronyms L1 and L2 conventionally to mean first language and target language.
Similarly I will use FL adjectivally to mean foreign language. The label “native speaker” will be
avoided.

Graham Low, Selby, North Yorkshire, UK, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-003
38 Graham Low

treat the points using the same topic areas used in 1988, but limit the viewpoint
taken and examples in the main to EFL learners. I should perhaps add as a pro-
viso that in this sort of reflection on one’s previous work, however objective one
tries to be, there remains an inevitable element of subjectivity.

2 What Is Metaphor?
The 1988 paper was written for applied linguists not metaphor specialists, so it
adopted a definition, designed to be easily comprehensible, that more or less
covered various theoretical positions: “Treating X as if it were, in some ways, Y
(where X and Y can show varying degrees of difference)” (1988: 126).
It seems reasonable to wonder whether this definition is adequate, allowing
for some lexical reinterpretation, as a foreign-language (FL) classroom or cour-
sebook explanation, for older learners (meaning older teenagers and adults) at
least.
On the positive side, it allows for metaphoricity to be a cline (the greater
the apparent difference between X and Y, the greater, or at least clearer, the
metaphoricity), for metaphor to be treated selectively on the “surface” or in
terms of underlying X IS Y propositions, for metaphor to be primarily unidirec-
tional (TIME IS MONEY being different from MONEY IS TIME) and for metaphor to be
expressed visually, gesturally, or auditorily, as well as linguistically. The use of
“in some ways” also encompasses the need, stressed later by both Cameron
(2003) and the Pragglejaz group (2007), that to be metaphorically used, some
plausible account of the basic sense/literal-to-figurative change (implying one
or more connection or correspondence) needs to be specifiable, such that the
randomly named particle labels in Physics, like colour and charm, for example,
remain non-metaphoric. Lastly, where subsidiary clauses continue to treat X as
Y (as in “Anne seemed a rose, that bloomed in June and faded later in the
same year”), the reader can choose to see the clauses as “inheriting” (as Steen
1999 put it) the metaphoricity of the main clause, or identify rose, bloomed, and
faded as independently metaphorically used.
With hindsight, a treatment definition has the added advantages that it
allows for expressions such as “John is (the new) Fred” to be metaphoric or
not, depending on how one thinks of Fred, and it allows for multimodal real-
isations where X is in one modality and Y is in another (as in many cartoons,
advertisements and graffiti). There is also no requirement for the speaker/
writer and the hearer/reader to agree on the meaning of an utterance, leaving
the hearer/reader to guess on the basis of the available contextual evidence
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 39

whether a metaphor is involved or was intended, and if so, how.2 It would

seem to also cover the ideological position where a believer sees, say, heaven
as literally in the sky, but a non-believer sees the same link not as truth but as
metaphorical (to the non-believer, the “as if it were Y” applies).
Neither blending theory (e.g., Fauconnier and Turner 2008) nor primary
metaphor (Grady 1998; Grady and Johnson 2002) were around in the early
1980s. To take blending theory first, while some lexis that FL learners are likely
to encounter represents clear examples of figurative + non-figurative blending
(like frogman for example), these can be easily explained as a man treated as if
he were in some ways a frog. I would argue that the complex and abstract ter-
minology involved in blending theory and the difficulties with specifying “ge-
neric space” in particular, rule out its use in the majority of FL situations.
Primary metaphor constitutes a different ‘problem’, as several primary
metaphors (or variations on them) are highly relevant to FL contexts and easily
understandable. Thus PSYCHOLOGICAL (DISTANCE) IS PHYSICAL DISTANCE and SOCIAL
(DISTANCE) IS PHYSICAL DISTANCE are frequently involved in common interaction
expressions, explaining why “This is John” (vs “It’s John”, or “That’s John”)
works as a social introduction – making John not just visible but a member,
even if temporary, of the interaction group – and a worried listener might an-
swer the phone with “Who’s that?” or “Is that John?” (rather than “Is this
John?” “Is it John?”, or “Are you John?”). Many primary metaphors may well be
grounded in metonymic relations, as Grady claims, but for practical purposes
they can still be described as “treating X as if it were Y” even if one has to hunt
around and look “under the surface” to isolate appropriate Xs and Ys. There
would seem no need, however, to introduce either of the labels “primary meta-
phor” or “primary scene” to FL learners. In short, primary metaphor can be
dealt with in FL contexts using the 1988 treatment definition.
Sharifian (2017: 18 and elsewhere), arguing from a Culture Linguistics stance,
again developed after 1988, has argued that it is preferable to formulate X IS Y as X
AS Y, to emphasise the partial and culturally-constructed linking of the two (e.g.,
rephrasing Yu’s [2007: 27] wording to HEART AS RULER OF THE BODY in Chinese) –
something which again dovetails neatly with the “treatment as Y” definition.
Cameron’s (2003 and elsewhere) use of complexity theory to explain how
texts have different layers or narratives, with metaphors relating to different
layers and resonating with current and earlier contexts to differing degrees, fits

2 The notion of “deliberate metaphor” (Steen 2011a, 2011b) remains controversial. I am here
simply suggesting (with Deignan, Littlemore and Semino 2012: 22) that there are times when
the author’s aims and intentions are worth hypothesising.
40 Graham Low

the “treatment as Y” definition. Moreover, the idea of a metaphoric expression,

for example, chivalrous, reflecting, say, a cultural context from the past, while
at the same time having a role, such as “old-fashioned, geriatric” behaviour, in
the present discourse needs to be covered in FL situations. The hearer/reader
may well, of course, be guessing whether a link with the past is intended and
whether the speaker is trying to make a point or not about the present.
While the 1988 definition thus has some advantages and covers several situa-
tions that are likely to recur in FL contexts, including ideas developed post-1988,
there are a number of problems. The main one is that it excludes simile. “Treating
X as if it were in some ways Y” implies the identification (albeit partial3) of Y with
X. Simile, however, involves comparison (e.g., Teaching is like cooking), not identi-
fication (e.g., Teaching is cooking). There is good reason to keep metaphor separate
from simile, in that (1) not all similes are best treated as figurative – Your house is
like mine is most likely to be seen as a non-figurative, or literal, comparison – (2)
“X is Y” and “X is like Y” can have different meanings – examples can be found in
Littlemore and Low (2006: 42–43), including Life is a joke (definitely not funny, on
a par with You’re having a laugh!) versus Life is like a joke (where it could well
be) – and (3) the X-Y relationship is clearer with simile than the corresponding
metaphor (thus I suggested in Low (2015: 22) that research prompts in elicited met-
aphor studies could reduce confusion by using X is like Y format).
However, at the same time, it is often desirable to ally simile with metaphor,
and to be able to see a simile such as Teaching is like cooking as close in meaning
to Teaching is cooking. The justification is that (1) it is not always easy to estab-
lish whether markers like ‘kind of’ are flagging metaphor, simile or literal subca-
tegorisation (Low 2010: 293) and (2) speakers/writers often mix the two, either in
stretches of discourse/text involving expansion or development of a figurative
link, or in repair sequences. An example of the latter comes from a UK university
lecture on business strategy (the numbering represents tone groups):

1721 the I mean organisations are like sort of

1722 er you know
1723 have huge buffering power
1724 to actually stop any change
(from Low 2010: 298, example 4)4

3 Veale (2012: 335) calls the identification “absolute”, but this would seem to imply more than
“in some ways” and imply a sameness between X and Y.
4 The source is the British Academic Spoken English (BASE) corpus. My thanks to Hilary Nesi for
permission to cite extracts for research purposes. The BASE corpus was developed at the
Universities of Warwick and Reading under the directorship of Hilary Nesi and Paul Thompson.
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 41

So, with hindsight, the 1988 paper could usefully have linked simile and
metaphor better, and in a more detailed way. In self-defence, I can only say
that when I wrote the paper I had not examined the complex roles that the two
play in discourse.
The second main problem with the treatment definition is that it appears to
cover numerous examples of metonymy. Thus in The sandwich lunches sit at the
back of the room or Put the tools in the Ford it could be argued that the eaters are
being treated as the prototypical objects eaten, or that the vehicle/product is
being treated as the manufacturer/ producer. While the close connections between
metonymy and metaphor became abundantly clear with primary metaphor, and
there are (other) cases where it is not clear which of the two is involved, it was not
my intention in 1988 to conflate them. Perhaps a compromise position would be
to apply the treatment definition to both, but to regard part-whole examples as
metonymy. Close association examples could be metonymic or metaphorical de-
pending on the degree of closeness, or whether the hearer/reader felt that connec-
tion or discrepancy/difference was being emphasised. One could then explain
This firm needs some new blood (= we need new people, and we need new ideas)
as treating both people and ideas as blood, with people as blood being part-whole
metonymy and ideas as blood being either metaphor or (populist) close associa-
tion metonymy (greater blood flow to the brain generates more ideas).5
A third potential problem, and one highly relevant to FL learners, is the
variable metaphoricity of technical terminology, where a specialist author may
begin by treating a technical term as metaphoric, but in subsequent mentions
treat it as literal or minimally metaphoric, whereas a non-specialist reader sees
it throughout as highly metaphorical. Deignan, Littlemore and Semino (2013: 8)
would treat this an example of the more general topic of “insider discourse”. I
do not think this is a genuine problem for the “treatment” definition, however,
as individual words and phrases can be resubjected to it at different points in
the text, and it can be applied using the viewpoints of different protagonists.
Thus slang or jargon can at times be seen as talking to insiders, while excluding
outsiders.
Lastly, one might wonder if proverbs, which were not dealt with in the 1988
paper, represent another problem area for the treatment definition. I shall take

Corpus development was assisted by funding from BALEAP, EURALEX, the British Academy and
the Arts and Humanities Research Council. The URL is http://www2.warwick.ac.uk/fac/soc/al/re
search/collect/base/.
5 This is not to disagree with the point made by Steen (2007: 59) that when identifying exam-
ples of figurative use in research exercises, it is preferable to identify metaphor and metonymy
separately.
42 Graham Low

“a rolling stone gathers no moss” and “it’s the early bird that catches the worm”
as examples. In the first, a person who does not settle in a job/community is
treated as a stone that rolls (in a wood? Or where?). In the second, a person who
gets up early, or starts a task before others have committed to it, is treated as an
early-rising bird. Like most English proverbs, they have two parts and involve an
action > result storyline (or scenario6), so treating a job, home and/or family as
moss and a reward/payoff as a worm seems adequate for most FL purposes. In
practice, the FL learner may additionally be faced with having to identify the
proverb, as only the first half is often explicit, as well as establishing the moral
force of the utterance (eating worms may always be a good practice, but do we
want stones to have moss?), but neither of these tasks affects the viability of the
definition. Lakoff and Turner (1989: 162–166) suggest a complex creation/inter-
pretation pathway for proverbs via the conceptual metaphor generic is specific.
However, I would argue that most adult and teenage FL learners will know at a
general level how to interpret a proverb (equivalents seem to exist in most lan-
guages); what they need is to know is if a proverb is being referred to, and if so
which one, what its ‘dictionary’ sense is, how this relates to the current situation
and whether irony is involved. A treatment definition, plus explanations of the
scenario involved, is likely to be adequate for most FL purposes; recourse to com-
plex generic and specific logic may well confuse rather than help.7

3 What Core Functions Does Metaphor Have?

The 1988 paper isolated half a dozen core functions, likely to be encountered
by many learners. However, there are many types of learner, with differing
wants and needs concerning the L2, so the functions of metaphor that are most
likely to be relevant to learners does not constitute a fixed set (making it hard
to list functions as part of a general L2 proficiency profile). The 1988 set is nev-
ertheless worth briefly reviewing.

6 I use “scenario” for narratives at lexical level and “story” for narratives created at discourse
level. With proverbs, the two are more or less equivalent.
7 Sullivan and Sweetser (2010) argue that a blending theory approach can provide a better ex-
planation than conceptual metaphor theory, accounting for both more- and less-metaphorical
linguistic examples in a principled way. While I take their point, I would still argue that the
terminological complexity of blending theory and the difficulties with mapping generic spaces
rule it out for any but the most linguistics-oriented FL learners.
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 43

To Enable People to Talk at All about X

Abstract concepts like emotions, the mind, time, colour hue, saturation and lu-
minosity, life and death, friendship are all heavily metaphorised, often viewed
in terms of containers, journeys, locations, or rivers/seas/fluids. Many of these
topics have exponents that come into L2 courses early (e.g., a dark blue, vs a
deep blue, or a rich blue etc) and recur at all levels.

To Relate or Structure X

Clearly the famous conceptual metaphors with large amounts of transferred

structure, such as ARGUMENT AS WAR or LIFE AS A JOURNEY can be seen as structur-
ing aspects of argument or life reports; similarly, treating the mind as a com-
puter, a building or a person can help people develop (or reflect) mental
models of the mind. Less obviously, perhaps, there is the way vertical UP-DOWN
scales are used for so many concepts, or how scenarios involving actors, ac-
tions and events are used. One “problem” with structuring via metaphor is the
inconsistencies, or structural gaps, which show up in the collocational tenden-
cies of conventional metaphorical expressions. Thus I can “put my mind to a
problem”, “apply it to a problem” or “get my brain onto a problem”, but I can-
not conventionally move it (mind or brain), push it, place it, or put it back.
Again, a bank account is fairly consistently a boxlike container in English: you
open/close an account, pay into it, have money in it, and take money out of it
(though you don’t open it immediately prior to paying into it), but it sounds
very odd to talk about a balance being calculated inside an account. And “put
it on my account” represents a different type of account (one which is a list of
money owing) and seemingly a different underlying metaphor (a two-
dimensional “slate”). Collocational inconsistency is so widespread in English
and such a minefield for FL learners, that creating sets of what you can and
cannot say seems well worth the effort.

To Extend Thought

Extending thought is perhaps only relevant for some learners, such as those
following university courses in the L2, who are exposed to analogies for tech-
nical concepts or different theories in lectures and readings, and for whom
learning about critical thinking via metaphors is important. But even elemen-
tary-level learners can usefully extend conventional English via playing with
44 Graham Low

“Love is . . . ”, or “Anger is . . . ”; and the use of direct elicited metaphor

(“Learning English is . . . ”8) to help learners think about their own learning
(e.g., the papers in Wan and Low 2015; or the evaluation in Low 2017) seems
to be successful if appropriate methodological precautions are taken.

To Dramatise X

Metaphor is often used in discourse, in genres like verbal reports, or advertise-

ments, along with hyperbole, to dramatise an event, by exaggerating it positively
or negatively. Sometimes this involves the creation of metaphor-based stories,
(something not considered in the 1988 paper). Dramatisation occurs at both dis-
course and lexical levels, and learners need to be able to cope with both.
Dramatisation can be achieved lexically by use of markers or “flags” such as
literally: e.g., “At this point she literally hit the roof”. This sort of dramatization is
easy for FL learners to achieve, as long as they can cope with literally having two
very different senses!
Dramatisation at discourse level seems less frequently taught on EFL
courses. As a brief illustration of creating a metaphor-based story, which relies
on combining conventional metaphoric phraseology with discourse, the follow-
ing anecdote about whale meat is embedded inside a longer anecdote in the
previously cited UK university lecture on business strategy. The speaker is part
of a group in an Icelandic restaurant:

1347 in this actual room

1348 two two Norwegians
1349 and said the wonderful classic phrase (.)
1350 what is so wrong
1351 about killing minky whales
1352 and and to see the sort of head of Greenpeace
1353 virtually go through the ceiling
1354 before coming down (.)
1355 and sort of killing all in his path (.)
(from Cameron, Low and Maslen 2010)

8 As argued Sec. 2, a teacher might well generate a clearer prompt here by using a simile
(Learning English is like . . . ).
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 45

The speaker creates a small three-event story involving the head of Greenpeace
getting angry, via a conventional idiom (go through the ceiling), plus an intensi-
fier (virtually) flagging that the idiom is metaphoric as well as forceful, and
then acting wildly (killing all in his path). The metaphoric idiom and the rela-
tively fixed rampaging animal phrase are connected narratively by inserting a
bridge, coming down (which creatively extends the metaphor in the idiom), and
the three events are held together rhetorically by being framed between two po-
lysemous uses of killing (lines 1351, 1355).

To Prevaricate, Avoid Clarity, Not Actually Commit Oneself

The example in the 1988 paper came from Lerman’s (1985) study of a US presi-
dential speech and journalists’ reactions to it, whereby, having said nothing
literal the speaker is not committed to, or responsible for anything concrete.
While this function may not be appropriate for many younger and/or lower-
proficiency learners, it seems highly useful as regards both comprehension
and production for upper intermediate and advanced learners, especially
those studying politics.

To Allow/Facilitate Discussion of Emotionally Charged Topics

There is abundant evidence that writers/speakers often use more (though not
necessarily more creative) metaphors, leading to short bursts or clusters, when
showing emotions, and/or at “emotional centres” of discourse (Lerman 1983: 3).
A good recent example comes from the editor of a photography magazine enthu-
siastically introducing the issue; the opening sentence has seven or eight meta-
phorical words or phrases, but the rate then declines fast:

Have I got hues for you

Vivid images infused with rich hues sit alongside striking archive photogra-
phy in an issue that showcases the best of both worlds.
As he prepares for a headline appearance at The Photography Show, Miles
Aldridge explains the role of colour in his work. . ..
(K. Morgan, The RPS Journal, March 2018: 161)9

9 My thanks to Kathleen Morgan for permission to cite the extract from her editorial.
46 Graham Low

EFL learners seem to be increasingly faced with tasks requiring subjective

or emotive responses of various sorts. Littlemore et al. (2014) in their study of
200 written exam scripts by German and Greek EFL learners found a qualitative
difference at B2 (upper intermediate) level. Below B2, metaphors primarily in-
volved function words like prepositions, but at and above B2 they involved
open class words more. As MacArthur (2010: 419) put it, this seemed to be a
task effect, inasmuch as at/from elementary level, “in response to tasks requir-
ing that they state their personal opinions on certain issues or highlight their
personal significance, learners needed to use metaphorical language.”
It is also worth reiterating the research and/or therapy use of metaphor.
Wan (2011, 2012), Fisher (2013) and others have, as noted above, used direct eli-
cited metaphor with reasonable success to aid feedback on courses, both aca-
demic and foreign language.

Other Functions

The 1988 paper briefly mentioned the use of metaphor to summarise, and/or disen-
gage with a topic, and with hindsight, two related aspects of this function seem
worth stressing as useful to L2 learners. First there is the finding by Drew and Holt
(1995) that British and American speakers wanting to change topic in a conversa-
tion (or end the conversation) frequently add a summarising idiom or metaphor,
which the listener can accept or reject. Secondly, there is the finding (Low 1997)
that written texts such as editorials, novels or book reviews not infrequently add
metaphoric expressions near the start and end of text units like sections and
paragraphs, thereby creating metaphoric borders or frames, some of which mark
changes of topic. The RPS Journal extract above is an example of an initial text
border.
One last core function not mentioned in the 1988 paper is that of reassuring
the listener/reader and relating unfamiliar topics to familiar ones. In one sense
this relates to vast amounts of conventional metaphor, where abstract Xs are
associated with concrete, familiar Ys often based round the human body and
common bodily events. But an aspect of reassurance/familiarisation to which
listeners and readers of “authentic” English are likely to be frequently exposed,
but which is not commonly dealt with on EFL (or FL) courses with any degree
of systematicity, is the use of animacy or personification. As an example, in
Low (2005) I showed how even a fairly formal popular science article explain-
ing research in Darwinian evolution repeatedly (and ironically given the topic)
animated/personified elements undergoing change, seemingly allowing them
intentions. The moral is that animacy is a common instructional device, and its
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 47

use can override the logic of the topic being instructed. A second genre, to
which EFL learners are likely to be exposed, and which, in the UK at least, in-
volves repeated animacy, is TV weather forecasting, as in:

“the (cold) front nudges its way across (the UK) before it makes itself at
home (+ location)”
(O. W. Evans, BBC1, Late Look North, 20 Feb 2018)

In both the popular science and the weather forecasting cases, the aim seems to
be the opposite of “insider language” (“outsider language”?), or rather to give
outsiders non-threatening access to insider concepts, processes or argumentation.
It should be added that metaphor can serve the reverse function, via what
Halliday (1985/1994; see also Steen 2007: 85–88) called grammatical meta-
phor, where, say, an action, process or event is treated as an object. To the
extent that this involves a change of syntactic class, like nominalisation, the
label seems valid. The example below contrasts three (imagined) utterances:

(a) He left. Then the family had breakfast. >

(b) After he left, the family had breakfast. >
(c) After his departure, the family had breakfast.

The effect of (a) > (b) > (c) is to increasingly background the action of leaving, and
to increase the implication that the reader already knew he would leave, or had
left (thereby emphasising familiarity with the action by increasing, rather than de-
creasing, abstraction). Grammatical metaphor would thus seem to cover academic
arguing via abstract nouns, a practice FL teachers might want to encourage or dis-
courage depending on the situation of the learners and purpose of the course.
To sum up, with a couple of additions, the core function list seems as valid
today for foreign language learning as it did in 1988.

4 Metaphoric Competence
“Competence” can be envisaged in several ways, from the social to the psycho-
logical, focussing on the process or the result, and from abilities directly related
to the action/result to ones underlying it and only indirectly related to it. The
1988 paper took an approach based round a range of discourse-based skills,
which could in theory at least be directly tested, though how test scores could
be combined to form an overall index remained (and still remains) unclear.
48 Graham Low

Indeed, the whole topic of how a notion called metaphoric competence fitted
into general language competence (i.e. via proficiency testing, rather than test-
ing what had been taught) was not elaborated on. The important question of L2
identity was also overlooked in 1988; how far particular learners see themselves
as the sort of person who uses a lot of metaphor and plays with ideas/words in
the L2 may well influence the discourse they produce, as well as how they inter-
act (which will in turn affect their proficiency test scores); while there has been
research into differing identities in L1 and L2, I have not seen studies relating
this specifically to metaphor teaching, learning or use.
Metaphoric competence probably needs to be seen as encompassing more
than discourse (even though the skills envisaged in the 1988 paper had implica-
tions for grammar and lexis too). Indeed, when Littlemore and Low (2006a,
2006b) explored general models of communicative competence commonly used
in foreign language teaching/learning at the time, especially that of Bachman
(1990) and Bachman and Palmer (1996), it became clear that aspects of meta-
phor in lexis or discourse needed to be addressed by learners at all stages of
proficiency and in all four dimensions of competence (linguistic, sociolinguis-
tic, discourse and strategic).
Littlemore (2001) took a different approach to testing metaphoric compe-
tence from Low (1988) and developed a series of psychologically-based tests,
some of which tested underlying skills thought to correlate with metaphoric
competence. Apart from test sets by Azuma (2003, 2005) and a discursive re-
view of metaphoric competence by MacArthur (2010), little research has been
published at the time of writing to reconcile and compare empirically the social
and psychological approaches to metaphoric competence.
Until it is, I think that listing things that learners need to be able to do in and
with the L2 represents a valid, if not comprehensive, approach to teaching meta-
phoric competence, and for general purpose FL contexts (rather than specific pur-
poses contexts) I would endorse the general summary I gave in Low (2008: 221)10:

Productively, speakers need to know how to use non-specific metaphor to “decouple”

from a narrative or conversational topic, in order to summarise it, evaluate it, withdraw
gracefully from the argument, or simply change topic. Receptively, listeners need to be
able to pick up on the previous speaker’s metaphor, use their knowledge of the target
culture and discourse practices to guess what the speaker is implying, and choose to
“run with” the metaphor, extend it or even close it down. They need moreover to be
aware of the implications of the strategy they themselves adopt. They need to recognise
where style jumps take place, where speakers and writers stop being metaphoric. They
need to recognise where the speaker is extending or elaborating beyond conventional

10 My thanks to Cambridge University Press for permission to cite this extract.

 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 49

language and why – are they being friendly, humorous, sarcastic or even addressing a
third party? Learners need to recognise where the speaker is avoiding a topic, or refusing
to take responsibility (Lerman, 1983). Lastly, they need to recognise when texts or speak-
ers are operating simultaneously on multiple levels (as in many, possibly most, jokes,
advertisements and banter) and to establish what effects and messages are being hinted
at on each level.

To the above one could add the need to understand euphemisms (like conven-
tional and mildly creative expressions for lavatory) and their appropriate con-
texts of use, and to know how to intensify (or tone down) a metaphoric
utterance (e.g., by using really, seriously, literally, kind of etc.).

5 Teaching and Learning Implications

I begin this section by noting the stark conclusions drawn by MacArthur (2017)
that after all the years of work by linguists of various sorts, there has been little
attempt to integrate metaphoric competence into national FL syllabuses, exami-
nation profiles, or textbooks. Despite the existence of a few standalone materials
and activity books/sets, metaphor (and especially its use in discourse/pragmat-
ics) is “still marginal” (ibid.: 418) to mainline FL language teaching and testing.
Limiting FL metaphor teaching to vocabulary, with a concomitant lack of
focus on teaching metaphor use in discourse, is clearly worrying, given the list
of skills and the examples in the previous two sections. In 1988, I was very in-
terested as an applied linguist in instructional activities that involved multiple
texts and multiple tasks, based round one or more “storylines”, resulting in a
mixture of text types/genres, with jumps of formality, emotion, and mode, and
allowing for multimodal products like advertisements (drafts as well as final-
ised) with pictures and texts. The process and output would be enhanced by
small group work, as different participants with different perspectives could
be represented.11 The advantage was that this type of activity covered the met-
aphor-related skills listed above, but the downside was that they were very
time-consuming to create, took up a lot of class time, and did not by them-
selves explain how the metaphors operated. While task-based learning has
over the years developed certain aspects of multi-text, multi-task activities, I
have rarely seen tasks that foreground metaphor-related skills. The possibility
of creating them, even shortish ones, as part of mainline instructional pack-
ages remains, though.

11 Outlines of several possible multi-text, multi-task activities were given in (1988: 139–141).
50 Graham Low

Both the 1988 and 2008 papers touched on the key pedagogical question of
whether the fact that you can treat, say in horror as metaphorical (PSYCHOLOGICAL
STATE AS CONTAINER) means that you should necessarily teach it as such. My expe-
rience with talking to language teachers has shown that there is still considerable
unwillingness to teach much prepositional use or tenses as metaphorical. There
is likely to be no ideal solution to this problem, but one might suggest that there
are advantages to showing learners where multiple structural details apply (“I told
you not to let him get into danger; he’s in danger now, so it’s your job to get him
out of danger”) and where they do not (in haste, in anger, in horror).
In Low (2008) I emphasised a related problem that simply teaching learners
about metaphor does not per se guarantee that the form or meaning will be re-
membered for long, or that productive use of them will improve. Having said
that, much of the research into metaphor teaching has involved recommending
(e.g., Golden 2010) or giving explanations of conceptual metaphor structure and/
or etymology and continues to do so. While pictures and etymology discussions
can add a degree of critical engagement by learners, MacArthur (2017), comment-
ing on Boers et al. (2008), Boers et al. (2009), MacArthur (2006) and Szczepaniak
and Lew (2011), notes that pictures do not seem good at cueing the exact word-
ing, or appropriate use, of new (= to the learner) metaphorical phrases; indeed
there is the danger that what is remembered is the picture not the phrase.
I shall base the last half of this section around two recently published stud-
ies: the first qualitative, the second quantitative, both on teaching vocabulary/
conventional phraseology. In the first study, Niemeier (2017) asked 26 German
pre-intermediate level EFL students aged 13 to 15 years about colour idioms in-
volving seven expressions each for blue, red, green and yellow. Learners were
first asked to collect colour idioms from memory, dictionaries, or a previous
school project on colour symbolism. In lesson 2 they brainstormed in groups
which objects were literally blue etc, which of the 7x4 given idioms were the
same in German and English and which were English only. Then in lesson 3
they reported on the earlier tasks, speculated on the motivations/etymologies
involved and thought up contexts for using the idioms. There was lastly a
homework writing task which had to include four of the taught idioms. Finally,
after three weeks, learners completed a gap-fill post-test. Despite the seeming
absence of a pretest or control group, Niemeier reported that the learners re-
mained motivated and their vocabulary knowledge “improved noticeably”
(ibid.: 280). It is hard from the report to isolate specific factors contributing to
success, but one might speculate, based on research into the value of reflection
and dialogic approaches on learning generally (e.g., Willis (1996) on reflection
and Pontefract and Hardman, or Hardman (2008) on dialogic approaches) that
the group negotiation/discussion (student-student and student-teacher), limited
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 51

input data per colour, making the main task one of actively placing idioms with
and without L1 equivalents on diagrams (as if playing a board game), and the
creation of a coherent narrative production task, may have helped.
Niemeier also raised the vexed question of whether to use pictures, but de-
cided against using them. The problem was not so much recall of exact wording,
but that posed by confusing and inaccurate details – a difficulty earlier raised
briefly by Low (2008, re diagrams of the movement of time) and far more seriously
by Boers and MacArthur (2009) and MacArthur (2010), so if comprehensive pretest-
ing of pictures was not possible, Niemeier’s solution was probably the right one.
The second study, Pan (2019), was again small scale, but involved a more
rigorous research design, with pre-test, control group, and immediate and de-
layed post-tests. The aim was to test out the viability of the “conceptual meta-
phor approach” (CMA) with young learners and to teach 43 elementary-level
EFL learners in China (aged 10 to 12 years) anger expressions relating to ANGER
IS A HOT FLUID IN A CONTAINER. This involved explaining the details of the concep-
tual metaphor to the metaphor/experimental group, consistently teaching lit-
eral senses before metaphoric ones (thus steam before steam up), creating
paired associates (following Boers and Lindstromberg 2009), so that seeing
steam should cue steamed up as well as vice versa. The delayed post-test
showed that the metaphor group and the higher proficiency students (i.e. in the
metaphor group and the control group) did significantly better than the non-
CMA/control group and lower proficiency students, so the conclusion was that
a conceptual metaphor approach to teaching could work successfully with ele-
mentary level students. As in Niemeyer’s study, the metaphor group students
were keen and interested. Unlike in Niemeier’s study, no attempt was made to
establish how Chinese expressions for anger differed from English ones.
One needs to be careful before generalising about the success of CMA teach-
ing. Firstly, age may be a significant factor; CMA may cause problems, or require
different teaching methods, with some young learners (MacArthur 2006;
MacArthur and Piquer-Píriz 2007), who interpret FL expressions via metonymy
rather than metaphor – though Pan’s lower age limit was 10 years. Secondly,
most research studies apart from those by Boers and colleagues have been small
scale, involving small numbers of students, few teaching sessions (Niemeyer
had but three, Pan had just one, and even Boers did not integrate the idiom
teaching into a whole EFL course curriculum), unmotivating control group activ-
ities (frequently emphasising rote-learning of lists [MacArthur 2017: 420]),
small numbers of expressions taught (indeed Pan reported some surprise at
just how little content could be taught per session), and a focus on learning
lexical items rather than discourse-level skills. Pan also reported, and this is
an important point, that the regular teacher who actually taught the class
52 Graham Low

found the CM method exerted a heavy toll on her (something which matches
my anecdotal finding above, that multi-text, multi-task activities are hard and
time-consuming to create). The “effort and tiredness” problem can affect stu-
dents as well as teachers. While Pan’s and Niermeier’s learners appear to
have remained motivated, Condon’s (2008: 249) experimental group students
were reportedly very tired when asked to engage with metaphorical verbs at
the end of the teaching day. The implication is that it would be hard for teach-
ers to integrate CM (or indeed non-CM) metaphor teaching into a regular or
full syllabus/curriculum, unless publishers created extensive course materials
using it, with explanations for teachers.
How far FL metaphor teaching should focus on explicit comparisons with
the first language remains unclear. Nacey and Jensen (2017: 284) noted that
Norwegian EFL learners’ main problem with selecting prepositions was that
they tended to go for the equivalent used in Norwegian. One might generalise
by concluding that where first and target languages are close, like English and
Norwegian or German, teaching across languages is useful, as Niemeier found.
But it is important to bear in mind Philip’s (2010: 75) proviso: “attributing all
errors to language interference is somewhat simplistic,” so the moral would
seem to be to use L1 patterning at times, but not to rely overmuch on it.
Both Niemeier and Pan made sure that basic senses were taught before
metaphoric ones were introduced. While this makes logical sense in most cases
(MacArthur and Piquer-Píriz 2007), including the two situations considered, the
points made in Low (1988) and (2008) remain valid, that the basic sense may
be hard to find/agree on, and that there will be numerous contexts where the
metaphoric sense is either encountered early on (as with classroom manage-
ment language (e.g., write down vs write up, or sit down vs sit up) see also Low
and Littlemore 2009) or before the literal sense (as with technical phrases like
buttress an argument), assuming the literal sense is ever encountered at all; in
these last two situations the “basic/literal sense first” approach will not always
be appropriate (a point echoed by MacArthur 2017).

6 Conclusion
The 1988 paper On teaching metaphor inevitably omitted approaches like blend-
ing theory (for lexis) or Cameron’s use of complexity theory (for discourse), de-
tails of much corpus research, especially that by (e.g.) Philip, Nacey, or Golden
of learner language and strategies, and details of numerous empirical studies
of teaching metaphor as part of foreign language training.
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 53

However, despite the omissions, the “treatment” definition of metaphor seems

usable with (or adaptable to) FL language learners, and the core functions of meta-
phor remain largely valid: indeed, the point that the skills necessary for using an
L2 well involve interactional, context-based, ones, and not just the prior learning
of conventional lexis, remains something which has not been researched, or inte-
grated into teaching programmes, much in the intervening thirty years. While the
notion of generalised “metaphoric competence” can no doubt be refined and re-
vised via the addition of contributory psychological skills, I would argue that for
most FL teaching situations, a “concrete” approach based round what a learner
needs to be able to do with the L2 when hearing/reading it or producing it remains
useful. As does the suggestion that multi-text, multi-task, storyline activities can
aid the learning of interactional and discourse (spoken and written) aspects of L2
metaphor.
In short, On teaching metaphor, together with the 2008 update Metaphor and
education are still valid sources of ideas for teaching and learning, as well as for
future applied research. There is little excuse for the continuing marginality, as
MacArthur (2017) put it, of metaphor to EFL teaching or FL teaching in general.

References
Alexander, Richard. 1983. Metaphors, connotations, allusions: Thoughts on the language-
culture connexion in learning English as a foreign language. Trier: University of Trier,
L. A. U. T. Papers, Series B, No. 91.
Azuma, Masumi. 2003. Metaphorical competence in an EFL context. Paper presented at the
5th International Conference of Researching and Applying Metaphor (RAAM V), Paris
September 3–5. Program and Abstracts 54.
Azuma, Masumi. 2005. Metaphorical competence in an EFL context. Tokyo: Toshindo
Publishing.
Bachman, Lyle. 1990. Fundamental considerations in language testing. Oxford: Oxford
University Press.
Bachman, Lyle F. & Adrian S. Palmer. 1996. Language testing in practice. Oxford: Oxford
University Press.
Boers, Frank, Seth Lindstromberg, Jeannette Littlemore, Hélène Stengers & June Eyckmans.
2008. Variables in the mnemonic effectiveness of pictorial elucidation. In Frank Boers &
Seth Lindstromberg (eds.), Cognitive linguistic approaches to teaching vocabulary and
phraseology, 189–216. Berlin: Mouton de Gruyter.
Boers, Frank, Ana Piquer-Píriz, Hélène Stengers & June Eyckmans. 2009. Does pictorial
elucidation foster recollection of figurative idioms? Language Teaching Research 13(4).
367–388.
Boers, Frank & Seth Lindstromberg. 2009. Optimising a lexical approach to instructed second
language acquisition. Basingstoke: Palgrave Macmillan.
54 Graham Low

Boers, Frank & Fiona MacArthur. 2009. The use of pictorial and gestural modes in fostering
the learning of idioms and conventional metaphors in instructed second language
acquisition: An overview. Paper presented at RaAM 2009 Workshop ‘Metaphor,
metonymy and multimodality, University of Amsterdam, June 4 –5.
Cameron, Lynne. 2003. Metaphor in educational discourse. London: Continuum.
Cameron, Lynne, Graham Low & Robert Maslen. 2010. Finding systematicity in metaphor use. In
Lynne Cameron & Robert Maslen (eds.), Metaphor analysis: Research practice in applied
linguistics, social sciences and the humanities, 116–146. London: Equinox.
Deignan, Alice, Jeannette Littlemore & Elena Semino. 2013. Figurative language, genre and
register. Cambridge: Cambridge University Press.
Drew, Paul & Elizabeth Holt. 1995. Idiomatic expressions and their role in the organisation of
topic transition in conversation. In Martin Everaert, Erik-Jan van der Linden, André Schenk &
Rob Schreuder (eds.), Idioms: Structural and psychological perspectives, 117–131. Hillsdale,
NJ: L. Erlbaum.
Fauconnier, Gilles & Mark Turner. 2008. Rethinking metaphor. In Raymond Gibbs, Jr (ed.), The
Cambridge handbook of metaphor and thought, 53–66. Cambridge, UK & New York:
Cambridge University Press.
Fisher, Linda. 2013. Constructing beliefs in the foreign language classroom using metaphor as
a sociocultural tool. Cambridge: University of Cambridge, PhD dissertation.
Golden, Anne. 2010. Grasping the point: A study of 15-year-old students’ comprehension of
metaphorical expressions in schoolbooks. In Graham Low, Zazie Todd, Alice Deignan &
Lynne Cameron (eds.), Researching and applying metaphor in the real world, 35–62.
Amsterdam: John Benjamins.
Grady, Joseph. 1998. The Conduit Metaphor revisited: A reassessment of metaphors for
communication. In Jean-Pierre Koenig (ed.), Conceptual structure, discourse and
language 2, 205–218. Stanford, CA: CSLI Publications.
Grady, Joseph & Christopher Johnson. 2002. Converging evidence for the notions of subscene
and primary scene. In René Dirven & Ralf Pörings (eds.) Metaphor and metonymy in
comparison and contrast, 533–554. Berlin: Mouton de Gruyter.
Halliday, Michael. 1985/1994. An introduction to functional grammar (2nd ed.). London:
Edward Arnold.
Hardman, Frank. 2008. Promoting human capital: The importance of dialogic teaching in
higher education. Asian Journal of University Education 3(1). 31–48.
Lakoff, George. 1982. Categories and cognitive models. Trier: University of Trier,
L. A. U. T. Papers, Series A, No. 96.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago, IL: Chicago University
Press.
Lakoff, George & Zoltán Kövecses. 1983. The cognitive model of anger inherent in American
English. (Mimeo). Berkeley, CA: University of California at Berkeley, Cognitive Science
Program, Institute of Human Learning.
Lakoff, George & Mark Turner. 1989. More than cool reason: A field guide to poetic metaphor.
Chicago, IL: Chicago University Press.
Lerman, Claire. 1983. The functions of metaphor in discourse: Masking metaphor in the Nixon
conversations. Paper presented at the 1983 Whim Conference “Metaphors be with you:
Humour and metaphor.” (Mimeo.) English Department, Arizona State University. (Short
version published in Don L. F. Nilsen (ed.), Whimsy 2. 133–148.)
 Taking Stock after Three Decades: “On Teaching Metaphor” Revisited 55

Lerman, Claire. 1985. Media analysis of a presidential speech: Impersonal identity forms in
discourse. In Teun van Dijk (ed.), Discourse and communication: New approaches to the
analysis of mass media discourse and communication, 185–215. New York: Walter de
Gruyter.
Littlemore, Jeannette. 2001. Metaphoric competence: A language learning strength of
students with a holistic cognitive style? TESOL Quarterly 35(3). 459–491.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sarah Turner. 2014. Investigating figurative
proficiency at different levels of second language writing. Applied Linguistics 35(2). 117–144.
Littlemore, Jeannette & Graham Low. 2006a. Metaphoric competence and communicative
language ability. Applied Linguistics 27(2). 268–294.
Littlemore, Jeannette & Graham Low. 2006b. Figurative thinking and foreign language
learning. Basingstoke: Palgrave Macmillan.
Low, Graham. 1988. On teaching metaphor. Applied Linguistics 9(2). 125–146.
Low, Graham. 1997. Figurative boundary frames in written text. (Mimeo).
Low, Graham. 2005. Explaining evolution: The use of animacy in an example of semi-formal
science writing. Language and Literature 14(2). 129–148.
Low, Graham. 2008. Metaphor and education. In Raymond W. Gibbs, Jr (ed.), The Cambridge
handbook of metaphor and thought, 212–231. Cambridge, UK & New York: Cambridge
University Press.
Low, Graham. 2010. Wot no similes? The curious absence of simile in university lectures. In
Graham Low, Zazie Todd, Alice Deignan & Lynne Cameron (eds.), Researching and
applying metaphor in the real world, 291–308. Amsterdam: John Benjamins.
Low, Graham. 2015. A practical validation model for researching elicited metaphor. In Wan
Wan & Graham Low (eds.), Elicited metaphor analysis in educational discourse, 15–37.
Amsterdam: John Benjamins.
Low, Graham. 2017. Eliciting metaphor in education research: Is it really worth the effort? In
Francesca Ervas, Elisabetta Gola & Maria-Grazia Rossi (eds.), Metaphor in
communication, science and education, 249–266. Berlin: de Gruyter.
Low, Graham & Jeannette Littlemore. 2009. The relationship between conceptual metaphors and
classroom management language: Reactions by native and non-native speakers of English.
Iberica, 17 (Special issue on metaphor and language for specific purposes). 25–44.
MacArthur, Fiona. 2006. The effects of pictorial representations on the learning of imageable
idioms in EFL. Paper presented at the 6th International Conference on Researching and
Applying Metaphor (RaAM6), Leeds, April.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where are we and where should we
be going? A view from the language classroom. AILA Review 23. 155–173.
MacArthur, Fiona. 2017. Using metaphor in the teaching of second/foreign languages. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
413–425. Abingdon & New York: Routledge.
MacArthur, Fiona & Ana M. Piquer-Píriz. 2007. Staging the introduction of figurative
extensions of familiar vocabulary items in EFL: Some preliminary considerations. Ilha do
Desterro 53. 123–136.
Nacey, Susan & Bård U. Jensen. 2017. Metaphoricity in English L2 learners’ prepositions. In
Francesca Ervas, Elisabetta Gola & Maria-Grazia Rossi (eds.), Metaphor in
communication, science and education, 283–304. Berlin: de Gruyter.
56 Graham Low

Niemeyer, Susanne. 2017. Teaching (in) metaphors. In Francesca Ervas, Elisabetta Gola &
Maria-Grazia Rossi (eds.), Metaphor in communication, science and education, 267–282.
Berlin: de Gruyter.
O’Reilly, David. 2017. An investigation into metaphoric competence in the L2: A linguistic
approach. University of York, PhD dissertation.
Pan, Molly Xie. 2019. The effectiveness of the Conceptual Metaphor Approach to English idiom
acquisition by young Chinese learners. Metaphor & the Social World 9(1). 59–82
Philip, Gill. 2010. “Drugs, traffic and many other dirty interests”: Metaphor and the language
learner. In Graham Low, Zazie Todd, Alice Deignan & Lynne Cameron (eds.), Researching
and applying metaphor in the real world, 63–80. Amsterdam: John Benjamins.
Pontefract, Caroline & Frank Hardman. 2005. The discourse of classroom interaction in Kenyan
primary schools. Comparative Education 42(1). 87–106.
Pragglejaz group. 2007. MIP: A method for identifying metaphorically used words in
discourse. Metaphor and Symbol 22(1). 1–40.
Sharifian, Farzad. 2017. Cultural linguistics. Amsterdam: John Benjamins.
Skehan, Peter. 1998. A Cognitive approach to language learning. Oxford: Oxford University
Press.
Steen, Gerard. 1999. From linguistic to conceptual metaphor in five steps. In Raymond
W. Gibbs & Gerard Steen (eds.), Metaphor in cognitive linguistics, 55–77. Amsterdam:
John Benjamins.
Steen, Gerard. 2007. Finding metaphor in grammar and usage. Amsterdam: John Benjamins.
Steen, Gerard. 2011a. When is metaphor deliberate? In Nils-Lennart Johannesson, David
C. Minugh & Christina Alm-Arvius, (eds.), Selected papers from the 2008 Metaphor
Festival, Stockholm, 43–63. Stockholm: Acta Universitatis Stockholmiensis.
Steen, Gerard. 2011b. What does ‘really deliberate’ really mean? More thoughts on metaphor
and consciousness. Metaphor and the Social World 1(1). 53–56.
Sullivan, Karen & Eve Sweetser. 2010. Is “generic is specific” a metaphor? In Fey Parril, Vera
Tobin & Mark Turner (eds.), Meaning, form and body, 309–328. Stanford, CA: CSLI
Publications.
Szczepaniak, Renata & Robert Lew. 2011. The role of imagery in dictionaries of idioms. Applied
Linguistics 32(3). 323–347.
Veale, Tony 2012. A computational exploration of creative similes. In F. MacArthur, José Luis
Oncins-Martínez, Manuel Sánchez-García & Ana María Piquer-Píriz (eds.), Metaphor in
use: Context, culture, and communication, 329–343. Amsterdam: John Benjamins.
Wan, Wan. 2011. An examination of the validity of metaphor analysis studies: Problems with
metaphor elicitation techniques. Metaphor and the Social World 1(2). 261–288.
Wan, Wan. 2012. Using metaphorical conceptualisation to construct and develop ESL students’
writing: An exploratory study. University of York, PhD dissertation.
Wan, Wan & Graham Low. (Eds.). 2015. Elicited metaphor analysis in educational discourse.
Amsterdam: John Benjamins.
Willis, Jane. 1996. A framework for task-based learning. Harlow: Longman.
Yu, Ning. 2007. Heart and cognition in ancient Chinese philosophy. Journal of Cognition and
Culture 7(1). 27–47.
Ana M. Piquer-Píriz
Figurative Language and Young L2
Learners
1 Introduction
For a long time figurative language was considered one the highest achievements
of cognitive development and not available to young children (Piaget and
Inhelder 1969). Nowadays, however, there seems to be a general consensus
among linguists, psychologists and psycholinguists (Cameron 1996, 2003;
Johnson 1999; Özçalışkan 2011, 2014; Nerlich, Clarke and Todd 1999; Stites and
Özçalışkan 2013; Vosniadou 1987; Winner 1988; Zurer Pearson 1990) that figura-
tive language is part of our communicative system from early childhood (for
some recent reviews, see Özçalışkan 2011, 2014). It is common to listen to chil-
dren using, for instance, metaphors (e.g. it’s a flying unicorn, referring to a mov-
ing, horse-shaped cloud) or personifications (it’s having a wee, referring to a
plant that has just been watered) from a very early age.
It has also been shown that children are able to cope with figurative meanings
in a second language (L2). When confronted with non-literal, semantic extensions
of words that are commonly employed in a literal sense in the L2 classroom (e.g.
give me a hand), young L2 learners were able not only to understand these mean-
ings but also to explain verbally the connections between the literal and figurative
senses (Piquer-Píriz 2005, 2008a, 2008b and 2010), showing that they are also
ready to benefit from a CL-oriented approach to L2 vocabulary learning in the way
that has been advocated for older learners (cf. Boers and Lindstromberg 2008).
This chapter aims to explore how children’s figurative competence may fa-
cilitate their systematic building of connections among the semantic extensions
of core meanings in an L2. In order to do so, section 2 presents a brief definition
of the theoretical notions of figurative competence and linguistic motivation fo-
cusing on how they are interrelated and may have an impact on the acquisition
of vocabulary in an L2. The rest of the chapter develops these two notions.
Section 3 focuses on the former and is divided into two parts: The first part of-
fers a brief overview of the existing research into children’s figurative thinking
in their L1 and the second part reports on my own research into children’s un-
derstanding of L2 meanings.

Ana M. Piquer-Píriz, English Philology (Faculty of Education), University of Extremadura,

Badajoz, Spain, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-004
58 Ana M. Piquer-Píriz

Section 4 is devoted to the notions of linguistic motivation and motivated

polysemy and analyses their potential in young learners’ development of vo-
cabulary in an L2.

2 Defining Figurative Competence and Linguistic

Motivation
Researchers from different areas have taken great interest in exploring what has
been termed ‘children’s figurative competence’ (for a complete definition, see
Levorato and Cacciari 1992), which, in this chapter, is broadly understood as child-
ren’s ability to understand and produce non-literal meanings of words and
phrases. It should be noted that the term ‘figurative’ is used here to denote all
kinds of non-literal expressions since, as will be seen in the next section, it has
been shown that children’s ability to deal with non-literal forms is not restricted to
metaphor but it also covers other forms such as metonymy, personification or
irony that are also present in children’s language and thought.
The concept of ‘figurative competence’ can be equated to the more widely
used notion of ‘metaphorical competence’ (Gardner and Winner 1978, Danesi 1986,
1992, 2008, 2016; Littlemore 2001, Littlemore and Low 2006a; Low 1988). The emer-
gence of this idea can be traced back to the 1970s in the work carried out by psy-
chologists such as Gardner and Winner (1978) and has, since then, been analysed
in the development of the L1 and other languages (see Nacey 2017 for an overview
of the literature on metaphorical competence; Littlemore and Low 2006a, 2006b
and MacArthur 2010 for insightful discussions of the importance of this compe-
tence in L2 learning; or Littlemore 2010 for a study that explores the relationship
between metaphor interpretation and production behaviour in L1 and L2). As can
be seen, this construct and, particularly, the role that it plays in L2 acquisition has
attracted a great deal of interest among researchers and, in fact, several chapters
included in the present volume explore this issue from different perspectives (see
the chapters by Low; Suárez Campos, Hijazo-Gascón and Ibarretxe-Antuñano;
Nacey; or Castellano-Risco and Piquer-Píriz).
At the same time, cognitive linguists have been emphasising the importance
of linguistic ‘motivation’. For them, language use is a reflection of general cogni-
tive processes, grounded in experiential correlations and mediated by culture and,
instead of treating linguistic manifestations as arbitrary, they look for plausible ex-
planations for their characteristics, highlighting how the different meanings of a
word are motivated and interrelated. This gave rise to a new conception of poly-
semy which, according to Lakoff (1987: 13), “arises from the fact that there are
 Figurative Language and Young L2 Learners 59

systematic relationships between different cognitive models and between elements

of the same model. The same word is often used for elements that stand in such
cognitive relations to one another”. This notion of motivated polysemy has in-
spired applied cognitive linguists to try out new methods to present new L2 vo-
cabulary to learners as well as to explore how effective these CL-oriented
practices may be on the basis that “learners who are aware that an L2 is much
more than a system of arbitrary form-meaning connections may be relatively
likely to adopt mnemonically fruitful practices of insightful learning rather
than less effective ones associated with blind memorisation” (Boers and
Lindstromberg 2008: 18).
In my view, the two notions of figurative competence and motivated polysemy
are intrinsically connected and this can have very important implications in the
process of learning an L2 even at very early ages. Most definitions of ‘metaphorical
competence’ generally describe it as the ability to understand and produce meta-
phor (Danesi 1986, 1992, 2008, 2016; Littlemore 2001; Low 1988; Nacey 2013) and,
in the specific case of the role it may play in the process of learning an L2, some
authors relate it to a more global capacity that has been labelled ‘conceptual flu-
ency’ (Danesi 1992) or ‘figurative reasoning’ (Littlemore and Low 2006a). And this
is precisely where the connection lies. The ability to use our concrete knowledge to
understand abstract concepts on the basis of our embodied and cultural experien-
ces and our interaction with the world corresponds to the basis of the Cognitive
Linguistic paradigm (Lakoff and Johnson 1980, 1999). It is quite simple to put this
tenet to the test informally. When I have tried to explain this idea to my university
students, colleagues and even friends I have tended to use a very simple example.
Depending on whether they are Spanish, English or Portuguese L1 speakers, I ask
them why they think that we use cabecero-a / head /cabeceira to refer to el/la ca-
becero/a de la cama / the head of the bed / a cabeceira da cama. There are always
two main answers that seem to reflect, on the one hand, associative (metonymi-
cally-based) reasoning, e.g. because that is the part where we lay our head on; and
analogical (metaphorically-based) processing, e.g. because it is the top of the bed.
This little test is certainly very unscientific but quite revealing: there seems to be a
natural, spontaneous ability that allows us to connect concrete and abstract mean-
ings in the way that cognitive linguists have advocated and this capacity has been
shown to be part of the human reasoning abilities from early infancy (Özçalışkan
2011, 2014) and also available to children with L2 figurative uses (Piquer-Píriz
2005, 2008a, 2008b, 2010).
Thus, this natural capacity to understand abstract concepts by relating them
in different ways to more concrete ones would be conceptual (see Gibbs’ [1999] for
a discussion on the difference between the mental process and the linguistic prod-
ucts) and, therefore, not restricted to the way we reason in our native language but
60 Ana M. Piquer-Píriz

would also be applicable to any additional languages we may speak. Some more
rigorous explorations have been carried out with L2 learners of different ages:
young adults (Piquer-Píriz and Boers 2010), teenagers (Piquer-Píriz, MacArthur
and Alejo-González 2010) and children (Piquer-Píriz 2005, 2008a, 2008b, 2010). As
will be seen in further detail below, it has been shown that even very young L2
learners (as young as five years of age) establish metaphorical and metonymical
connections when asked to explain non-literal L2 meanings. When asked, for in-
stance, to identify and explain the idiomatic expression give me a hand, most of
the learners, particularly at the younger ages, that took part in Piquer-Píriz’s study
(2005: 123) employed figurative reasoning (e.g. (1) “‘hand’ porque con las manos se
puede ayudar” – ‘hand’ because you can help with your hands) in preference to
other strategies such as interlingual identification (e.g. (2) “ ‘hand’ porque en
español es parecido a ‘échame una mano’ ” – ‘hand’ because it’s similar to ‘échame
una mano’ in Spanish [2005: 136]). If these young learners are able to establish
these connections, it would seem reasonable to think that they would also benefit
from an instructional approach that relies on fostering metaphorical competence
or, more generally, enhances figurative thought.

3 Figuration and Children

3.1 Children’s Figurative Thinking in Their L1

Research, from different perspectives, into the understanding and production of

figurative language at the early stages of the development of an L1 (e.g. Cameron
1996, 2003; Johnson 1999; Özçalışkan 2011; Pineda 2015; Stites and Özçalışkan
2013; Vosniadou 1987; Winner 1988; Zurer Pearson 1990) has mostly concluded
that children start to produce and understand these linguistic uses shortly after
they begin to speak and that their figurative abilities improve with age
(Özçalışkan 2011), thus, overcoming the contention that figurative language was
one of the highest achievements of cognitive development and not available to
an ‘immature’ mind (Piaget and Inhelder 1969). Children’s ability with figurative
forms is not restricted to metaphor (Billow 1975; Nippold, Leonard and Kail 1984;
Nippold and Sullivan 1987; Vosniadou 1987; Zurer Pearson 1990; Cameron 1996,
2003) and idioms (Gibbs 1987, 1991; Cacciari and Levorato 1989, 1998; Levorato
and Cacciari 1992; Abkarian, Jones and West 1992). Metonymy has also been ac-
knowledged to be present in children’s language and thought (Nerlich, Clarke
and Todd 1999) and children’s understanding of irony has also been studied
(Winner 1988).
 Figurative Language and Young L2 Learners 61

A number of important issues that have emerged from this literature are
briefly discussed below.

3.1.1 Developmental psychology: Awareness of a pre-existing similarity

between two elements (perceptual vs. psychological links)

From the mid-seventies onwards, developmental psychologists have been rigor-

ously analysing children’s figurative capacity and have shown that most of the
early ideas about children’s inability to understand figurative productions until
later childhood were partly a consequence of applying adults’ standards to mea-
sure and interpret children’s capacities (Winner 1988). The results were very differ-
ent in studies that employed methodological techniques adapted to children such
as elicited, verbal repetition (e.g., Zurer Pearson 1990) or the use of stories, some-
times enacted by puppets or toys (e.g., Levorato and Cacciari 1992 or Stites and
Özçalışkan 2013).
Establishing criteria that allowed researchers to classify children’s utterances
as metaphorical and non-metaphorical became central in the field (Vosniadou
1987, Vosniadou and Ortony 1983, Winner 1988). Most of these accounts of the de-
velopment of children’s figurative capacity were based on the idea that there is a
pre-existing similarity between two elements and that children should be aware
of it in order to classify their productions as metaphorical. The type of relationship
existing between the concrete and abstract elements was also considered a key
issue. A pioneering study in this regard is Asch and Nerlove’s (1960) analysis of
children’s (aged from three to twelve) understanding of adjectives that refer to
both physical properties and psychological traits such as cold, warm, deep, hard or
bright. They found that the physical sense of the term is acquired first, the psycho-
logical meaning only comes later and as an independent and, the dual property of
the lexeme is realised at last and, in most cases, not spontaneously.
In the eighties, Winner (1988) distinguished between sensory and non-
sensory metaphors and claimed that: “What children most often do is find a sen-
sory similarity between topic and vehicle when a nonsensory, relational similarity
is at issue” (Winner 1988: 36). She defined non-sensory metaphors as those based
on similarity that is not apprehended by our senses. They can be divided into rela-
tional and psychological-physical metaphors. The former are based on similarities
between objects, situations, or events that are physically dissimilar, but, owing to
parallel internal structures, function in a similar way (e.g. clouds and sponges
which function to hold and then release water). Psychological-physical metaphors
are based on a resemblance between the sensory attributes of a physical object,
62 Ana M. Piquer-Píriz

perceived through any sensory modality, and psychological, non-sensory attrib-

utes of a person (e.g. cheerful people are ‘sunny’, or cranky people are ‘sour’).
More recently, Özçalışkan (2011, 2014) has also highlighted that, from a very
early age, children are able to understand and produce comparisons based on per-
ceptual similarities (e.g. ‘The moon is like a balloon’) but that they experience
more difficulties with comparisons that involve two different domains of experi-
ence (e.g. ‘a warm person’). A very similar pattern has been observed when chil-
dren are reasoning in an L2. As will be seen in greater detail in the next section, in
a study that explored young Spanish EFL learners’ understanding of the figurative
meanings of warm and cold when referring to personality (Piquer-Píriz 2010), the
younger children (6- and 8-year-olds) based their answers on the concrete, physi-
cal sensations (a person is cold or warm depending on the weather or the clothes
they are wearing). In contrast, some of the 10-year-olds could establish the psycho-
logical link between the two meanings: coldness was clearly related to unfriendli-
ness and warmth with kindness.
The evidence offered by this body of research suggests that children’s percep-
tion of properties changes with development, with perceptual or sensory links ap-
pearing first, followed by relational or functional links, and finally physical-
psychological links. There are two possible explanations for these findings, in rela-
tion to children’s metaphorical competence. On the one hand, it could be argued
that children are not able to perceive the kinds of similarities on which non-
sensory metaphors are based and, therefore, their metaphorical competence
would not be fully developed until a certain age. On the other, the fact that chil-
dren are not able to perceive certain types of similarities may not be due to their
underdeveloped metaphorical competence but, rather, to their insufficient knowl-
edge of the domains to which the elements that are compared belong, thus limit-
ing the kinds of connections that children perceive. When the child has
sufficiently developed knowledge of the elements being linked, they are able to in-
terpret the metaphor even if the ground is non-sensory (Keil 1989). As is shown in
the next section, there is some evidence that support this second explanation.

3.1.2 The importance of children’s developing knowledge of the world

It seems that children’s developing knowledge of the world plays an important

role in their understanding and production of figurative language. Several scholars
(Carey 1985, Keil 1989, Winner 1988, Vosniadou 1989 or Cameron 1996) have ar-
gued that children’s ‘misunderstanding’ of metaphors is not due to their lack of
metaphorical capacity but to their developing ‘knowledge of the world’, particu-
larly, a lack of ‘domain knowledge’. These two notions refer to how children’s
 Figurative Language and Young L2 Learners 63

knowledge of concepts and events and their relations among them become en-
riched in response to different inputs and experiences (their own social experien-
ces with other people and the world that surrounds them, including the
explanations provided by their parents, teachers, siblings or friends but also sto-
ries, cartoons, TV programmes or the Internet, that is, all sort of inputs to which
they are exposed). Children are in the process of developing this knowledge and,
therefore, although even very young children may have the competence to per-
ceive all of the kinds of similarities that adults perceive, they may lack fully devel-
oped knowledge of the domains from which either the topic or vehicle is drawn,
and thus fail to see the similarity between them. One cannot know which aspects
of the vehicle domain to map onto the topic domain if little is known about one or
both domains. In order to test this hypothesis, Keil (1989) used metaphors based
on domains that are familiar to children and showed that when children had suffi-
cient knowledge of the two domains involved in the comparison, they demon-
strated no difficulty in understanding non-sensory metaphors. Cameron (1996)
argues that discourse context (which, according to her would include situational
context (participants, situation and goals), the immediate linguistic context of the
metaphor, and the textual or interactional context of the metaphor) “can contrib-
ute towards accessing and selecting domain knowledge” (Cameron 1996: 54).

3.1.3 Figurative productions as developmental strategies

For other authors, some of the figurative productions uttered by children are prod-
ucts of their language development. In this sense, Clark (1993) described overex-
tensions (e.g. when a child calls a horse ‘doggy’) as one of the most important
ways that children use to map the meaning that they want to express onto a form
which they have already acquired. In a similar vein, Nerlich, Clarke and Todd
(1999) see overextensions as communicative strategies used by children which can
be based on metaphorical, metonymical and synecdochical relationships. They fo-
cused on the analysis of metonymy and found that up to age 2:5, children make
use of what they call ‘compelled metonymical overextensions’. They are compelled
because they are based on the fact that at this age children’s vocabulary, category
and conceptual systems are still relatively small and unstructured and as soon as
the right word is learned, the use of overextensions stops. These compelled meton-
ymies, which are used by children to fill gaps in their limited lexicons contrast
with creative metonymies used to express something new by not using words al-
ready available in their lexicons. At these stages, children’s semantic systems, as
well as their systems of concepts and categories, are still incomplete and unstruc-
tured and will have to be adjusted and modified in the years to come.
64 Ana M. Piquer-Píriz

Overextension would be the mechanism used to fill some gaps. As soon as the
right word is learnt, it stops. This view raises the controversial issue of what we
might call ‘unacceptable’ vs. ‘acceptable’ extensions from an adult’s point of view.
Children’s early overextensions are ‘unacceptable’ from an adult point of view, but
as soon as children’s categories match the adults’, they are not regarded as over-
extensions anymore.

3.1.4 Cognitive Linguistics: Johnson’s ‘constructional grounding’

Johnson’s (1999) constructional grounding theory is, in my view, one of the most
comprehensive accounts, from a cognitive-linguistic perspective, of the acquisi-
tion of figurative language by children which, despite having been published 20
years ago, has unfortunately not been much further developed. Johnson argued
that children acquire metaphorical concepts by early conflations of two domains
of experience by means of what he calls ‘constructional grounding’. This idea is
closely related to Grady’s (1997) Primary Metaphor Theory, according to which
there are some conceptual metaphors – the most basic ones – which arise directly
from experiential correlations that are often highly embodied and that happen in
our early experience of the world. These metaphors would be unconsciously and
automatically incorporated to our understanding of concepts. Thus, the origin of
this type of metaphors is not some type of perceived similarity but the notion of
co-occurrence (conflation of experiences). One widely quoted example is QUANTITY
IS VERTICAL ELEVATION (also expressed as MORE IS UP) which is motivated by “the
correlation between quantity and levels in piles, fluids in containers, etc.” (Grady
1997: 285).
Johnson’s constructional grounding theory is based on constructional and cog-
nitive approaches to language, mainly Tomasello’s (2003, 2006) ideas on construc-
tion grammar from the point of view of L1 acquisition and Slobin’s (1981) notion of
the importance of simple scenes experienced by children and the role they play in
the process of acquiring their mother tongue. According to Johnson, “a sign that is
relatively easy for children to learn serves as the model for another more difficult
sign, because it occurs in contexts in which it exemplifies important properties of
the more difficult sign in a way that is especially accessible to children” (1999: 1).
Two important processes are necessary for constructional grounding to occur: In
the first place, there must be some kind of relation between the two signs (co-
occurrence or similarity) and they must be frequent utterances in the language
input to which children are exposed. All in all, these underlying processes are not
very different from those pointed out by developmental psychologists.
 Figurative Language and Young L2 Learners 65

3.1.5 Concluding remarks

As can be seen, the development of figurative language in children is a fascinat-

ing topic that has attracted the attention of researchers from different fields
(mostly psychologists and linguists). Research into it has produced mutually en-
riching findings that, when considered together, offer a more global (and yet in
need of further exploration) picture of this complex phenomenon. From the brief
overview of some of the most influential research into figuration and children in
their L1 presented above, we can infer that, when analysing children’s under-
standing and production of figurative language, it is important to take into ac-
count processing, production and contextual factors. Levorato and Cacciari
(1992: 146–147) attempted to define what children’s figurative competence may
consist of, including the following four abilities that take into account processing
(1, 2 and 4), production (4) and contextual (3) factors:
1. The ability to understand the dominant, peripheral and additional related
meaning of a word, its position in a given semantic domain and its paradig-
matic and syntagmatic relations.
2. The ability to suspend a purely literal-referential strategy. This is a prereq-
uisite not only for figurative language comprehension, but also for most of
the linguistic repertoire (for instance, polysemous words).
3. The ability to use contextual information in order to construct a coherent
semantic representation and to integrate it with the lexical and semantic
information carried by the figurative expression.
4. The ability to create and understand the figurative uses of a word, a sen-
tence or a given domain as well as to retrieve the conceptual structures
involved.

As a conclusion, figurative reasoning and its linguistic products seem to be a

continuum in a person’s life. In the early stages, some creative figurative uses
may be related to the need to fill gaps in the children’s growing conceptual and
linguistic system, as Nerlich, Clarke and Todd (1999) and Clark (1993) have
pointed out. These productions, strange by adults’ standards, seem to be com-
mon to children, who might simply be using a mechanism (seeing something in
terms of something else) which appears to be available to them from a very
early age. This mechanism will be used throughout their lives as a fundamental
tool for recognition, classification, learning, and even scientific discovery and
creativity (Vosniadou and Ortony 1989, Gentner 2003) and, consequently, when
enriching vocabulary in both their mother tongue and a second language. At
the same time, not only the ability to perceive something in terms of something
else but also the co-occurrence of highly embodied, experiential correlations of
66 Ana M. Piquer-Píriz

concrete and abstract notions accompanied by the exposure to the language that
expresses those concepts may explain why certain figurative uses are acquired the
way they are, as Johnson (1999) explains. Furthermore, children are being ex-
posed to conventional figurative forms through their interaction with adults and
peers, education, games, stories, television or the internet. These uses establish
the boundaries between accepted, customary, figurative uses of a specific cultural
community and creative, sometimes one-shot uses that an individual may employ
and that, if congruent with metaphorical usage generally, are very likely to be un-
derstood by other language users. Thus, a fundamental characteristic of human
figurative competence seems to be continuity: It is present from very early ages
and it develops into adult competence as the fundamental ability of analogical
reasoning that, in some individuals, might reach its heights, as Gibbs (1994) ar-
gues, in the form of scientific discovery or literary production.

3.2 Figurative Language and Young L2 Learners

The role of figurative language and thought at the earliest stages of the second
language learning process has been largely unexplored. My own research has
focused on children’s figurative ability when learning an L2. As stated above, if
children are able to interpret and produce non-literal meanings in their L1, they
would be expected to behave in a similar way in an L2. In order to analyse this
issue, I explored how children understood and reasoned about some figurative
meanings in two domains of experience that are very commonly used in the
young L2 leaners’ classroom: body part terms (Piquer-Píriz 2005, 2008a, 2008b)
and temperature terms when applied to colours (Piquer-Píriz 2006) and person-
ality (Piquer-Píriz 2010). A total of 148 children took part in the study that fo-
cused on body part terms and 140 participants were involved in the studies that
explored temperature terms. Their mean ages ranged from 5:8 to 11:9. They all
attended state schools in Extremadura (western Spain) and came from a variety
of social and economic backgrounds.
In the case of body part terms, eight semantic extensions of three different
body part terms were analysed (Piquer-Píriz 2005): HAND (give me a hand, the
hand of a watch and hand it to me), MOUTH (not open one’s mouth) and HEAD
(the head of a bed, the head of a hammer, the head of the stairs and the head of
a line of cars). The figurative meanings of warm and cold when applied to col-
ours (red, orange, yellow, light blue, dark blue and green) and personality
(warm and cold people) were also explored.
The main findings were that, first of all, the young (5- to 12-year-old)
Spanish learners of EFL that took part were, in general, able to identify different
 Figurative Language and Young L2 Learners 67

semantic extensions of English body part and temperature terms whose proto-
typical meaning they knew from their English lessons. However, not all the ex-
tensions were equally transparent for them. Some of the patterns identified in
the literature into children’s figurative abilities in their L1 reviewed in the previ-
ous section were also found in these studies. For example, those comparisons
based on perceptual similarities (such as the head of a hammer) were more eas-
ily identified by learners in all age groups than comparisons that involved two
different domains of experience (cold/warm people). In the case of the latter,
younger children (6- to 8-year-olds) related their interpretations of these figura-
tive meanings to bodily sensations, that is, these children concentrated mainly
on the prototypical meaning of warm and cold as temperature terms and partic-
ularly in the bodily sensation of being cold or warm. In the context of the stim-
uli with which they were provided (i.e., a short story), this was explained in
terms of the clothes worn by the dolls used to enact the story:

(3) C: “porque tiene más frío – porque lleva menos ropa [se refiere a ‘the fat
lady’]‘warm’ (R: ésta [thin lady] es ‘warm’¿por qué crees?) tiene más calor
porque lleva más cosas”
C: Because she feels colder – because she’s got less pieces of clothing on
[she refers to the fat lady] warm (R: this one [thin lady] is warm why do
you think so?) she feels warmer because she’s got more things on.

The youngest children that took part in this specific study (6-year-olds) mainly
stayed in the concrete realm and related cold and warm with their own bodily
sensations. This does not mean that they lacked figurative competence but it
seems that they had not made the links between temperature terms and person-
ality yet. The 8-year-olds appeared to be on their way to doing so but the pro-
cess did not seem to be complete yet. In contrast, most of the 10-year-olds
related both domains but, in the process of developing their world knowledge,
some notions were not clearly conventionalised yet and some of them made
their own interpretations (see Piquer-Píriz 2010, for further details).
The figurative uses analysed not only included perceptually-based compari-
sons but also metonymically-based semantic extensions. In fact, the expressions
give me a hand and hand it to me may well be motivated by experiential correla-
tions, in the sense that Johnson (1999) points out. Interestingly, the results con-
cerning hand it to me present the most uniform data across the different age
groups, not only in the number of correct responses (over 80%), but also in the
interpretation strategy employed in most cases, i.e., the matching of hand with
its function of picking up and giving out things, grounded in the children’s expe-
riences of manipulating and passing things, in what seems to reflect the
68 Ana M. Piquer-Píriz

importance of Slobin’s (1981) notion of simple scenes: children’s early experien-

ces manipulating objects. The clearly embodied nature of hand it to me seems to
have facilitated its comprehension by a very high percentage of children in the
three age groups, despite the fact that this particular use involves a shift in part
of speech (from noun to verb) which does not happen in Spanish where the pos-
sible expressions would be pásamelo/dámelo (con la mano) which literally mean
pass it on to me / give it to me (with the hand). In fact, ‘mano’ (hand) is not usu-
ally even part of the expression in Spanish. These children focused on the mean-
ing of the figurative extension and not on the linguistic form and successfully
interpreted it.
Thus, an important finding from these studies is that children’s ability to rea-
son figuratively played a very important role in their interpretation of non-literal
uses. In this sense, these results also replicate some of the findings in previous
research into the figurative capacity of monolingual children: Analogical reason-
ing and the transfer of knowledge from a concrete to an abstract domain play a
significant role in children’s abstract reasoning. Furthermore, they also indicate
that this competence operates with linguistic forms in the second language from,
at least, the age of five. These children’s figurative reasoning was often based on
their bodily experiences and their observation of and interaction with the world
that surrounds them. The influence of embodiment was especially evident in the
study devoted to body parts terms (Piquer-Píriz 2005). For example, many chil-
dren established metonymical connections of the type HAND FOR FUNCTION to ex-
plain different figurative uses of hand. HAND FOR HELPING was used to explain the
meaning of give me a hand:

(4) C: “ ‘hand’ porque con las manos se puede ayudar”

C: ‘hand’ because you can help with your hands

And HAND FOR PASSING THINGS in the case of hand it to me:

(5) C: “ ‘hand’ porque normalmente las cosas se dan con la mano”

C: ‘hand’ because you normally use your hand to give things out

Many examples of children’s awareness of their environment and the influence

of this in their figurative reasoning could be found in the study that dealt with
temperature terms when applied to colours (Piquer-Píriz 2006). For instance,
when explaining why orange is a warm colour, the following exchange of infor-
mation took place among three 8-year-old children:
 Figurative Language and Young L2 Learners 69

(6) C1: “porque también cuando está anocheciendo se pone [el sol] algunas
veces detrás de las nubes se pone naranja (C2: y morao [morado] – y
morao [morado] un poquinino [poquito]) (C3: ¿Al amanecer o al ano-
checer?) (C1: a los dos, a los dos) (C2: una especie de rosa)”
C1: because at dusk it sometimes hides behind the clouds and gets orange
(C2: and purple – and a bit purple) (C3: at dawn or at dusk?) (C1: both,
both) (C2: pinkish)

These findings seem to reveal not only the importance of motivation, i.e., figu-
rative language is not arbitrary but rather grounded in human sensorimotor
and social experiences, but also the fact that young L2 learners can reflect on it.
However, these studies also show that all figurative extensions are not equally
transparent for all children at all ages. It seems that in the process of grasping the
figurative meanings of polysemous words, these young learners go from concrete,
prototypical meanings to abstract senses, a tendency that has also been reported
in their L1 (Winner 1988). Furthermore, as happens in their L1, another key aspect
that influences children’s figurative reasoning is the different types of knowledge
children are able to differentiate at different ages, what, as has been mentioned
above, some authors have called ‘their developing knowledge of the world’ (Keil
1989 or Cameron 1996). The importance of the role played by contextual factors in
this process has also been emphasised by Cameron (1996) who points out that
metaphorical expressions are not used in isolation but within a discourse context.
The children participating in my studies showed the same tendency to refer to a
context which they seem to have closely linked to words and images when, for
example, they referred to el cabeza del pelotón de la vuelta ciclista o de una carrera
(the leader of the group of cyclists or the leader in a race) to justify why the first
car in a line is the head or when they remembered a similar image or situation in
los dibujos animados (cartoons) or video juegos (computer games).
This growing knowledge may lead them, in some cases, to wrong conclu-
sions. For example, for some of the younger learners that took part in the
study, the top part of any of the cars in a line was the head of the line of cars
because that was the highest part and the head is at the top. They were over-
generalising explanations based on the human body schema whereas the older
children, with a better differentiated and more developed knowledge of do-
mains, were more flexible in the use of other schemas and applied the animal
body schema to reason about the head of a line of cars (see MacArthur and
Piquer-Píriz 2007 for further details).
Similarly, most of the participants explained that the head of the stairs is the
top part because it is the part placed at the top of a staircase. They were making a
projection grounded in their knowledge of the human body schema. In this case,
70 Ana M. Piquer-Píriz

their knowledge of ‘stairs’ and ‘heads’ matches that of an adult. However, some
of them said that the head of a line of cars was a police car placed in the middle
of the line because its lights were slightly above the roofs of the other cars. Here,
they were using the same mechanism, but in this case, it does not match the con-
ventional adult link which relates this use of ‘head’ to the notion of being in first
rather than in top place.
Studies on children’s development of conceptual domains will therefore
clarify the domain differentiations which children are able to establish at differ-
ent stages. This idea is especially relevant if the problem is approached from a
cognitive linguistic perspective. If we accept the cognitive view of metaphor
that our abstract reasoning is based on our understanding of concrete concepts
via a metaphorical projection from source/concrete to target/abstract domains,
then any methodology that aims to enhance this projection needs to take into
account what kind of domain differentiation a child is capable of at various
ages. For instance, trying to clarify the connection between coldness and un-
kindness before certain age may not be as efficient as it could be expected.
To sum up, it seems, then, that children are ready and can be helped to
make optimal use of a limited lexicon, if guided, to extend meanings through fig-
urative reasoning. This is the question I turn to in the next section.

4 Motivated Meanings, Polysemy and L2

Vocabulary Learning at Early Stages
As has been pointed out above, one of the main tenets of Cognitive Linguistics
is that figuration pervades our language and thought and that, in order to un-
derstand abstract concepts, we mostly resort to metaphorical or metonymical
connections based on our bodily sensations and our interaction with the sur-
rounding environment (Lakoff and Johnson 1980, 1999 or Lakoff 1993 for a de-
tailed description of Conceptual Metaphor Theory). This implies that, for
cognitive linguists, language is not arbitrary but mostly motivated. One of the
most comprehensive treatments of linguistic motivation in this discipline is
Panther and Radden (2004). Starting from some definitions of motivation pro-
posed by both linguists and psychologists, they put forward their own: “a lin-
guistic unit (target) is motivated if some of its properties are shaped by a
linguistic source (form and/or content) and language-independent factors. We
thus assume that linguistic motivation involves a causal relation. As suggested
by the term ‘shape’ in the above definition, however, the notion of causation is
non-deterministic.” (2004: 4)
 Figurative Language and Young L2 Learners 71

Interestingly they also noted that “traditionally, the term ‘motivation’ is ap-
plied to the form of linguistic units; more recently scholars have also applied it to
the extension of senses” (2004: 2).
The understanding of polysemy and semantic extension as motivated by meta-
phor and metonymy (understood here as mental mechanism) and grounded in ex-
periential correlations of a physical and a social nature has important implications
for L2 instruction (for some in-depth analyses of polysemy from a Cognitive
Linguistics perspective, see Cuyckens and Zawada 2001, Cuyckens, Dirven and
Taylor 2003 or Taylor 2003).
To begin with, it gave rise to models of semantic networks (Lakoff 1987,
Langacker 1990) based on the notion that the different meanings of a given lex-
eme “form a radially structured category, with a central member and links de-
fined by image-schema transformations and metaphors” (Lakoff 1987: 460).
The notion of embodiment (Johnson 1987 or Gibbs 2003) is particularly im-
portant in understanding the motivation for polysemy: in this view, human ab-
stract reasoning is embodied, that is, grounded in our sensorimotor experiences
and our interaction with the world. According to Johnson, metonymy and meta-
phor are the means by which it is possible “to ground our conceptual systems
experientially and to reason in a constrained but creative fashion” (1992: 351).
Thus, cognitive linguists consider language and its use as a reflection of gen-
eral cognitive processes, mediated by culture, and, instead of treating language
as arbitrary, plausible explanations for its characteristics are sought. The notion,
derived from these theoretical tenets, of models for semantic networks (Lakoff
1987 or Langacker 1990) in which the different meanings of a given lexeme form
radially structured categories which consist of a central member connected to
the others via metonymy, metaphor or image schema implies that whole clusters
of lexical items may be linked in systematic, non-arbitrary relations. These ideas
have had important implications for L2 instruction. Taking this premise as their
starting point, Applied Cognitive Linguists have been exploring, for over two dec-
ades, the pedagogical potential of enhancing figurative awareness to facilitate the
comprehension and retention of vocabulary in an L2 (see Boers and Lindstromberg
2006 for some theoretical considerations and Boers 2011 for a review of different
CL-oriented studies). This research has concentrated mainly on older (intermediate
or advanced) learners (Ponterotto 1994; Kövecses and Szabò 1996; Lazar 1996,
2003; Deignan, Gabrys and Solska 1997; Boers and Demecheleer 1998; Boers 2000;
Saaty 2016). However, EFL has been expanding and introduced at ever younger
ages in the educational systems of many countries and, as shown above, CL has a
contribution to make with younger learners.
In an instructed second language setting, learners’ grasp of figurative mean-
ings will largely depend on the range of senses which they meet in the graded
72 Ana M. Piquer-Píriz

input of the classroom (MacArthur and Piquer-Píriz 2007) and, therefore, which
meanings should be introduced and at what stage of the learning process are im-
portant issue to be taken into account when designing L2 syllabi (Piquer-Píriz
2011).
In general terms, MacArthur (2010: 159) describes metaphorical thinking as
a great potential for L2 learners to expand their vocabulary:

Metaphor as a way of thinking about language use is new for most learners and teachers;
it is interesting and flexible. Unlike grammar, it has no hard and fast rules: there are no
‘correct’ or ‘incorrect’ metaphors, but rather better, communicatively more successful
ones in comparison with less successful ones. It is often picturesque and imagistic, fa-
vouring the holistic learner over the analytic one (Littlemore 2001). Most important of all,
with regard to the learners of English with knowledge of a relative impoverished stock of
words, metaphor helps to make meaning from many everyday, highly familiar words, for
among all the forces that drive semantic extension, the most powerful is metaphor
(Taylor 2003). Metaphor is thus the foreign language learners’ best ally in the quest for
greater expressive powers . . . metaphor is used by learners of English, and they can be
helped to use it better, but only if it is given a prominent place in classroom discussion.

Metaphor certainly seems to have great potential for L2 learners, including very
young learners at the earliest stages of the learning process who have a re-
stricted lexicon. In fact, some of the most effective CL-oriented techniques such
as the use of total-physical response (TPR) activities (Lindstromberg and Boers
2005, Saaty 2016) are well-suited for children.
However, the application of CL-oriented pedagogies in the L2 classroom
and, particularly, with young learners needs to be cautious. In fact, CL-oriented
pedagogies have not reached L2 classrooms as much as might be expected. We
need to bear in mind that the fact that figurative meanings are motivated does
not mean that they are predictable. But this is, precisely, according to Boers
(2013), why enhancing and promoting figurative awareness can be of an aid for
L2 learners and why teachers and teaching materials should foster the underly-
ing connections.
This author has suggested that rather than encouraging learners’ ability to
generate L2 metaphors, it may be more successful to instil a “metaphor aware-
ness” in learners to “organize the steady stream of figurative language they are
exposed to” (Boers 2000: 564). In this same line is MacArthur’s (2010) paper on
metaphorical competence in EFL where she argues for the importance of adding
‘metaphor awareness’ generally to the experience of learning English in instructed
L2 setting: “In this approach, the specific pedagogical techniques employed [. . .]
are seen to be less significant than the general foregrounding of metaphor and the
effect(s) this may have on learners’ growing awareness of how metaphor perme-
ates language (their own and the L2 to be learnt).” (MacArthur 2010: 157)
 Figurative Language and Young L2 Learners 73

The sooner this awareness of metaphor is introduced in the L2 classroom,

taking into account all the aspects that have been pointed out above, the better
for the learning process.

5 Conclusions
Children, like adults, are endowed with the unique capacity of humans to under-
stand one thing in terms of another, and this ability is also available to them in
an L2 and could, therefore, be exploited to help them build their own L2 lexicon
in a systematic way. Even very young learners are ready to extend the possibilities
of a limited vocabulary in the L2 and benefit from an instructional approach that
relies on metaphorical competence or, more generally, figurative thought, in the
way applied cognitive linguists have been advocating. In fact, since the L2 lexicon
of a beginner is particularly restricted, helping these learners to use their vocabu-
lary as productively as possible will be especially advantageous for them.
Furthermore, learning a language is a developmental process and its first stages
are fundamental in terms of establishing a solid basis for future development.
Encouraging young learners to exploit this capacity and clarifying the prototypi-
cal meanings of core lexical items would lay the foundations for later extensions
in other areas. Stimulating this capacity alone may not suffice to ensure effective
learning and factors such as the children’s growing knowledge of the world and
their communicative needs at different ages should also be taken into account
when selecting what meanings should be taught and when they should be intro-
duced in the young learners’ L2 classroom. However, making young learners
aware of their own ability to establish connections among the different senses of
a word will empower them to build a systematic, rich L2 lexicon.

References
Abkarian, Gene G., Alison Jones & Gretchen West. 1992. Enhancing children’s communication
skills: idioms ‘fill the bill’. Child Language Teaching and Therapy 6(3). 246–254.
Asch, Solomon & Harriet Nerlove. 1960. The development of double function terms in children.
In Bernard Kaplan & Seymour Wapner, (eds.), Perspectives in psychological theory,
47–60. New York: International Universities Press
Billow, Richard. 1975. A cognitive developmental study of metaphor comprehension.
Developmental Psychology 11(4). 425–423.
Boers, Frank. 2000. Metaphor awareness and vocabulary retention. Applied Linguistics 21.
553–571.
74 Ana M. Piquer-Píriz

Boers, Frank. 2011. Cognitive semantic ways of teaching figurative phrases: An assessment.
Review of Cognitive Linguistics 9. 227–261.
Boers, Frank. 2013. Cognitive Linguistic approaches to teaching vocabulary: Assessment and
integration. Language Teaching 46. 208–224.
Boers, Frank & Murielle Demecheleer. 1998. A cognitive semantic approach to teaching
prepositions. English Language Teaching Journal 53. 197–204.
Boers, Frank & Seth Lindstromberg. 2008. How cognitive linguistics can foster effective
vocabulary teaching. In Frank Boers & Seth Lindstromberg (eds.), Cognitive linguistic
approaches to teaching vocabulary and phraseology, 1–61. Berlin: Mouton de Gruyter.
Boers, Frank & Seth Lindstromberg. 2006. Cognitive linguistic approaches to second or
foreign language instruction: Rationale, proposals and evaluation. In Gitte Kristaensen,
René Dirven, Michel Achard & Francisco J. Ruiz-Mendoza (eds.), Cognitive linguistics:
current applications and future perspectives, 305–358. Berlin: Mouton de Gruyter.
Cacciari, Cristina & M. Chiara Levorato. 1989. How children understand idioms in discourse.
Journal of Child Language 16. 387–405.
Cacciari, Cristina & M. Chiara Levorato. 1998. The effect of semantic analyzability of idioms in
metalinguistic Tasks. Metaphor and Symbol 13 (3). 159–177.
Cameron, Lynne. 1996. Discourse context and the development of metaphor in children.
Current Issues in Language and Society 3(1). 49–64.
Cameron, Lynne. 2003. Metaphor in educational discourse. London: Continuum.
Carey, S. 1985. Conceptual changes in childhood. Cambridge, MA: MIT Press.
Clark, Eve V. 1993. The lexicon in acquisition. Cambridge: Cambridge University Press.
Cuyckens Hubert, René Dirven & John R. Taylor. 2003. Cognitive approaches to lexical
Semantics. Berlin: Mouton de Gruyter.
Cuyckens, Hubert & Britta Zawada (eds.). 2001. Polysemy in cognitive linguistics. Amsterdam:
John Benjamins.
Danesi, Marcel. 1986. The role of metaphor in second language pedagogy. Rassegna Italiana
di Linguistica Applicata 18. 1–10.
Danesi, Marcel. 1992. Metaphorical competence in second language acquisition and second
language teaching: The neglected dimension. In James E. Alatis (ed.), Language
communication and social meaning (Georgetown University Round Table on Languages
and Linguistics), 489–500. Washington: Georgetown University Press.
Danesi, Marcel. 2008. Conceptual errors in second-language learning. In Sabine De Knop &
Antoon De Rycker (eds.), Cognitive approaches to pedagogical grammar, 231–256.
Berlin: Mouton de Gruyter.
Danesi, Marcel. 2016. Conceptual fluency in second language teaching: An overview of
problems, issues, research findings and pedagogy. International Journal of Applied
Linguistics & English Literature 5(1). 145–153.
Deignan, Alice, Danuta Gabryś & Agnieszka, Solska. 1997. Teaching English metaphors using
cross-linguistic awareness-raising activities. ELT journal 51(4). 352–360.
Gardner, Howard & Ellen Winner. 1978. The development of metaphoric competence:
Implications for humanistic disciplines. Critical Inquiry 5(1). 123–141.
Gardner, Howard, Mary Kircher, Ellen Winner & David Perkins. 1975. Children’s metaphoric
productions and preferences. Journal of Child Language 2. 125–141.
Grady, Joseph E. 1997. Foundations of meaning: Primary metaphors and primary scenes.
PhD Dissertation, University of California, Berkeley.
 Figurative Language and Young L2 Learners 75

Gentner, Dedre. 2003. Why we’re so smart. In Dedre Gentner & Susan Goldin-Meadow (eds.),
Language in mind, 195–235. Cambridge, MA: MIT Press.
Gibbs, Raymond. W. 1987. Linguistic factors in children’s understanding of idioms. Journal of
Child Language 14. 569–586.
Gibbs, Raymond. W. 1991. Semantic analyzability in children’s understanding of idioms.
Journal of Speech and Hearing Research 34. 613–620.
Gibbs, Raymond. W. 1994. The poetics of mind. New York: Cambridge University Press.
Gibbs, Raymond W. 1999. Researching metaphor. In Graham Low & Lynne Cameron (eds.),
Researching and applying metaphor, 29–47. Cambridge: Cambridge University Press.
Gibbs, Raymond W. 2003. Embodied experience and linguistic meaning. Brain and Language
84. 1–15.
Johnson, Mark. 1987. The Body in the mind: The bodily basis of meaning, imagination, and
reason. Chicago: University of Chicago Press.
Johnson, Christopher. 1999. Constructional grounding: The role of interpretational overlap in
lexical and constructional acquisition. PhD Dissertation, University of California,
Berkeley.
Keil, Frank. 1989. Concepts, kinds and cognitive development. Cambridge, MA: MIT Press.
Kövecses, Zoltán & Péter Szabò. 1996. Idioms: A view from cognitive semantics. Applied
Linguistics 17(3). 326–355.
Lakoff, George. 1987. Women, fire, and dangerous things: What categories reveal about the
mind. Chicago: University of Chicago Press.
Lakoff, George. 1993. The Contemporary theory of metaphor. In Andrew Ortony (ed.),
Metaphor and thought, 202–251. 2nd edn. Cambridge: Cambridge University Press.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago
Press.
Lakoff, George & Mark Johnson. 1999. Philosophy in the flesh: The embodied mind and its
challenge to western thought. New York: Basic Books.
Langacker, Ronald W. 1987. Foundations of cognitive grammar. Vol. 1, Theoretical
prerequisites. Stanford, CA: Standford University Press.
Langacker. Ronald W. 1990. Concept, Image and Symbol. The Cognitive Basis of Grammar.
Berlin: Mouton de Gruyter.
Lazar, Gillian. 1996. Using figurative language to expand students’ vocabulary. English
Language Teaching Journal 50(1). 43–51.
Lazar, Gillian. 2003. Meanings and metaphor: Activities to practise figurative language.
Cambridge: Cambridge University Press.
Levorato, M. Chiara & Cristina Cacciari. 1992. Children comprehension and production of
idioms: The role of context and familiarity. Journal of Child Language 19. 415–433.
Lindstromberg, Seth & Frank Boers. 2005. From movement to metaphor with
manner-of-movement verbs. Applied Linguistics 26(2). 241–261.
Littlemore, Jeannette. 2001. Metaphoric competence: a possible language learning strength of
students with a holistic cognitive style? TESOL Quarterly 35(3). 459–491.
Littlemore, Jeannette. 2010. Metaphoric competence in the first and second language:
similarities and differences. In Martin Pütz & Laura Sicola (eds.), Cognitive processing
in second language acquisition: inside the learner’s mind, 293–315. Amsterdam: John
Benjamins.
Littlemore, Jeannette & Graham D. Low. 2006a. Figurative thinking and foreign language
learning. Basingstoke: Palgrave Macmillan.
76 Ana M. Piquer-Píriz

Littlemore, Jeannette & Graham D. Low. 2006b. Metaphoric competence and communicative
language abilities. Applied Linguistics 27(2). 268–294.
Low, Graham D. 1988. On teaching metaphor. Applied Linguistics 9(2). 125–147.
MacArthur, Fiona. 2006. Metaphor in the real world of EFL: Coursebooks for learners aged
12–18. Paper presented at the RaAM 6 Conference, Researching and Applying Metaphor:
Ten Years On, University of Leeds, 10–12 April.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where are we and where should we
be going? A view from the language classroom. AILA Review 23. 155–173.
MacArthur, Fiona & Ana M. Piquer Píriz. 2007. Staging the introduction of figurative
extensions of familiar vocabulary items in EFL: Some preliminary considerations. Ilha do
desterro: A Journal of English Language, Literatures in English and Cultural Studies, 53.
123–134.
Nacey, Susan. 2013. Metaphors in learner english. Amsterdam: John Benjamins.
Nacey, Susan. 2017. Metaphor comprehension and production in a second language. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
503–515. Taylor & Francis.
Nerlich, Brigitte, David. D. Clarke & Zazie Todd. 1999. Mummy, I like being a sandwich.
In Klaus U Panther, and Günter Radden (eds.), Metonymy in language and thought,
361–383. Amsterdam: John Benjamins.
Nerlich, Brigitte, Zazie Todd & David D. Clarke. 2003. Emerging patterns and evolving
polysemies: The acquisition of get between four and ten years. In Brigitte Nerlich, Zazie
Todd, Vimala Herman and David Clarke (eds.), Flexible patterns of meaning in mind and
language, 333–357. Berlin: Mouton de Gruyter.
Nippold, Marilyn, Laurence B. Leonard & Robert Kail. 1984. Syntactic and conceptual factors in
children’s understanding of metaphors. Journal of Speech and Hearing Research 27.
197–205.
Nippold, Marilyn & Michael Sullivan. 1987. Verbal and perceptual analogical reasoning and
proportional metaphor comprehension in young children. Journal of Speech and Hearing
Research 30. 367–376.
Özçalışkan, Şeyda. 2011. Acquisition of metaphor. In Patrick C. Hogan (ed.), Cambridge
encyclopedia of language sciences, 486–488. Cambridge:
Cambridge University Press.
Özçalışkan, Şeyda. 2014. Development of metaphor. In Patricia Brooks & Vera Kempe (eds.),
Encyclopedia of Language Development, 374–375. New York: Sage Publishers.
Radden, Günter & Klaus U. Panther. 2004. Introduction: Reflections on motivation. In Günter
Radden & Klaus U. Panther (eds.), Studies in linguistic motivation, 1–26. Berlin: Mouton
de Gruyter.
Piaget, Jean & Bärbel Inhelder, 1969. Psicología del niño. Madrid: Ediciones Morata, S.A.
Piaget, Jean. 1962. Play, dreams and imitation in childhood. New York: Norton
Pineda, Patricia. 2015. Looking for and making sense of “special” words. Metaphor
recognition and interpretation by school children. Utrecht: LOT dissertation series.
Piquer-Píriz, Ana M. 2005. La comprensión de algunas extensiones semánticas de los lexemas
“hand”, “mouth” y “head” en las primeras etapas del aprendizaje del Inglés. PhD
Dissertation, University of Extremadura. Cáceres: Servicio de publicaciones de la
Universidad de Extremadura.
 Figurative Language and Young L2 Learners 77

Piquer-Píriz, Ana M. 2006. The embodied and social grounds for figurative meanings: Colour
terms in EFL. Paper presented at the II International Workshop on Metaphor and
Discourse, University Jaume I, Castellón (Spain).
Piquer-Píriz, Ana M. 2008a. Reasoning figuratively in early EFL: Some implications for the
development of vocabulary. In Frank Boers & Seth Lindstromberg (eds.), Cognitive
linguistic approaches to teaching vocabulary and phraseology, 219–240. Berlin: Mouton
de Gruyter
Piquer-Píriz, Ana M. 2008b. Young learners’ understanding of figurative language. In Mara
S. Zanotto, Lynne Cameron & M. Chiara Calvacanti (eds.), Confronting metaphor in use:
An applied linguistic approach, 183–198. Amsterdam: John Benjamins.
Piquer-Píriz, Ana M. 2010. Can people be cold and warm?: Developing understanding of some
figurative meanings of temperature terms in early EFL. In Graham D. Low, Zazie Todd,
Alice Deignan & Lynne Cameron (eds.), Researching and applying metaphor in the real
world, 21–33. Amsterdam: John Benjamins.
Piquer-Píriz, Ana M. 2011. Motivated word meanings and vocabulary learning: The polysemy of
hand in the english for young learners classroom. Metaphor and the Social World 1 (2).
154–173.
Piquer-Píriz, Ana M. & Frank Boers. 2010. Individual differences in L2 metaphor processing:
A think-aloud experiment. Paper presented at the VII Congreso Internacional de la
Asociación Española de Lingüística Cognitiva (AELCO). University of Castilla la Mancha
(Spain), 30–2 October.
Piquer Píriz, Ana M., Fiona MacArthur & Rafael Alejo. 2010. Does learning an L2 promote
metaphorical thinking? Some preliminary findings from a study of Spanish secondary
school learners in CLIL and non-CLIL contexts. Paper presented at the Eighth
International Conference on Researching and Applying Metaphor (RaAM 8). VU University
(The Netherlands), 30–3 July.
Ponterotto, Diane. 1994. Metaphors we can learn by. English Teaching Forum 32 (3). 2–7.
Saaty, Rawan. 2016. Teaching L2 metaphor through awareness-raising activities: experimental
studies with saudi EFL learners. Unpublished Doctoral Dissertation. University of
Birmingham.
Stites, Lauren J. & Şeyda Özçalışkan. 2013. Developmental changes in children’s
comprehension and explanation of spatial metaphors for time. Journal of Child Language
40. 1123–1137.
Taylor, J. R. 2003. Polysemy’s Paradoxes. Language Sciences 25. 637–655.
Tomasello, Michael. 2003. Constructing a language: A usage-based theory of language
acquisition. Cambridge, MA: Harvard University Press.
Tomasello, Michael. 2006. Construction Grammar for kids. Constructions 1. 1–11.
Vosniadou, Stella. 1987. Children and metaphors. Child Development 58. 870–885.
Vosniadou, Stella & Andrew Ortony. 1983. Intelligence of the literal-metaphorical-anomalous
distinction in young children. Child Development 54. 154–161.
Vosniadou, Stella. 1989. Similarity and analogical reasoning: A synthesis. In Stella Vosniadou
and Andrew Ortony, eds., Similarity and analogical reasoning, 1–18. Cambridge:
Cambridge University Press.
Winner, Ellen. 1988. The point of words: Children’s understanding of metaphor and irony.
Cambridge, MA: Harvard University Press.
Zurer-Pearson, Barbara. 1990. The comprehension of metaphor by preschool children. Journal
of Child Language 17. 185–203.
Laura Suárez-Campos, Alberto Hijazo-Gascón
and Iraide Ibarretxe-Antuñano
Metaphor and Spanish as a Foreign
Language

1 Introduction: Conceptual Metaphor

and Metaphorical Competence
Since the publication of Lakoff and Jonson’s Metaphors we live by (1980), where
the basic tenets of Conceptual Metaphor Theory (henceforth, CMT) were estab-
lished, metaphor stopped being considered just a rhetorical figure and is now rec-
ognised as an essential mechanism in the construction of language. A conceptual
metaphor is defined as a cognitive mechanism that establishes mappings between
a source domain –usually but not necessarily a concrete domain– and a target
domain, generally a more abstract domain, e.g. UNDERSTANDING IS SEEING. A meta-
phorical expression, on the other hand, is characterised as the multimodal lin-
guistic encoding of a conceptual metaphor particular to each language, e.g. I see
what you mean. The mappings between conceptual domains are not arbitrary, but
motivated. This means that they are experientially grounded in the world around
us. In other words, the experiential basis of conceptual metaphors is embodied,
that is, based on our interaction with the physical, social, and cultural dimen-
sions of our world (see Johnson 1987).
This embodied nature explains why the same conceptual metaphors can be
found cross-linguistically. For example, the metaphor UNDERSTANDING IS SEEING

Note: This research has been supported by grants FFI2013-45553-C3-1-P and FFI2017-82460-P
from the Spanish Ministry of Economy and Competitiveness and by the Government of Aragon
(Psylex H11–17R). Special thanks to the editors for their kind invitation to take part in this vol-
ume and to Chris Skedgel for his help. We would like to dedicate this chapter to Fiona, not
only for her inspiring work on metaphor and metaphorical competence in different contexts,
but also, and above all, for being herself, a truly and fully “competent horsewoman”.

Laura Suárez-Campos, Departamento de Lingüística General e Hispánica, Universidad de

Zaragoza, Zaragoza, Spain, e-mail: [email protected]
Alberto Hijazo-Gascón, School of Politics, Philosophy, Language and Communication
Studies, University of East Anglia, Norwich, United Kingdom,
e-mail: [email protected]
Iraide Ibarretxe-Antuñano, Departamento de Lingüística General e Hispánica, Universidad
de Zaragoza, Zaragoza, Spain

https://doi.org/10.1515/9783110630367-005
80 Laura Suárez-Campos et al.

also works in Spanish in metaphorical expressions such as Ya veo lo que quieres

decir (Lit. [I] already see what [you] want [to] say), Tienes que echarle otro vistazo
(Lit. [you] must throw it another look), or El problema estaba bien claro (Lit. the
problem was well clear) (see MacArthur, Krennmayr, and Littlemore 2015).
Over the years since Lakoff and Johnson’s publication, CMT has been used
to unveil the conceptualisation of a wide variety of domains and fields of
knowledge. For instance, CMT has explored the scope of metaphorical source
domains such as perception (Speed et al. 2019) or body parts (Maalej and Yu
2011), but also metaphorical target domains such as time (Yu 1998) or emo-
tions (Kövecses 2000). Although English is the predominant language in this
type of studies, other languages have also been explored. One example is Yu’s
research on anger and happiness in Chinese (Yu 1998). There are also some
cross-linguistic studies comparing several languages, such as Ogarkova and
Soriano (2018). As far as Spanish is concerned, work on conceptual metaphor
is long and well-established in some areas such as emotions (Barcelona and
Soriano 2004; Soriano 2003; Soriano and Valenzuela 2009), but still incipient
in other areas such as, for instance, body parts (but see, Gutiérrez Pérez 2010;
López Rodríguez 2009; Olza 2011).
Another important contribution of CMT has been to offer plausible solu-
tions and explanations to traditional problems and long standing issues such
as the question about how and when figurative (abstract) concepts are ac-
quired, i.e. metaphoric competence in the L1 (Cometa and Eson 1978). Danesi
(1988) coins the term metaphorical competence for the specific acquisition of
figurative concepts in an L2. In both cases, the key question is to explain
how a language codifies abstract concepts through metaphorical reasoning,
but whereas this competence develops naturally and unconsciously in an L1,
a set of skills need to be learned in an L2 to achieve a good level of metaphor-
ical competence (Danesi 1988).
The L2 skills that learners need to develop in order to acquire this meta-
phorical competence, however, do not start from scratch. They can be trans-
ferred from the learner’s L1 (Danesi 1992; Johnson 1991; Littlemore 2010), but
they should be also specifically incorporated in foreign language teaching
(Danesi 1988, 1992, 2008; Littlemore 2001a, 2001b; Littlemore and Low 2006;
Low 1988). As Danesi (1988) points out, L2 learners can use metaphor both as
an interaction strategy and as a resource for meaning inference. In order to do
so, however, learners need to be aware of how concepts are encoded and re-
flected in language through metaphorical reasoning. This is what Boers (2000)
calls metaphorical awareness; that is, the ability to recognise the non-arbitrary
relationship between the form of a metaphorical expression and its meaning, as
 Metaphor and Spanish as a Foreign Language 81

well as the acceptance of cultural and linguistic differences in metaphorical ex-

pressions (Boers, Demecheleer, and Eyckmans 2004). Metaphorical awareness
is not just crucial to help learners improve their metaphorical competence, and
hence, their linguistic skills in an L2, it will also improve their memorising
(Boers and Lindstromberg 2008).
The acquisition of metaphorical competence and its implementation in the
foreign language class is becoming a hot topic in second language acquisition
(see MacArthur 2017 for a review). Although most of the earlier studies focused
on the acquisition and teaching of English as a foreign language (henceforth
EFL), nowadays there is a growing body of research that pays attention to other
languages. The goal of this chapter is to offer an overview of the research car-
ried out around the role of conceptual metaphor and metaphorical competence
in Spanish as a foreign language (henceforth, SFL). Section 2 summarises some
of the main studies on the acquisition, both comprehension and production, of
metaphorical competence in SFL. Section 3 presents some of the pedagogical
implementations of CMT in SFL and discusses how and to what extent this prac-
tice could be beneficial for second language learners.

2 The Acquisition of Metaphorical Competence

in SFL
As mentioned above, there is a growing interest in the development of meta-
phorical competence in L2 acquisition and its practical application in the class-
room, and SFL is no exception (see Suárez-Campos and Hijazo-Gascón 2019, for
an overview). Recent research on metaphor in SFL shows that metaphor reason-
ing brings benefits to the acquisition of linguistic patterns, such as the verbal
system (Nieto and Martínez Vázquez 2014; Real Espinosa 2009), verbal mor-
phology (Ruiz Campillo 2014), and the use of prepositions (Llopis-García 2015;
Mendo Murillo 2019). Most of the studies on metaphor and SFL focus on vocab-
ulary learning and on fostering metaphorical competence, following previous
studies in EFL (Boers 2000, 2006, 2013; Danesi 1992, 2008; Littlemore 2001a,
2001b, 2010; Littlemore et al. 2012, 2013; Low 1988; MacArthur 2010).
The main findings of the studies on metaphor in SFL will be discussed in
this section. First, studies dealing with comprehension will be presented
and, second, those focused on metaphor production in SFL. Finally, we will
review the studies that propose pedagogical materials using metaphors to be
applied in SFL instruction.
82 Laura Suárez-Campos et al.

2.1 Comprehending Metaphors in Spanish

Learners’ comprehension of the different elements that are involved in a meta-

phorical expression is crucial to achieve metaphorical competence. Studies in
EFL show that metaphorical comprehension depends on the strategy used by
the student to infer the meaning of the expressions (Azuma 2009; Littlemore
2004). This choice of strategy is related to students’ cognitive style (Boers and
Littlemore 2000). While analogical learners focus on the correspondences be-
tween the source and the target domain, holistic learners understand both do-
mains as a whole. However, L2 learners can also adapt their strategies to the
situation, by using different cognitive processes, such as associative thinking,
analogical reasoning, image formation, and metaphoric extension (Littlemore
2008; Littlemore and Low 2006).
There are several strategies used to infer meaning in an L2. They can be
based on linguistic knowledge (about the L2 in general or in relation to its simi-
larities with the L1) or on cognitive reasoning (Polisse 1990). One of the most
frequent non-linguistic strategies used by L2 students is finding metaphorical
motivations. In Ferreira’s study (2008), participants used this strategy even in
cases where they could rely on the context to infer meaning. Students can also
search for conceptual similarities between the L1 and the L2. However, this is a
more challenging strategy as they lack awareness about these similarities be-
tween languages (Chen and Lai 2013).
The studies focused on strategies to infer metaphorical expressions in SFL
are consistent with previous research in EFL (De Cock and Suñer 2018; García
Caballero 2017; Mayo Martín 2017). Their results also point out that the influence
of the learning styles of the students is crucial for the choice of the strategies.
They also show that the importance of context is minimal and that strategies
based on conceptual similarities are not straightforward for L2 learners.
García Caballero (2017) notes that the students’ cognitive styles and the type
of metaphor are conditioning factors for the comprehension of metaphors in
SFL. In her study, she checked the ability of a group of SFL students to identify
metaphors in a text. The SFL students had a high level of proficiency (B2–C1)
and different L1s. Taking into account their cognitive style, students were classi-
fied according to two axes: visual/verbal and intuitive/perceptive. Learners with
a visual cognitive style are better at interpreting metaphors with an external
image structure, as in en las penumbras del amor (Lit. in the shadows of the
love) ‘in the dark hours of love’ while verbal cognitive style learners find it
easier to process metaphors with a simile structure, as in como Pedro por su
casa (Lit. like Pedro around his house) ‘like he owned the place’. Perceptive
 Metaphor and Spanish as a Foreign Language 83

learners prefer metaphors that are also lexicalised in the L1 whereas intuitive
learners favour novel metaphors.
Mayo Martín (2017) focuses on colloquial language and on the role of con-
text and cultural distance to infer metaphorical meaning. Her results show that
the conceptual distance between the languages seems to affect the comprehen-
sion of metaphorical expressions. L2 learners could infer more easily the mean-
ing of expressions that belong to conceptual metaphors shared in the L1 and
the L2. For example, they were presented with two euphemistic metaphorical
expressions meaning ‘to die’, one with a counterpart in their L1s, no estar (‘not
be’), and one without, palmarla (Lit. print the palm of the hand). Students were
able to infer the meaning of the former but not the latter. Context was not al-
ways used as a strategy to infer the meaning of these expressions, which is con-
sistent with Ferreira’s (2008) results for EFL.
A similar methodology is used by De Cock and Suñer’s (2018) study. French-
speaking Belgian learners of Spanish had to explain the meaning of sexual taboo
metaphorical expressions that were presented to them with and without context.
The expressions chosen present some similarities with the L1, but also some con-
ceptual and sociocultural differences. For instance, the Spanish term los huevos
(Lit. the eggs) referring to the testicles and the expression la tiene morcillona (Lit.
he has it like a blood sausage) to indicate the state before an erection are not
shared with French and their comprehension needs some cultural knowledge.
Interestingly, the level of proficiency (B2 and C1) in De Cock and Suñer’s study
did not play a role in the comprehension of metaphorical expressions. In fact, the
variable that proved to be determinant was whether participants had an immer-
sion experience in a Spanish-speaking country, as learners who had lived in a
Spanish-speaking country had better results. Since these are taboo expressions re-
lated to sex, they are rarely treated in an L2 class, and therefore they can only be
acquired through frequent exposure to different registers and types of discourse.
These studies identify some of the difficulties that SFL students face when
they have to infer the meaning of Spanish metaphorical expressions. For exam-
ple, their lack of awareness of conceptual similarities between languages seems
to be a problematic area that hinders the correct comprehension of metaphorical
expressions. These studies also highlight some of the determining factors for
meaning comprehension, such as the cognitive style of the students and their ex-
posure to the target language. They also show that, interestingly enough, the
level of proficiency does not seem to be particularly relevant when students have
to interpret the meaning of metaphorical expressions. It is also very revealing to
find that learners do not tend to rely on context to infer these meanings. More
studies on metaphor comprehension in SFL are needed (on different registers
84 Laura Suárez-Campos et al.

and contexts) to achieve a better understanding on how metaphorical meanings

are inferred in SFL.

2.2 Producing Metaphors in L2 Spanish

The production of metaphors has been mostly studied in relation to written dis-
course in EFL (Littlemore et al. 2012, 2013; Nacey 2013; MacArthur 2010; Paris
2014a, 2014b), although there are a few studies in other L2s, such as Norwegian
(Golden 2012) and Spanish (Gómez Vicente 2013, 2019; Masid 2014). There is
nevertheless a growing interest in oral production, mainly in EFL (MacArthur
2011, 2016a, 2016b; MacArthur, Krennmayr, and Littlemore 2015; Paris 2014a).
These studies show differences in the use of figurative language in the L1
and in the L2. As one would assume, native speakers use more metaphorical ex-
pressions than L2 learners. As Nacey (2013) explains, however, it is important to
clarify that this difference in frequency is related to the different types of meta-
phors that native and non-native speakers produce. L1 speakers tend to use more
conventional metaphors, while L2 speakers use novel metaphors with more fre-
quency. These novel metaphors in the L2 are in many cases transferred from
their L1 and consequently they tend to be considered inappropriate for the L2. It
is also noted that metaphors in the L2 are more frequently used by learners with
higher levels of proficiency (Golden 2012; Littlemore et al. 2012, 2013).
Results in these studies in SFL are consistent with the findings for EFL. Gómez
Vicente (2013, 2019) explores how emotions are expressed through metaphor in
SFL by L1 French students. Her results confirm that metaphor is very frequently
used in the expression of emotions. As expected, metaphor is more used by native
speakers. However, it is important to note that non-figurative language is preferred
in both groups. Literal statements are more used than metaphorical expressions,
even in the case of native speakers’ narratives. When looking into the use of meta-
phors, native speakers use more figurative language than L2 speakers and, conse-
quently, a lower metaphorical density is found in L2 texts. The results also show
that the vast majority of metaphorical expressions used are related to the spatial
domain, closely linked with embodied motivation (Gómez Vicente 2013). This
study shows the importance of metaphor to express abstract concepts such as
emotions, and its benefits to L2 learners to improve their written performances.
The number of studies in metaphorical language production in L2 is still lim-
ited, in relation to both written and oral discourse. However, production is cru-
cial to understand the acquisition of metaphors in an L2. It seems clear that L2
learners need to understand and produce metaphors in their written texts and in
their oral expression. Future research is needed to shed more light on how
 Metaphor and Spanish as a Foreign Language 85

learners use figurative language in Spanish L2 and into what strategies they use
during these processes, especially in the understudied area of oral production.

2.3 Metaphor in Language Instruction

There are a number of studies that include both comprehension and production
of metaphor in Spanish L2. Their aim is to assess the impact of pedagogical in-
tervention in the production and comprehension of metaphors at different pro-
ficiency levels (Acquaroni 2008; Acquaroni and Suárez-Campos 2019; Lavín
Carrera 2016; Masid 2014, 2017). Participants in these studies completed a pre-
test before receiving instruction on metaphor, and a post-test following the in-
struction. The results of both tests are contrasted in order to find whether there
is an improvement in the metaphorical competence of the students, indepen-
dent of their cognitive styles or their level of proficiency.
In Acquaroni’s (2008) longitudinal study, SFL learners with mixed levels
of proficiency (from B1 to C2), enrolled in a creative writing course during a
full term. These students were trained in understanding and producing meta-
phors in Spanish. During this course, they had different tasks, individually or
in groups, with the aim of improving their writing skills. Their level of meta-
phorical competence was compared to a control group of SFL students. Results
show that students who prefer a visual cognitive style have better results in
the pre-test. However, the increase in metaphorical density is similar for all the
students in the post-test, regardless of their cognitive style. Therefore, the ped-
agogical intervention is beneficial for all the students.
Masid’s research (2014, 2017) tests the efficiency of a pedagogical inter-
vention designed to develop metaphorical competence with the aid of a web-
site. There were two groups of informants: one group had informants from
Slovakia and received online training whereas the other group had German
informants who followed the instruction as part of their on-site Spanish clas-
ses. Results show a positive effect of pedagogical intervention in both produc-
tion and comprehension regardless of the informants’ level of proficiency
(levels A2 to B2). Masid concludes that the explicit treatment of metaphor in
the classroom improves both the comprehension and the production of the
metaphorical expressions in SFL at any level. Consequently, the introduction
of activities with metaphor in the SFL classes is beneficial and complements
the learning of the L2.
Similarly, Lavín Carrera (2016) argues for the benefits of introducing meta-
phors in classroom instruction. In this study, students’ metaphor awareness im-
proves after being exposed to a corpus of metaphorical expressions compiled
86 Laura Suárez-Campos et al.

from a sample of textbooks and classified on the basis of the domains involved
in their underlying mappings.
The abovementioned studies show the benefits of metaphor instruction to
SFL students. These are relevant to all levels of proficiency and independent of
the student’s learning style. Future research should include replications of these
studies comparing these learning gains with non-metaphor-oriented interven-
tions. It is also necessary to carry out studies with metaphors in the teaching of
Spanish in lower levels of proficiency and in child second language acquisition.

3 Pedagogical Proposals for Teaching Metaphor

in SFL
Previous studies in English and Spanish show that introducing metaphors in lan-
guage instruction is beneficial for all types of learners, at any age (MacArthur and
Piquer-Píriz 2007; Piquer-Píriz 2008a, 2008b, 2011) and at any level of proficiency
(Littlemore et al. 2012, 2013). However, these findings on the benefits of meta-
phorical competence are not reflected in language teaching instruction and meta-
phors are still considered a difficult area of language in relation to literary skills.
This absence of metaphorical competence in second language instruction is
probably caused by the fact that the use of metaphor has not been systematised
in the descriptors in the CEFRL (Common European Framework of Reference for
Languages 2001). Consequently, there are only minimal references to metaphor in
the descriptors of the CPCI (Curriculum Plan of the Cervantes Institute 2006),
which specifies CEFRL descriptors for the teaching of Spanish. The CPCI includes
metaphorical expressions only from a B2 level onwards.1 As a result, explicit treat-
ment of metaphors is absent in SFL materials and textbooks or, if they are in-
cluded, they are always in relation to literary texts for higher levels of proficiency.
There are, however, some signs of change in language teachers’ views on
metaphor and an increasing awareness of the potential of metaphor in SFL learn-
ing process. In a survey carried out by Rivera León (2016), more than 80% of the
SFL teachers surveyed agreed or strongly agreed with the idea that metaphor
plays a major role in the compensatory and interactive aspects of strategic

1 The CEFRL levels have been compared to the levels established by the American Council for
the Teaching of Foreign Languages (ACTFL). The B2 Level of the CEFRL can correspond to an
Advanced Mid, following Vandergrift (no date). Martínez Baztán (2008) considers that the oral
evaluation of B2 corresponds to an Intermediate High/Advanced Low of the ACTFL. For the
equivalences of CEFRL levels with other institutions see Vandergrift (no date).
 Metaphor and Spanish as a Foreign Language 87

competence. More than 67% were interested in integrating metaphor in their SFL
classes. However, most teachers believed that explicit exposure to metaphor in
the classroom is insufficient. They perceived that the use of metaphor in class
tends to be restricted to high level groups (from B2). Only 10% of SFL teachers
claimed to integrate activities involving metaphor in the classes for lower profi-
ciency levels. In addition, the vast majority of them considered that the resources
they can find to integrate this practice into the classroom are scarce.
The reviews of SFL textbooks made by Masid (2014) and Mayo Martín (2017)
confirm this lack of metaphor-related content in textbooks and pedagogical ma-
terials. Therefore, most publications on metaphor in SFL aim at investigating
the practical application of CMT in SFL class, and to apply the findings of re-
search on the acquisition of metaphorical competence to pedagogical proposals
that can be used in SFL instruction.

3.1 Metaphor in the Classroom

As mentioned above, one of the main difficulties faced by students when they deal
with metaphorical expressions in an L2 is the lack of awareness of the cultural sim-
ilarities and differences that often underlie the meaning of a metaphorical expres-
sion (Danesi 1992). Boers (2000, 2006, 2013) argues that metaphorical awareness
by L2 students can be fostered if both the context of use and the origin of meta-
phorical expressions are explained. He also suggests focusing students’ attention
on the source domain in order to establish a system of meanings associated with
this source domain. Other authors in the field of L2 metaphor acquisition have also
claimed the need for metaphors to be explicitly taught in the classroom (Boers and
Lindstromberg 2008; Kalyuga and Kalyuga 2008), and the need to develop specific
pedagogical proposals based on thematic groupings, the promotion of metaphori-
cal extensions, and analogical reasoning (Littlemore 2001b). In some cases, the
relevance of using photographs and audiovisual materials is also highlighted
(Littlemore and Low 2006).
Some of the studies on Spanish provide guidelines for the introduction of
metaphor in the teaching of SFL (Bandera-Nápoles 2016; Gómez Vicente 2013;
Guillén Solano 2012; Hijazo-Gascón 2011; Lantolf and Bobrova 2014; Masid
2014). These are concerned with the appropriate choice of the metaphorical ex-
pressions and suggest how to introduce them in the class. They also agree with
Boers (2000, 2006) in presenting metaphors explicitly, paying attention to the
context and their motivation. Finally, they are also systematic in the activities
with metaphor in the class. Let us review some of these studies.
88 Laura Suárez-Campos et al.

Hijazo-Gascón (2011) argues for the importance of metaphors to develop

learning and communication strategies. He focuses on metaphors for emotions
and considers that the use of metaphor in class benefits analogical reasoning,
awareness of the differences between L1 and L2, and fosters strategic competen-
ces. He proposes the use of conceptual metaphors to talk about emotions and in
relation to other linguistic content such as verbs of change ponerse ‘get’, volverse
‘turn’, and quedarse ‘remain’, vocabulary for emotions as well as pragmatic and
discourse-related strategies.
Based on teaching observations of SFL classes, Guillén Solano (2012) singles
out three factors in the use of metaphorical expressions in class. First, the need to
take into account whether the metaphorical expression is frequent in all varieties
of Spanish or whether it is specific to one dialectal area. Second, the type of dis-
course in which the metaphorical expression occurs so that it can be contextual-
ised and analysed in relation to style. Finally, the potential morphosyntactic
combinations in which metaphorical expressions occur since sometimes these ex-
pressions cannot be modified. For example, in an expression such as tomar el
pelo (Lit. take the hair) ‘to mock someone’, it is not possible to change the definite
for an indefinite article (*tomar un pelo (Lit. take one hair)) or in the expression
por si las moscas (Lit. for if the flies) ‘just in case’, the noun moscas ‘flies’ must
take the plural form and not the singular one (*por si la mosca (Lit. for if the fly)).
Guillén Solano also identifies the three most successful strategies to favour meta-
phor comprehension: (i) to group phraseological units by their domains; (ii) to
make explicit the relationship between form, meaning, and function; and (iii) to
contextualise metaphorical expressions.
Gómez Vicente (2013) presents a series of pedagogical units to apply meta-
phor to SFL teaching. Her activities include the metalinguistic, the conceptual,
and the linguistic dimensions in pedagogical progression. This means that it is
necessary to reflect on the concept of metaphor, and then to explain the concep-
tual structures that underlie the expressions. Gómez Vicente also argues that it is
crucial to teach the linguistic elements that convey these conceptual schemes.
Among her recommendations, she proposes some complementary techniques to
integrate the teaching of the vocabulary of emotions. One of those recommenda-
tions is to replace random lists of expressions for thematically-grouped sets of
expressions. Following this procedure, students have to search for the motivation
of a given expression with the help of their own intuition and to establish com-
parisons with the students’ own L1.
Masid (2014) suggests choosing the expressions and vocabulary items to be
taught in class in relation to their frequency and productivity. According to this
author, the most productive and general expressions should be selected. They
should be gradually integrated into the teaching activities, until reaching the
 Metaphor and Spanish as a Foreign Language 89

most complex meanings. Masid also recommends introducing metaphor in

class from the beginning of the learning process and in a systematic way.
Similarly, Lantolf and Bobrova (2014) also agree with a gradual introduc-
tion of metaphors in the SFL classroom. Students should be familiarised with
metaphorical structures first and should be aware of the presence of metaphor
in everyday language. After that, teachers should explain its underlying con-
ceptual structure, which can be done by comparing the metaphorical expres-
sions found in Spanish and their counterparts in the learners’ L1.
With the aim of specifying how to gradually introduce metaphors and sys-
tematise their practice in the SFL class, Bandera-Nápoles (2016) recommends
encouraging the student to engage in independent work. This practice needs to
be integrated within the task-based approach, since this methodology is be-
lieved to foster different skills and competences simultaneously. Metaphorical
competence will therefore be developed through associated learning tasks,
which will both consolidate previous knowledge and prepare the student to
deal with new metaphorical content.

3.2 Practical Activities with Metaphor for the SFL Class

Many scholars in the field of SFL acquisition and teaching argue for the positive
aspects of introducing metaphor explicitly in the L2 Spanish classes (Gómez
Vicente 2013; Hijazo-Gascón 2009, 2011; Kaitian 2011; Lamartí 2011; Masid 2013;
Rivera León 2016). In some cases, they also develop specific teaching materials
to develop metaphorical competence and to raise awareness of metaphors.
These pedagogical proposals also provide students with strategies to prepare
them to understand and produce metaphors in Spanish. In order to achieve this
goal, it is necessary for students to be familiarised with conceptual metaphors
and to be able to carry out activities in which they relate these conceptual meta-
phors with their corresponding metaphorical expressions.
Different types of activities are proposed with this goal in mind. For in-
stance, identifying conceptual metaphors and metaphorical expressions in a
text, asking the students to find different metaphorical expressions correspond-
ing to conceptual metaphors in their L1 and L2, and translating the Spanish
metaphorical expressions into the students’ L1. In other cases, activities are
based on the crosslinguistic comparison of metaphorical expressions (Lamartí
2011; Pérez Bernal 2004). This type of activity is argued to help students (i) to
understand the metaphorical expressions, (ii) to be aware of conceptual differ-
ences among languages, and (iii) to achieve a better understanding of the cul-
tural framework in the target language. In this sense, as Encina and Franco
90 Laura Suárez-Campos et al.

(2010) explain, metaphors encapsulate a great amount of cultural content that

should also be explored in the SFL class.
Another crucial factor for the teaching of metaphor is to present metaphori-
cal expressions in real discourse contexts. For instance, Bandera-Nápoles’
(2016) pedagogical proposal focuses on the use of metaphors in real audio-
visual productions. By a series of pre- and post-activities related to a documen-
tary, students are induced to reflect not only on the metaphorical expressions
used in the documentary but also on thematically-related metaphors.
Visual productions are not the only means to introduce the student meta-
phorical expressions. Any kind of material such as images, music, pieces of art
or recreational activities can also be used with the same purpose. Metaphors,
which can be found in all of these contexts, will foster students’ creativity and
analogical reasoning (Rodríguez Santos and Foncubierta 2016). Some pedagogi-
cal proposals also include work with dictionaries (Gutiérrez Pérez 2004), on-
line corpora (Oster 2010), and poetry (Acquaroni 2011).
Another possibility is to work with creative metaphors in Spanish L2.
Acquaroni (2008), for instance, focuses on these metaphors mainly in relation to
creative writing workshops. Students have to use figurative language consciously,
individually or in groups, to create their own texts in Spanish. They also acquire
knowledge of the metaphorical conceptualisation through the analysis of Spanish
literary texts, where they identify and relate conceptual metaphors with their cor-
responding expressions. However, it is important to note that metaphors are not
exclusive to literature and poetry. Metaphor can be found in all kinds of dis-
course, as explained in Section 2.1. in relation to taboo metaphors in colloquial
language.
One of the semantic domains that often turns up in L2 classroom interven-
tions is that of emotions. There are several converging reasons that explain why
this semantic domain is popular in this context. One reason is that emotions tend
to be expressed through metaphor, and therefore, metaphorical expressions for
emotion are easily found. Another reason relates to the students’ own experi-
ence. They need to communicate what they feel in different contexts, and meta-
phors are excellent evaluative mechanisms to encapsulate emotions, attitudes
along with any additional semantic and pragmatic meaning. There are different
pedagogical proposals in relation to certain emotions such as lust (Hijazo-Gascón
2009), pride (Oster 2010), happiness and sadness (Acquaroni 2011), love (Rivera
León 2016), and to feelings in general (Gómez Vicente 2013; Hijazo-Gascón 2011).
Activities and pedagogical proposals to work with metaphors in SFL are
mostly aimed at adult learners at intermediate and high proficiency levels (usu-
ally B2). However, metaphor is neither restricted to adult population nor B2 lev-
els. Studies on metaphorical competence by young learners of L2 English
 Metaphor and Spanish as a Foreign Language 91

(Piquer-Píriz 2008a, 2008b, 2011) and on metaphor use at lower proficiency levels
of the CEFRL (Littlemore et al. 2012, 2013) have shown that metaphor is present
in both child and adult learners at all proficiency levels. These studies in English
make clear the current need for pedagogical proposals to address the use of
metaphors in class with young learners and at beginner levels in SFL. Martínez
Franco (2015) is a pioneering study where metaphor activities related to topics
covered at initial levels of proficiency (daily routines, cooking recipes, describing
the house, talking about the weather, etc.) are developed.

4 Conclusions
This chapter has offered an overview of how Conceptual Metaphor Theory has
been applied to the acquisition and teaching of Spanish as a Foreign Language.
Similar to previous results in EFL, studies on metaphor comprehension in SFL
have revealed the importance to encourage the use of different strategies to
infer meaning from metaphorical expressions, since comprehension depends
on many variables such as the cognitive style of the students, their exposure to
the language, and the conceptual distance between languages. Studies on met-
aphor writing production, on the other hand, have evidenced the need to use
metaphor and, therefore, have argued that it is crucial to make students be
aware of the embodied motivation that grounds the conceptual metaphor, since
it can be used as a compensatory strategy. Research on the implementation of
metaphor in the SFL class has demonstrated that it is beneficial for students to
teach metaphor in class because it improves the students’ metaphorical compe-
tence as well as their communication skills. Despite the importance of meta-
phor in all these SFL contexts, the number of studies devoted to investigate the
impact of metaphor on the acquisition and teaching of SFL is still limited.
There are areas such as the role of metaphor in oral comprehension and pro-
duction or the role of age (young learners) and proficiency level (beyond B2)
that remain yet untouched.
Research reviewed in this chapter (Hijazo-Gascón 2009, 2011; Gómez 2013;
Lantolf and Bobrova 2014; Masid 2014; Acquaroni and Suárez-Campos 2019)
has claimed that a systematic teaching of metaphor will provide students with
a coherent framework to acquire figurative language, avoiding traditional
learning by memorisation. Researchers in this field are increasingly aware of
the potential benefits of introducing the teaching of metaphor right from the
beginning of the learning process. Metaphor helps students developing their
cognitive skills and enhancing their autonomy in their future learning of
92 Laura Suárez-Campos et al.

figurative language. As a result, the number of pedagogical proposals related to

metaphor in SFL is increasing but not quickly enough. At the moment, and
probably (but not only) due to the lack of metaphorical descriptors in the
CEFRL and the CPCI, the body of materials available to systematically teach
metaphor in the SFL classroom is scanty.
In conclusion, conceptual metaphor in SFL is an area of study with an enor-
mous potential, with multiple didactic applications, and, in short, with endless
paths for future research.

References
Acquaroni, Rosana. 2008. La incorporación de la competencia metafórica a la enseñanza-
aprendizaje del español como segunda lengua a través de un taller de escritura creativa:
estudio experimental. Madrid: Universidad Complutense de Madrid dissertation.
Acquaroni, Rosana. 2011. Metáfora y poesía como instrumento para la comunicación
intercultural en el aula de ELE: la conceptualización de la tristeza y de la alegría a través
de un poema de Miguel Hernández. Paper presented at FIAPE IV International Conference,
University of Santiago de Compostela, 17–20 April 2011.
Acquaroni, Rosana & Laura Suárez-Campos. 2019. El desarrollo de la competencia metafórica en
la enseñanza del español LE/L2. In Iraide Ibarretxe-Antuñano, Teresa Cadierno & Alejandro
Castañeda Castro (eds.), Lingüística cognitiva y español LE/L2, 371–391. London:
Routledge.
Azuma, Masumi. 2009. Positive and negative effects of mother-tongue knowledge on the
interpretation of figurative expressions. Papers in Linguistic Science 15. 165–192.
Bandera-Nápoles, Aneyansis. 2016. La metáfora cognitiva y la conversación en la clase de ELE.
[Special issue] Maestro y Sociedad 2. 169–180.
Barcelona, Antonio & Cristina Soriano. 2004. Metaphorical conceptualization in English and
Spanish. European Journal of English Studies 8(3). 295–307.
Boers, Frank. 2000. Metaphor awareness and vocabulary retention. Applied Linguistics 21(4).
553–571.
Boers, Frank. 2006. Expanding learners’ vocabulary through metaphor awareness: What
expansion, what learners, what vocabulary. In Gitte Kristiansen, Michel Achard, René
Dirven & Francisco J. Ruiz de Mendoza Ibáñez (eds.), Cognitive linguistics: Current
applications and future perspectives, 211–232. Berlin: Mouton de Gruyter.
Boers, Frank. 2013. Cognitive linguistics approaches to teaching vocabulary: Assessment and
integration. Language Teaching 46(2). 208–224.
Boers, Frank & Seth Lindstromberg. 2008. Cognitive linguistic approaches to teaching
vocabulary and phraseology. Berlin: Mouton de Gruyter.
Boers, Frank & Jeannette Littlemore. 2000. Cognitive style variables in participants’
explanations of conceptual metaphors. Metaphor and Symbol 15(3). 177–187.
Boers, Frank, Murielle Demecheleer & June Eyckmans. 2004. Cross-cultural variation as a
variable in comprehending and remembering figurative idioms. European Journal of
English Studies 8(3). 375–388.
 Metaphor and Spanish as a Foreign Language 93

Chen, Yi-Chen & Huei-Ling Lai. 2013. The influence of cultural universality and specificity on
EFL learners’ comprehension of metaphor and metonymy. International Journal of Applied
Linguistics 23. 312–336.
Cometa, Michael S. & Morris E. Eson. 1978. Logical operations and metaphor interpretation:
A Piagetian interpretation. Developmental Psychology 49. 649–659.
Danesi, Marcel. 1988. The development of metaphorical competence: A neglected dimension
in second language pedagogy. In Albert N. Mancini, Paolo Giordano & Pier Raimondo
Baldini (eds.), Selected papers from the Proceedings of the Third Annual Conference of
the American Association of Teachers of Italian, 1–10. River Forest, IL: Rosary College.
Danesi, Marcel. 1992. Metaphorical competence in second language acquisition and second
language teaching: the neglected dimension. In James E. Alatis (ed.), Georgetown
University round table on languages and linguistics, 489–500. Washington, D.C.:
Georgetown University Press.
Danesi, Marcel. 2008. Conceptual errors in second-language learning. In Sabine De Knop &
Teun De Rycker (eds.), Cognitive approaches to pedagogical grammar, 231–256. Berlin:
Mouton de Gruyter.
De Cock, Barbara & Ferrán Suñer. 2018. The influence of conceptual differences of processing
taboo metaphors in the foreign language. In Andrea Pizarro Pedraza (ed.), Linguistic
taboo revisited. Novel insights from cognitive perspectives, 201–222. Berlin: Mouton de
Gruyter.
Encina, Pablo & Penélope Franco. 2010. Metáfora y metonimia en las clases de L2 en China.
III Jornadas de formación de profesores de ELE en China. Suplementos SinoEle 3. 1–15.
Ferreira, Luciane. 2008. A psycholinguistic study on metaphor comprehension in a foreign
language. Revista Virtual de Estudos da Linguagem. ReVEL 6(11). 1–23.
García Caballero, Ana. 2017. La importancia del componente metafórico en el aprendizaje de
E/2L. Acercamiento pragmático. RedELE 29. https://www.mecd.gob.es/dam/
jcr:98e9dc4f-06ec-42c0-b660-389a1917bf5c/redele-2017-29-7-garcia-caballero-ana.pdf
(accessed 10 February 2018).
Golden, Anne. 2012. Metaphorical expression in L2 production. The importance of the text
topic in the corpus research. In Fiona MacArthur, José Luis Oncins-Martínez, Manuel
Sánchez-García & Ana María Piquer-Píriz. (eds.), Metaphor in use. Context, culture and
communication, 136–148. Amsterdam: John Benjamins.
Gómez Vicente, Lucía. 2013. L’expression métaphorique de l’évenèment émotionel en français
(L1) et en espagnol (L1/L2): Les images schéma haut/bas et dedans/dehors. Analyse
descriptive et proposition didactique. Recherches en didactique des langues et des
cultures 10(1). 81–102.
Gómez Vicente, Lucía. 2019. La expresión de las emociones en la enseñanza de español LE/
L2. In Iraide Ibarretxe-Antuñano, Teresa Cadierno & Alejandro Castañeda Castro (eds.),
Lingüística cognitiva y español LE/L2, 340–370. London: Routledge.
Guillén Solano, Patricia. 2012. La incorporación del léxico coloquial en la clase de español
como segunda lengua: estrategias para el estudio de metáforas lexicalizadas. Revista
Kariña XXXVI. 93–100.
Gutiérrez Pérez, Regina. 2004. La metáfora en la enseñanza de español como segunda
lengua. In Mª Auxiliadora Castillo Carballo, Olga Cruz Moya, Manuel García Platero &
Juan Pablo Mora Gutiérrez (eds.), Las gramáticas y los diccionarios en la enseñanza del
español como segunda lengua: deseo y realidad. XV Congreso Internacional ASELE,
444–448. Sevilla: Centro Virtual Cervantes.
94 Laura Suárez-Campos et al.

Gutiérrez Pérez, Regina. 2010. Estudio cognitivo-contrastivo de las metáforas del cuerpo.
Análisis empírico del corazón como dominio fuente en inglés, francés, español, alemán e
italiano. Frankfurt: Peter Lang.
Hijazo-Gascón, Alberto. 2009. Cross-cultural differences in the conceptualization and their
application in L2 instruction. In Reyes Gómez Morón, Manuel Padilla Cruz, Lucía
Fernandez Amaya & María de la O Hernández López (eds.), Pragmatics applied to
language teaching and learning, 42–59. Newcastle: Cambridge Scholars.
Hijazo-Gascón, Alberto. 2011. Las metáforas conceptuales como estrategias comunicativas y
de aprendizaje: una aplicación didáctica de la lingüística cognitiva. Hispania 94(1).
142–154.
Johnson, Janice. 1991. Developmental versus language-based factors in metaphor
interpretation. Journal of Educational Psychology 83(4). 470–483.
Johnson, Mark. 1987. The body in the mind. The bodily basis of meaning, reason and
imagination. Chicago: University of Chicago Press.
Kaitian, Lu. 2011. Metáforas en la enseñanza del español a alumnos chinos: propuestas
didácticas para mejorar la conciencia y la competencia metafórica. In Nicolás Arriaga
Agrelo, Rocío Blasco García, Ana Mª Ducasse, Santiago González Fernández-Corugedo,
Armando Mateo Pérez, Francisco Javier Menéndez Sánchez, Alberto José Sánchez Griñán,
Juan Robisco García & José Miguel Blanco (eds.), Competencias y estrategias docentes en
el contexto de Asia-Pacífico. Selección de artículos del II Congreso de Español como
Lengua Extranjera en Asia-Pacífico, 329–346. Manila: Centro Virtual Cervantes.
Kalyuga, Marika & Slava Kalyuga. 2008. Metaphor awareness in teaching vocabulary.
Language Learning Journal 36(2). 249–257.
Kövecses, Zoltán. 2000. Metaphor and emotion. Language, culture and body in human
feeling. Cambridge: Cambridge University Press.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago.
Lamartí, Rachid. 2011. La metáfora y la competencia conceptual en estudiantes sinófonos de
ELE. In Nicolás Arriaga Agrelo, Rocío Blasco García, Ana Mª Ducasse, Santiago González
Fernández-Corugedo, Armando Mateo Pérez, Francisco Javier Menéndez Sánchez, Alberto
José Sánchez Griñán, Juan Robisco García & José Miguel Blanco (eds.), Competencias y
estrategias docentes en el contexto de Asia-Pacífico. Selección de artículos del II
Congreso de Español como Lengua Extranjera en Asia-Pacífico, 170–181. Manila: Centro
Virtual Cervantes.
Lantolf, James & Larysa Bobrova. 2014. Metaphor instruction in the L2 Spanish classroom:
Theoretical argument and pedagogical program. Journal of Spanish Language Teaching
1(1). 46–61.
Lavín Carrera, Elda. 2016. La metáfora conceptual: una estrategia para el aprendizaje de
vocabulario en E/L2 en el nivel B2. Santander: Universidad de Cantabria dissertation.
Littlemore, Jeannette. 2001a. Metaphoric competence: a language learning strength of
students with a holistic cognitive style? TESOL Quarterly 35(3). 459–491.
Littlemore, Jeannette. 2001b. Metaphoric intelligence and foreign language learning.
Humanising Language Teaching 3 (2). http://www.hltmag.co.uk/mar01/mart1.htm
(accessed 3 August 2018).
Littlemore, Jeannette. 2004. Interpreting metaphors in the language classroom. Les Cahiers
de l’APLIUT 23(2). 57–70.
Littlemore, Jeannette. 2008. The relationship of associative thinking, analogical reasoning,
image formation and metaphoric extension strategies. In Mara Sophia Zanotto, Lynne
 Metaphor and Spanish as a Foreign Language 95

Cameron & Marilda C. Cavalcanti (eds.), Confronting metaphor in use: An applied

linguistic approach, 199–222. Amsterdam: John Benjamins.
Littlemore, Jeannette. 2010. Metaphoric competence in the first and second language.
Similarities and differences. In Martin Pütz & Laura Sicola (eds.), Cognitive processing
in second language acquisition: Inside the learner’s mind, 293–316. Amsterdam: John
Benjamins.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sarah Turner. 2012. Investigating
figurative proficiency at different levels of second language writing. Research Notes 47.
14–28.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sarah Turner. 2013. An investigation
into metaphor use at different levels of second language writing. Applied Linguistics 35
(2). 117–144.
Littemore, Jeannette & Graham Low. 2006. Figurative thinking and foreign language learning.
New York: Palgrave Macmillan.
Llopis-García, Reyes. 2015. Las preposiciones y la metáfora del espacio: aportaciones y
potencial de la lingüística cognitiva para su enseñanza. Journal of Spanish Language
Teaching 2(1). 51–68.
López Rodríguez, Irene. 2009. Que no se te vaya la olla. Estudio lingüístico-cognitivo del
campo semántico de la “cabeza”. Tonos digital. Revista electrónica de estudios
filológicos 17. http://www.tonosdigital.es/ojs/index.php/tonos/article/view/306/217
(accessed 5 March 2018).
Low, Graham. 1988. On teaching metaphor. Applied Linguistics 9(2). 125–147.
Maalej, Zouheir A. & Ning Yu (eds.). 2011. Embodiment via body parts: Studies from various
languages and cultures. Amsterdam: John Benjamins.
MacArthur, Fiona. 2010. Metaphorical competence in EFL. Where are we and where should be
going? A view from the language classroom. AILA Review 23(1). 155–173.
MacArthur, Fiona. 2011. On the use of metaphor in office hours’ consultations carried out in
English between lecturers and students with different first languages. International
Journal of Innovation and Leadership in the Teaching of Humanities 1(1). 23–44.
MacArthur, Fiona. 2016a. Beyond engaged listenership: Assessing Spanish undergraduates’
active participation in academic mentoring sessions in English as academic lingua franca.
In Jesús Romero-Trillo (ed.), Yearbook of Corpus Linguistics and Pragmatics 2016. Global
implications for society and education in the networked age, 153–178. Cham: Springer.
MacArthur, Fiona. 2016b. Overt and covert uses of metaphor in the academic mentoring in
English of Spanish undergraduate students at five European universities. Review of
Cognitive Linguistics 41(1). 23–50.
MacArthur, Fiona. 2017. Using metaphor in the teaching of second/foreign languages. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
413–424. London: Routledge.
MacArthur, Fiona, Tina Krennmayr & Jeannette Littlemore. 2015. How basic is “UNDERSTANDING IS
SEEING” when reasoning about knowledge? Asymmetric uses of sight metaphors in office
hours consultations in English as academic lingua franca. Metaphor and Symbol 30(3).
184–217.
MacArthur, Fiona & Ana María Piquer-Píriz. 2007. Staging the introduction of figurative
extensions of familiar vocabulary items in EFL: Some preliminary considerations. Ilha do
Desterro 53. 123–134.
96 Laura Suárez-Campos et al.

Martínez Baztán, A. 2008. Oral evaluation equivalence between the ACTFL guidelines and
some of the CEFR scales. Granada: Universidad de Granada dissertation.
Martínez Franco, Sandra P. 2015. “Hoy mi cerebro no funciona”. El uso de las metáforas en la
enseñanza del español como lengua extranjera. 4º Encuentro Internacional de Español
como Lengua Extranjera. http://www.spanishincolombia.gov.co/
cuartoEncuentroLenguaEspanola/ponencias/Mart%C3%ADnez%20Franco%20Sandra%
20Patricia.pdf (accessed 20 January 2018).
Masid, Ocarina. 2013. La metáfora lingüística en el aula de ELE. In Narciso Contreras Izquierdo
(ed.), La enseñanza del español como LE/L2 en el siglo XXI. XXIV Congreso Internacional
ASELE, 895–903. Jaén: Centro Virtual Cervantes.
Masid, Ocarina. 2014. La metáfora lingüística en el desarrollo de la competencia léxica de ELE.
Propuesta semántica y didáctica sobre el léxico somático desde un punto de vista
cognitivo. Madrid: Universidad Complutense de Madrid dissertation.
Masid, Ocarina. 2017. La metáfora lingüística en español como lengua extranjera (ELE):
estudio pre-experimental en tres niveles de competencia. Porta Linguarum: Revista
Internacional de Didáctica de las Lenguas Extranjeras 27. 155–170.
Mayo Martín, Paula. 2017. Estudio sobre expresiones metafóricas tabú de uso frecuente para
su aplicación a la enseñanza de Español Lengua Extranjera (ELE). E-eleando 5. http://
www.meleuah.es/e-eleando/estudio-sobre-expresiones-metaf%C3%B3ricas-tab%C3%
BA-de-uso-frecuente-para-su-aplicaci%C3%B3n-la (accessed 20 January 2018).
Mendo Murillo, Susana. 2019. El significado de las preposiciones en la enseñanza del español
LE/L2: el caso de por y para. In Iraide Ibarretxe-Antuñano, Teresa Cadierno & Alejandro
Castañeda Castro (eds.), Lingüística cognitiva y español LE/L2, 220–234. London:
Routledge.
Nacey, Susan. 2013. Metaphors in learner English. Amsterdam: John Benjamins.
Nieto, Haydée & Julián Martínez Vázquez. 2014. La metáfora como base de un modelo
didáctico. A propósito de ‘Fundamento teórico de un modelo para trabajar los verbos
haber, ser y estar en el aula de ELE’ de Martha Jurado-Salinas. Revista Nebrija de
Lingüística Aplicada a la Enseñanza de Lenguas 17. 45–78.
Ogarkova, Anna & Cristina Soriano. 2018. Metaphorical and literal profiling in the study of
emotions. Metaphor and Symbol 33(1). 19–35.
Olza, Inés. 2011. On the (meta) pragmatic value of some Spanish idioms based on terms for
body parts. Journal of Pragmatics 43. 3049–3067.
Oster, Ulrike. 2010. Pride-Stolz-orgullo: A corpus-based approach to the expression of emotion
concepts in a foreign language. In Isabel Moskowich, Begoña Crespo, Inés Lareo & Paula
Lojo (eds.), Language widowing through corpora II, 593–610. La Coruña: University of La
Coruña.
Paris, Justin. 2014a. The expression of emotions by second language learners: Metaphor as a
linguistic vehicle. Poznan Studies in Contemporary Linguistics 50(1). 99–121.
Paris, Justin. 2014b. Pour une prise ne compte des transferts de la L1 dans l’apprentissage et
l’enseignement des expressions figuratives d’une L2. Mélanges CRAPEL 35. 135–150.
Pérez Bernal, Marián. 2004. Fraseología y metáfora: materiales para la enseñanza de la
fraseología en una L2. In Mª Auxiliadora Castillo Carballo, Olga Cruz Moya, Manuel
García Platero & Juan Pablo Mora Gutiérrez (eds.), Las gramáticas y los diccionarios en la
enseñanza del español como segunda lengua: deseo y realidad. XV Congreso
Internacional ASELE, 646–654. Sevilla: Centro Virtual Cervantes.
 Metaphor and Spanish as a Foreign Language 97

Piquer-Píriz, Ana M. 2008a. Reasoning figuratively in early EFL: Some implications for the
development of vocabulary. In Frank Boers & Seth Lindstromberg (eds.), Cognitive linguistics
approaches to teaching vocabulary and phraseology, 219–240. Berlin: Mouton de Gruyter.
Piquer-Píriz, Ana M. 2008b. Young learners’ understanding of figurative language. In Mara
Sophia Zanotto, Lynne Cameron & Marilda Cavalcanti (eds.), Confronting metaphor in
use: An applied linguistic approach, 183–198. Amsterdam: John Benjamins.
Piquer-Píriz, Ana M. 2011. Motivated word meanings and vocabulary learning: The polysemy of
hand in the English for young learners’ classroom. Metaphor and the Social World 1(2).
154–173.
Poulisse, Nanda. 1990. The use of compensatory strategies by Dutch learners of English.
Berlin: Mouton de Gruyter.
Real Espinosa, Juan Manuel. 2009. Gramática: la metáfora del espacio. MarcoELE 8. 1–56.
Rivera León, Lorena. 2016. La metáfora como recurso didáctico en el aula de ELE: un estudio a
partir de la lingüística cognitiva. Boletín de la Asociación para la Enseñanza del Español
como Lengua Extranjera (ASELE) 54. 13–46.
Rodríguez Santos, José María & José Manuel Foncubierta. 2016. Metáforas, símbolos y
comunicación en el aula de ELE. Mosaico 34. 22–29.
Ruiz Campillo, José Plácido. 2014. La lógica del espacio. Un mapa operativo del sistema verbal
en español. Journal of Spanish Language Teaching 1(1). 62–85.
Soriano, Cristina. 2003. Some anger metaphors in Spanish and English. A contrastive review.
International Journal of English Studies 3(2). 107–122.
Soriano, Cristina & Javier Valenzuela 2009. Emotion and colour across languages: implicit
associations in Spanish colour terms. Social Science Information 48(3). 421–455.
Speed, Laura J., Carolyn O. Meara, Lila San Roque & Asifa Majid (eds.). 2019. Perception
metaphors. Amsterdam: John Benjamins.
Suárez-Campos, Laura & Alberto Hijazo-Gascón. 2019. La metáfora conceptual y su aplicación
a la enseñanza de español LE/L2. In Iraide Ibarretxe-Antuñano, Teresa Cadierno &
Alejandro Castañeda Castro (eds.), Lingüística cognitiva y español LE/L2, 235–252.
London: Routledge.
Vandergrift, Laurens. No date. Proposal for a Common Framework for Reference for Languages
for Canada. http://elpimplementation.ecml.at/Home/IMPEL/Documents/Canada/
ProposalofaCFRforCanada/tabid/122/language/en-GB/Default.aspx#Chapter6
(accessed 03 April 2018).
Yu, Ning. 1998. The contemporary theory of metaphor: A perspective from Chinese.
Amsterdam: John Benjamin.
Part II. New Empirical Studies
Part II. 1. Learners’ Use and Knowledge of L2
Figurative language
Jeannette Littlemore, Paula Pérez-Sobrino, Nina Julich
and Danny Leung
Is Comfort Purple or Green?
Word-colour Associations in the First
and Second Language
1 Introduction and Background to the Study
Abstract concepts, in particular emotions, are often associated with particular
colours, and many of these associations have a bodily basis. For example, for
many speakers of English, anger is red, jealousy is green and purity is white. It
has been suggested that there is a degree of universal convergence regarding the
associations that people form between abstract concepts, emotions and colours
because they are, at some level, bodily-based. At the same time, it has also been
argued that culture and language play an important role in determining the asso-
ciations (see Kövecses 2005 for further discussion of this issue). However, we
know very little about the extent to which these associations are bodily-based
and how their perceived embodiment relates to their universality.
A number of studies have investigated similarities and differences in the asso-
ciations that people from different cultures and linguistic backgrounds form with
colours, and different reasons have been postulated for these findings. For in-
stance, associations between the colour red and anger have been found in English
(Waggoner and Palermo 1989), Hungarian (Kövecses 2005), Chinese (Chen et al.
2014), Japanese (Matsuki 1995) and Polish (Mikolajczuk 1998). However, in
Chinese, red has additional connotations. The Chinese word for red, ‘hóng’, has a
number of positive meanings, such as ‘hóng shì’ (red event) (wedding), ‘zõu hóng’
(walk red) (become famous), and ‘hóng rén’ (red) (famous person) (Xing 2008).
Other cultures, including English-speaking cultures, have also developed a posi-
tive association towards red as it is a symbol of love. Similar sorts of variation
have been found for the colour blue (Barchard et al. 2017) and white (Xing 2008).

Jeannette Littlemore, English Language and Linguistics, University of Birmingham, West

Midlands, United Kingdom, e-mail: [email protected]
Paula Pérez-Sobrino, Modern Philologies, University of La Rioja, Logroño, Spain,
e-mail: [email protected]
Nina Julich, Institute for British Studies, Leipzig University, Leipzig, Germany,
e-mail: [email protected]
Danny Leung, Applied Language Studies, The Open University of Hong Kong, Hong Kong,
China, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-006
104 Jeannette Littlemore et al.

In this chapter, we explore why it is that some word-colour associations are

more universal than others, and postulate that one explanatory factor might be
the extent to which the associations have a physical basis. Some associations
have a clear physical basis, such as the association of the colour red with
anger, as this reflects a clear physical response to an emotional experience.
Other associations are more easily explained in terms of the interactions that
we have through our bodies with the physical environment. This explains, for
example, the associations that people in Western societies have with the colour
yellow and joy (presumably linked to the warmth of the sun). Associations such
as these might also include the association of WHITE/BRIGHT with GOOD/
MORAL, and BLACK/DARK with BAD/IMMORAL as these are based on experi-
ences of self-efficacy in light vs darkness conditions, and experiences of clean-
liness vs dirt (Sherman and Palermo 1989). Other associations are very difficult
to explain either in terms of the body itself or of its interactions with the envi-
ronment, such as the association that jealousy has with the colour green for
many people living in the Western world.
A second question that is of interest is whether these associations carry over
to a second language. We do not know whether when people learn a second lan-
guage they transfer the word-colour associations from their first language or
whether they adopt the associations that are common in the second language, or
whether they form associations that are a mixture of L1 and L2 associations.
Furthermore, we do not know what specific factors drive language learners to
adopt L2 word colour associations rather than retaining their L1 associations,
even when speaking the L2. Some associations will be acquired through exposure
to the target language, but another way in which they may be acquired is through
exposure to the visual culture, particularly through online environments in which
the language is encountered. As we will see below, one of the findings made in
this study was that Cantonese-speaking people associate the word ‘comfortable’
with the colour green, whereas English-speaking people do not do this. A Google
image search for the Cantonese equivalent of comfortable (自在) results in a
screen full of green images, most of which are to do with nature, so we can see
immediately that the word has very different connotations in Cantonese than it
does in English. If people are exposed to L2 colour contexts on a regular basis
they are arguably more likely to internalise them when speaking the target lan-
guage. One question that has not been addressed is which word-colour associa-
tions are most likely to be acquired by second language learners. One possibility
is that they are more likely to acquire those which have a strong bodily-based mo-
tivation as they make more intuitive sense, whereas those that are more cultur-
ally-based will take longer to acquire. In this study we test this hypothesis,
 Is Comfort Purple or Green? Word-colour Associations 105

investigating whether more bodily-based word-colour associations are (a) more

universal and (b) more likely to be acquired by second language learners.

2 Research Questions and Working Hypotheses

In order to address these questions, we conducted a study that compared the
colours that are associated with a range of emotions and abstract concepts by
participants from two very different cultural backgrounds: English and Cantonese.
The study was designed to answer the following research questions:

RQ 1. To what extent do the colours that are associated with abstract con-
cepts vary cross-culturally between speakers of English and Cantonese?
Hypothesis 1: We expect some colour-word pairs to be shared across
groups, such as “angry” and “red” due to their strong physical basis
but we cannot predict the degree of cross-cultural variation across the
two languages.
RQ 2. Does performing a word-colour association task in one’s second lan-
guage engender a move towards second language-type associations?
Hypothesis 2: We expect the speakers to move from typical L1 word-
colour associations to those of the L2
RQ 3. To what extent does level of agreement both within and across lan-
guages correlate with perceived embodiment?
Hypothesis 3: We expect higher levels of agreement both within and
across languages for associations that have a physical basis.
RQ 4. Are physically-based L2 word-colour associations more likely to be
adopted by L2 speakers than culturally-based ones?
Hypothesis 4: We expect that physically-based associations will be
more likely to be adopted by L2 speakers than culturally-based ones.

3 Methodology
In order to answer the above questions, we conducted a two-part study, which in-
volved the administration of two online Qualtrics questionnaires1 to native

1 All materials including the questionnaires, R-scripts and the raw data are available at
https://osf.io/yrbp3/
106 Jeannette Littlemore et al.

speakers of English and Cantonese and to Cantonese speakers answering in

English. In the first part of the study we sought to answer research questions 1 and
2 by identifying the extent to which the colours that are associated with abstract
concepts and emotions vary cross-culturally between speakers of English and
Cantonese, and the extent to which performing a word-colour association task in
one’s second language engenders second language-type associations. We con-
sulted native speakers of English and Cantonese as well as dictionaries and lan-
guage corpora in an attempt to explain cases of variation. In the second part of the
study, we sought to answer research questions 3 and 4 by identifying the extent to
which the strength of a particular word-colour association both within and across
the two languages correlates with its perceived degree of embodiment and whether
bodily-based L2 word-colour associations are more likely to be adopted than cul-
turally-based ones.

3.1 Part 1: Variation In Word-Colour Associations in L1 and L2

3.1.1 Participants

In order to answer research questions 1 and 2, we administered a survey that

was designed to identify those colours that are most commonly associated with
abstract concepts in English and Cantonese2. We distributed the survey to 420
participants divided into three groups. The first group, “English L1”, consisted
of 99 English participants reading and responding in English (45 female, 54 male).
The second group, “Cantonese L1”, consisted of 195 bilingual Cantonese/English
speakers reading and responding in Cantonese (125 female, 70 male). The third
group, “Cantonese/English L2”, consisted of 126 bilingual Cantonese/English
speakers reading and responding in English (98 female, 28 male). The reason
for including this last group was that we were interested in exploring whether
the responses provided by bilingual English/Cantonese speakers start to ap-
proximate those provided by native speakers of English as an L1. Because the
participants were randomly selected we have no reason to expect that their
levels of English were different, however we acknowledge that our lack of in-
formation about their exact level of English is a potential weakness of the
study. All of the Cantonese participants had had at least twelve years of formal

2 All analyses were performed in R (R Core Team 2017).

 Is Comfort Purple or Green? Word-colour Associations 107

English language education and Hong Kong is a bilingual society.3 The distri-
butions across age groups are shown in Table 1.

Table 1: Age groups and language backgrounds of the participants in part 1 of the study.

Group/questionnaire language – – – – – –

English L      
Cantonese L      
Cantonese/English L      

3.1.2 Materials and stimuli

Prior to the study, we compiled a list of 41 words that people associate with a
particular colour in English. In order to generate this list of words, four inform-
ants, all of whom were native speakers of English, were asked to list as many
words for non-concrete phenomena that they could think of that were associated
with the colours yellow, orange, red, blue, green, purple, brown, black, white
and grey. The group then discussed these associations and those that made
sense to at least three people were retained for the study. Some of the informants
included flavours in their list of associations. After some discussion, we decided
to retain these items as they involve cross-sensory metaphor, are relatively ab-
stract and may be susceptible to cross-cultural variation. For the sake of com-
pleteness, we decided to include four basic flavours (sweet, sour, salty and
bitter) but decided to exclude ‘umami’ as this would be unfamiliar to many of the
English-speaking participants. The resulting list of 41 words consisted of words
for abstract concepts (e.g. intelligence), emotions (e.g. anger) and flavours (e.g.
sweet). The full list is available at https://osf.io/yrbp3/.
In the study, participants were asked to complete an online questionnaire pre-
pared in Qualtrics. This questionnaire contained the aforementioned 41 words
gathered for emotions and abstract concepts in their own language (for the
English and Cantonese L1 groups) and in their L2 (English) for the Cantonese/
English L2 group. These words were translated by a bilingual speaker of English
and Cantonese. Back translations were conducted to confirm that the translations
were appropriate. For each word, participants were asked to select from ten

3 All of the participants in the study were adults which means that they had all received twelve
years of compulsory English Language education in accordance with the country’s educational
regulations.
108 Jeannette Littlemore et al.

different colours (yellow, orange, red, blue, green, purple, brown, black, white,
grey), which they felt it was most strongly associated with. The exact wording was:

You will be shown a number of words, each followed by a list of colours. Please select
which colour you associate most strongly with each word you see.

They were given the option to say “I do not associate this word with a colour” if
they so wished, as well as “I associate this word with a colour that is not listed
here” and “I don’t know”. The sequence of words was randomised for each partici-
pant. The questionnaire was administered via a Snowball sampling technique via
social media and email.

3.1.3 Statistical procedures

In order to test whether the differences in word-colour associations across the

three language groups were significant, we performed a Cochran-Mantel-Haenszel
Test for three-dimensional contingency tables, the three dimensions being abstract
concept / emotion, colour, and the three language groups. The distribution was
significant (M2(480)=33,385, p<.001), indicating that the language groups preferred
different word-colour associations. This analysis was complemented by a within-
group analysis as well as an across-group analysis for each of the 41 words.
Within-group analysis: We performed a chi-squared test and Fisher Exact
Test4 for each language group and analysed the residuals to establish which
associations were most prominent in each language group.5
Across-group analysis: To test whether prominent colour associations differed
across the groups, we performed Fisher Exact Tests comparing the distribution of
colours selected for each word across the three language groups. For this analysis,
we were particularly interested in the behaviour of the Cantonese/English L2
group. We wanted to find out whether this group was more likely to respond in a
similar way to the Cantonese L1 group, thus retaining their native culture’s

4 In the data set, some cells have expected frequencies below 5. For this reason, the p-value was
obtained by performing a Fisher Exact Test. Residuals, however, are based on the results of the
chi-squared test (see footnote 5).
5 The residuals of a chi-squared test indicate how much an observed value in a cell differs
from what would be expected given the overall distribution. The residuals obtained here are
Pearson residuals which are calculated by the following formula: observed frequency minus
expected frequency divided by the square root of the expected frequency. Residuals above 2
and below -2 are usually interpreted as significant. High positive residuals indicate that a col-
our is more frequently associated with a particular emotion than expected given the overall
distribution. High positive residuals thus indicate strong word-colour associations.
 Is Comfort Purple or Green? Word-colour Associations 109

association, or converge with the English L1 group, thereby revealing a move to-
wards the L2 associations in their responses.

3.2 Part 2: The Physical Basis of Word-Colour Associations

3.2.1 Participants

In order to answer research questions 3 and 4 (i.e. to ascertain how ‘bodily-based’

the most prominent word/colour relations for each of the three groups were per-
ceived to be), we compiled and administered a survey to two new groups of partic-
ipants. They were contacted via the same social media and email networks as in
Part 1 of the study and steps were taken to ensure that the people who had partici-
pated in Part 1 of the study did not participate in Part 2. This was done so as to
avoid a familiarity effect. The first group, “English L1”, consisted of 51 English par-
ticipants reading and responding in English (29 female, 19 male, 3 unspecified).
The second group, “Cantonese L1”, consisted of 41 Cantonese speakers reading
and responding in Cantonese (28 female, 12 male, 1 unspecified). With this survey
we aimed to establish whether or not the degree of perceived embodiment sup-
porting a word-colour association related to the strength of that association within
a particular language as well as the tendency of the L2 speakers to adopt the asso-
ciation in their L2. The distributions across age groups are shown in Table 2.

Table 2: Age groups and language backgrounds of participants in

part 2 of the study.

Group – – – – –

English L     
Cantonese L     

3.2.2 Materials

We selected the most frequently associated colour for each of the concepts from
Part 1 in both languages and asked English and Cantonese participants to rate the
associations according to their perceived degree of embodiment. For concepts
where no colour was strongly associated (e.g. for dishonest in English, are shown
in Table 2. We still used the most frequent colour. We excluded ‘Angry is Red’ be-
cause it was used as an introductory item. This yielded 40 items for English and
110 Jeannette Littlemore et al.

Cantonese participants each, which were presented to participants in a rando-

mised order. Participants were asked to rate on a sliding scale ranging from 0
(“very weak”) to 100 (“very strong”) the extent to which they perceived the colour-
word pair to have a physical origin. There was also a possibility to rate the pair as
“not having a physical origin”.

3.2.3 Statistical procedures

We then correlated the perceived embodiment ratings for word-colour associa-

tions with their relative frequency of selection in the first study to see whether
degree of perceived embodiment was related to the popularity of the responses.
Furthermore, for those items where the association in Cantonese and English
were different, we were interested in establishing whether Cantonese speakers
of English as an L2 would be more likely to adopt bodily-based associations
than non-bodily-based ones. In order to do this, we inspected the perceived em-
bodiment ratings in all cases where speakers of the two cultures differed in
their associations (which we termed ‘culture-specific associations’) to establish
whether the associations used by Cantonese/English L2 speakers converged to
the more bodily-based association.

4 Findings
Our findings are discussed in relation to each of the four research questions
listed in Section 2.

4.1 RQ 1. To What Extent Do the Colours That Are Associated

with Abstract Concepts Vary Cross-Culturally between
Speakers of English and Cantonese?
In order to answer this research question, we identified four types of word-
colour association: (1) those words for which the two languages exhibited the
same word-colour association; (2) those where the two languages had the same
main association, but differed in the rest of choices; (3) those where there was
some variation in the order but where the overall choices were broadly similar;
and (4) those words where completely different word-colour associations were
reported. Here we discuss each of these groups in turn.
 Is Comfort Purple or Green? Word-colour Associations 111

Table 3 shows the words where the two languages converged in response
patterns (group 1)6.

Table 3: Cases where Cantonese and English associations were similar.

Word Cantonese English Significance of

difference

Angry red (., %) red (., %) similar (p=.)

Dangerous red (., %) red (., %) different (p<.)

Disgusted brown (., %) no colour (., %) similar (p=.)

no colour (., %) brown (., %)
colour not listed (., %) colour not listed (., %)

Dishonest no colour (., %) no colour (., %) similar (p=.)

don’t know (., %)
grey (., %)

Formal black (., %) black (., %) similar (p=.)

white (., %) white (., %)

Furious red (., %) red (., %) similar (p=.)

Old grey (., %) grey (., %) different (p<.)

brown (., %) brown (., %)

Safe green (., %) green (., %) similar (p=.)

white (., %)
blue (., %)

The associations between the words ‘angry’ and ‘furious’ and the colour red might
be related to the physical reactions that people have to these emotions or experien-
ces, in that they cause blood to rush to the surface of the skin. The motivation for
the association between danger and red relate to the natural world where (for ex-
ample) red-coloured insects and mushrooms being particularly poisonous or it
may derive from the fact that in many cultures the colour red is often used in signs
warning of danger because this colour stands out more than other colours. The

6 Numbers in brackets indicate residuals and percentages. We only report associations with re-
siduals above a value of 2. The higher the residuals, the stronger the association. Percentages
indicate which proportion of the group chose the colour for the given emotion.
7 The last column indicates whether the distributions of selected colours for the emotion were
significantly different or not (i.e. similar) across the different language groups, based on a
Fisher Exact Test for each emotion.
112 Jeannette Littlemore et al.

motivation of the relationship between the word ‘disgust’ and the colour brown
may relate to mud and dirt. With respect to the choice of black for ‘formal’, it can
be argued that it may originate in the wearing of black suits at formal occasions
(which is accepted practice in both Hong Kong and in the UK). The association of
‘old’ with grey may have been due to the fact that colours fade when things grow
old, and people’s hair turns grey with ageing. In Cantonese people are more likely
to talk about hair becoming ‘white’ (rather than grey), which may explain the dif-
ference in English and Cantonese association patterns. Finally, the association of
the word ‘safe’ with the colour green is more difficult to explain. It could be related
to nature or to the cultural convention of green lights meaning that it is safe to go
ahead when driving. It should be noted however that these explanations are highly
speculative, and we address the issue of causal factors in a more systematic way
below.
Table 4 displays the cases where the overall distribution was different but
the first word-colour association was the same (group 2):

Table 4: Cases where the overall distribution was different but the first word-colour
association was the same.

Word Cantonese English Significance of

difference

Bored grey (., %) grey (., %) different (p<.)

brown (., %)

Calm blue (., %) blue (., %) different (p<.)

white (., %) green (., %)

Corrupt black (., %) black (., %) different (p<.)

brown (., %)
grey (., %)

Depressed grey (., %) grey (., %) different (p<.)

blue (., %) black (., %)
purple (., %) blue (., %)

Dirty brown (., %) brown (., %) different (p<.)

black (., %)
colour not listed (., %)

Energetic orange (., %) orange (., %) different (p<.)

yellow (., %) yellow (., %)
green (., %)

Evil black (., %) black (., %) different (p<.)

purple (., %) red (., %)
 Is Comfort Purple or Green? Word-colour Associations 113

Table 4 (continued )

Word Cantonese English Significance of

difference

Extrovert orange (., %) orange (., %) different (p<.)

yellow (., %) don’t know (., %)
don’t know (., %)
green (., %)

Fresh green (., %) green (., %) different (p<.)

white (., %)

Mysterious purple (., %) purple (., %) different (p<.)

black (., %) black (., %)
grey (., %)

New white (., %) white (., %) different (p<.)

green (., %)

Passionate red (., %) red (., %) different (p<.)

Sour yellow (., %) yellow (., %) different (p<.)

orange (., %)

Spontaneous no colour (., %) no colour (., %) different (p<.)

don’t know (., %) orange (., %)
purple (., %)

Sweet colour not listed (., %) colour not listed (., different (p<.)
%)
orange (., %) orange (., %)

Truthful white (., %) white (., %) different (p<.)

no colour (., %) blue (., %)
blue (., %) no colour (., %)
don’t know (., %)

It is not easy to find reasons for all of these cases of variation but it is possible
to identify possible sources in some cases. For example, the fact that ‘depressed’ is
linked to blue in English but to black in Cantonese could be explained by the idio-
matic association between sadness and blue in English (e.g. ‘feeling blue’).
However, the motivation for this association is unknown. A similar hesitation
holds for ‘evil’: the fact that it is more likely to be associated with red in English
than in Cantonese perhaps reflects the fact that this is how the Devil is sometimes
portrayed; but, again, we do not know what factors lie behind this association.
Table 5 shows the words for which the same associations were offered, but
in a different order of preference (group 3).
114 Jeannette Littlemore et al.

Table 5: Cases where the same associations were offered but in a different order of
preference.

Word Cantonese English Significance of

difference

Cheerful orange (., %) yellow (., %) different(p<.)

yellow (., %) orange (., %)
colour not listed (., %)

Excited red (., %) orange (., %) different(p<.)

orange (., %) red (., %)

Happy orange (., %) yellow (., %) different(p<.)

yellow (., %) orange (., %)
red (., %)

Harmonious green (., %) blue (., %) different(p<.)

white (., %) green (., %)
blue (., %) white (., %)
colour not listed (., %)

Joyful orange (., %) yellow (., %) different(p<.)

red (., %) orange (., %)
yellow (., %)

Reliable blue (., %) brown (., %) different(p<.)

no colour (., %) no colour (., %)
brown (., %) blue (., %)

Sad grey (., %) blue (., %) different(p<.)

blue (., %) grey (., %)

Again, whilst we cannot comment on all the patterns in this table, it is inter-
esting to note that both languages associate orange and yellow with the concepts
“happy” and “joyful”, but it is only in Cantonese where both concepts are also as-
sociated with red. This is in line with the culture-specific association of red with
positive things in Chinese culture (Xing 2008). Table 3 also shows a shared connec-
tion between “calm” and green and blue (yet with different degrees of saliency in
each language), perhaps hinting at the common understanding of nature and the
sea as peaceful locations. Finally, it is interesting to note that both languages refer
to blue and grey when asked about “sad”. We have already mentioned that the
choice of blue can be accounted for by the English idiom, but it can also be argued
that this connection now has international reach as it refers to a well-known musi-
cal style. The choice of grey, in turn, might have a greater physical basis, due to
the paleness of the skin in depressed emotional states.
 Is Comfort Purple or Green? Word-colour Associations 115

Table 6 shows the words that were associated with completely different col-
ours (group 4).

Table 6: Cases where the words were associated with completely different colours.

Word Cantonese English Significance of

difference

Bitter brown (., %) yellow (., %) different (p<.)

black (., %) brown (., %)
no colour (., %)
don’t know (., %)

Comfortable green (., %) purple (., %) different (p<.)

blue (., %) green (., %)
white (., %) blue (., %)
purple (., %) brown (., %)

Erotic yellow (., %) red (., %) different (p<.)

purple (., %) purple (., %)
colour not listed (., %)

Fearful black (., %) no colour (., %) different (p<.)

Grown no colour (., %) green (., %) different (p<.)

no colour (., %)
don’t know (., %)

Intelligent blue (., %) no colour (., %) different (p<.)

white (., %)
purple (., %)
no colour (., %)

Jealous red (., %) green (., %) different (p<.)

purple (., %) red (., %)

Salty no colour (., %) white (., %) different (p<.)

brown (., %) no colour (., %)
don’t know (., %)
white (., %)

Shy colour not listed (., %) no colour (., %) different (p<.)
red (., %) don’t know (., %)
no colour (., %)

Stable blue (., %) brown (., %) different (p<.)

blue (., %)
116 Jeannette Littlemore et al.

Cases of clear divergence are somewhat easier to account for. Some can be
explained in very practical terms. For example, for speakers of English, the con-
cept of ‘salty’ is associated with salt but for speakers of Cantonese it is associated
with soy sauce or fish sauce, which are both brown. The same association may
also explain why ‘bitter’ is brown for speakers of Cantonese and yellow for speak-
ers of English; whereas soy sauce is bitter as well as salty, lemons are the most
prototypical bitter fruit for English speakers. An interesting case is “comfort-
able”, which is associated with the colour green in Cantonese but purple in
English. This might be explained by the fact that the word for ‘comfortable’ in
Cantonese connotes nature, countryside and wellbeing, due to the small size of
apartments in Hong Kong. In turn, in English, ‘comfortable’ is perhaps more
likely to be associated with soft furnishing inside the house. The word therefore
appears to have a slightly different meaning in the two languages in that in
English it implies physical contact with an object (such as an armchair or a bed)
whereas in Cantonese it refers to the environment more generally.
We now turn to culture to provide possible explanations for the different col-
our associations for two of the items: ‘jealous’ and ‘erotic’. The fact that the word
‘jealous’ is associated with the colour green in English can be explained by the
idiom ‘green with envy’, which does not exist in Cantonese; this might explain
why Cantonese rely on a more physical word-colour connection (our faces turn red
when we experience a strong emotion such as jealousy). Similarly, the fact that
‘erotic’ is associated with the word ‘yellow’ in Cantonese probably relates to the
fact that pornographic magazines are referred to as ‘yellow magazines’, though the
origin of this term is unclear. English speakers, on their part, rely on red and thus
prime the physical basis of this word colour association. This leaves us with the
following items: ‘fearful’, ‘grown’, ‘intelligent’, ‘shy’ and ‘stable’. These items are
difficult to explain but will be returned to below.
The similarities and differences observed in Tables 1–4 lead us to retain
Hypothesis 1 insofar as some colour-word pairs appeared to be universal across
groups, but there was also a degree of cross-cultural variation across the two
languages that will be explored in more detail in the following sections.

4.2 RQ 2. Does Performing a Word-Colour Association Task in

One’s Second Language Engender a Move
towards Second Language-Type Associations?
In order to answer this research question, we looked only at cases where the
English and Cantonese native speakers’ responses differed significantly from
each other (group 4). When this was the case, we were interested to see whether
 Is Comfort Purple or Green? Word-colour Associations 117

the L2 speakers stuck to the most common L1 associations, according to study

1, or whether they converged towards the L2-type associations. Cases of conver-
gence are highlighted in Table 7 in bold.

Table 7: Word associations provided by L2 speakers shown in comparison with native speaker
associations in the two languages.

Word Cantonese native Bilingual Cantonese/ English ‘Lʹ vs. ‘Lʹ vs.
speakers English speakers native Cantonese English
answering in English speakers
(referred to here as
‘Lʹ)

Bitter brown (., %) brown (., %) yellow (., similar different
%) (p=.) (p<.)
black (., %) black (., %) brown (.,
%)
no colour (.,
%)
don’t know (.,
%)

Comfortable green (., green (., %) purple (., similar different
%) %) (p=.) (p<.)
blue (., %) white (., %) green (.,
%)
white (., %) blue (., %) blue (.,
%)
purple (., %) brown (.,
%)

Erotic yellow (., purple (., %) red (., different different
%) %) (p<.) (p<.)
purple (., %) don’t know (., %) purple (.,
%)
red (., %) colour not
listed (.,
%)
colour not listed (.,
%)
yellow (., %)

Fearful black (., %) black (., %) no colour different different
(., %) (p<.) (p<.)
purple (., %)
no colour (., %)
118 Jeannette Littlemore et al.

Table 7 (continued )

Word Cantonese native Bilingual Cantonese/ English ‘Lʹ vs. ‘Lʹ vs.
speakers English speakers native Cantonese English
answering in English speakers
(referred to here as
‘Lʹ)

Grown no colour (., green (., %) green (., different different
%) %) (p<.) (p<.)
no colour (., %) no colour
(., %)
don’t know (., %) don’t know
(., %)

Intelligent blue (., %) blue (., %) no colour similar different
(., %) (p=.) (p<.)
white (., %) no colour (., %)
purple (., %)
no colour (.,
%)

Jealous red (., %) purple (., %) green (., similar different
%) (p=.) (p<.)
purple (., %) red (., %) red (.,
%)

Salty no colour (., no colour (., %) white (., different similar
%) %) (p<.) (p=.)
brown (., %) white (., %) no colour
(., %)
don’t know (., blue (., %)
%)
white (., %)

Shy colour not listed no colour (., %) no colour different different
(., %) (., %) (p<.) (p<.)
red (., %) colour not listed (., don’t know
%) (., %)
no colour (.,
%)

Stable blue (., %) blue (., %) brown (., similar different
%) (p=.) (p<.)
white (., %) blue (.,
%)
black (., %)
no colour (., %)
 Is Comfort Purple or Green? Word-colour Associations 119

These results show that in most cases, the responses given in the second lan-
guage resemble those that are given in the L1. There was only one case where the
L2 association converged towards English: ‘salty’, where the Cantonese/English
group chose white over brown (which was the most prominent choice of the
English group). The findings for ‘salty’ could be explained by the fact that the
word contains the word ‘salt’ itself, thus leading the speakers to think of salt it-
self in their responses. For this reason, this word might therefore be eliminated
from the analysis, allowing us to conclude that the strongest tendency was to re-
tain the L1 association.
There were however some cases where the associations produced in the L2
clearly diverged from Cantonese. These were: ‘erotic’, ‘fearful’, ‘shy’ and ‘grown’.
In the case of ‘shy’, there was no strong association with any particular colour. In
the case of ‘grown’, the main association is green as was the case in English. Yet,
there was still a significant difference between the overall distribution of colours
associated with ‘grown’ in the L2 and English. This might be interpreted as a weak
form of conversion. ‘Fearful’ behaved in a similar way in that the L2 diverged from
Cantonese (black was less strongly associated) and moved towards the English as-
sociation (where black was not associated at all with fearful). Interestingly, for the
word ‘erotic’, the L2 associations differed significantly from Cantonese but they did
not converge towards the English pattern (red) in any significant way. In these
cases, the L2 speakers appear to have developed a kind of ‘interlanguage’ in terms
of their associations. Consequently, Hypothesis 2 is only partly confirmed: speakers
move from typical L1 word-colour associations to those of the L2, but they also con-
tinue to make use of associations that are used in their L1, as one would expect.
Above we saw that for ‘happy’ and ‘joyful’, associations in English and
Cantonese are fairly similar. However, Cantonese (in contrast to English) also
associates red with these positive concepts. Interestingly, red does not emerge
as a significant association for ‘happy’ and ‘joyful’ in the Cantonese/English
L2 group (see Table 8). This may reflect the cultural rather than bodily-based
origin of the association between red and happy / joyful.
So far in our study, we have managed to explain some of the associations in
terms of practical considerations (e.g. the colour of soy sauce) or idioms whose
etymology is unclear (e.g. ‘green with envy’). However, we saw above that it may
be the case that some word-colour associations both in the L1 and the L2 can be
explained by their levels of perceived embodiment. It is to this issue that we now
turn by looking first at the role of perceived embodiment in shaping L1 associa-
tions (RQ3) and second at its role in shaping L2 behaviour (RQ4).
120 Jeannette Littlemore et al.

Table 8: L2 behaviour for happy and joyful.

Word Cantonese Cantonese L English L vs. L vs.

Cantonese English

happy orange (., %) yellow (., %) yellow (., %) different different
(p<.) (p<.)
yellow (., %) orange (., %) orange (., %)
red (., %)

joyful orange (., %) orange (., %) yellow (., %) different different
(p<.) (p<.)
red (., %) yellow (., %) orange (., %)
yellow (., %)

4.3 RQ 3. To What Extent Does Level of Agreement Both

within and across Languages Correlate with Perceived
Embodiment?
Perceived embodiment ratings and strength of association (operationalised by
how frequently a colour was selected for a given word in the first study) were
highly correlated for both English and Cantonese (for English: t = 10.483, df = 38,
r = .86, p < .001; for Cantonese: t = 5.3159, df = 38, r = .65, p < .001). This finding
suggests that, indeed, strongly associated word-colour pairs are more bodily-
based. We thus accept Hypothesis 3 of the study: the overall trend shows a tight
correlation between the most prominent word colour choices and their perceived
degree of embodiment. Figure 1 (English) and Figure 2 (Cantonese) plot the corre-
lation between the mean perceived embodiment rating for each association and
the strength of the association operationalised by the relative frequency of selec-
tion, i.e. percentage of how often the colour was chosen for the given concept
(please note that for reasons of display the figures do not present the full range
of the rating scale, which was 1 to 100).
It should be noted that the standard deviation for the mean ratings is rather
high. The standard deviation indicates how strongly individual ratings for an as-
sociation differ from the mean rating for the association. If the standard deviation
is small, the mean is a good summary for the distribution of ratings for a particu-
lar association. The higher the standard deviation, however, the more the data
points vary within the distribution. For the English perceived embodiment ratings,
the standard deviation on average was 27.8. The smallest standard deviation was
 Dangerous_red
80
Fresh_green
Passionate_red
Furious_red
Old_grey Dirty_brown
Formal_black
Erotic_red
Grown_green Calm_blue Cheerful_yellow
Depressed_grey Happy_yellow
Joyful_yellow

60 New_white
Salty_white
Bored_grey
Truthful_white Jealous_green
Sour_yellow Sad_blue
Harmonious_blue
Sweet_orange Safe_greenEvil_black

Fearful_black Energetic_orange

Perceived embodiment rating

Corrupt_black
Excited_orange
Intelligent_blue
Extrovert_orange
40 Stable_brown
Bitter_yellow
Dishonest_black Mysterious_purple
Disgusted_brown Reliable_brown
Spontaneous_orange
Shy_blue
Comfortable_purple
Is Comfort Purple or Green? Word-colour Associations

25 50 75
Strength of association (% frequency of selection)

Figure 1: Perceived degree of embodiment for frequently associated word-colour pairs (English).
121
 80
122

Dangerous_red

Furious_red
New_white

Fresh_green
Formal_black
Comfortable_green Calm_blue
Safe_green
60 Depressed_grey Fearful_black
Sad_grey Excited_red
Bored_grey
Harmonious_green Evil_black
Stable_blue
Truthful_white Sour_yellow
Energetic_orange
Jeannette Littlemore et al.

Happy_orangeMysterious_purple Jealous_red Old_grey

Reliable_blue
Happy_orange Cheerful_orange
Extrovert_orange
Dirty_brown Erotic_yellow
Joyful_orange
Bitter_brown

Perceived embodiment rating

Disgusted_brown Corrupt_black
40
Intelligent_blue Shy_red
Passionate_red
Sweet_orange
Dishonest_grey

Spontaneous_green
Grown_blue
Salty_brown

25 50 75 100
Strength of association (% frequency of selection)

Figure 2: Perceived degree of embodiment for frequently associated word-colour pairs (Cantonese).
 Is Comfort Purple or Green? Word-colour Associations 123

found for ‘fresh is green’ (SD = 21, variation coefficient 0.27).8 Thus, this pair ex-
hibited the least amount of variation among participants. The largest standard de-
viation was found for ‘shy is blue’ (SD = 29, variation coefficient = 0.91). Thus,
this pair exhibited the highest amount of variation, or divergence, among raters.
For the Cantonese perceived embodiment ratings, the standard deviation
was equally high. On average, the standard deviation was 27.3, indicating that
the individual ratings varied strongly from the mean for a particular associa-
tion. ‘Bitter is brown’ showed least variation (variation coefficient = 0.32) and
‘salty is white’ showed most variation (variation coefficient = 1.54).
Furthermore, it should be noted that when rating the degree of bodily-based
motivation for a given word-colour pair, participants also had the option to select
“don’t associate”. It turns out that for some word-colour pairs participants selected
this option quite often (e.g. “don’t associate” was selected 61% of the times for
Jealous is Green in the Cantonese data and even 66.7% for Comfortable is Purple
in the English data).9 The frequency with which this option was selected for a
given pair is negatively correlated both with strength of association (t = -6.3435,
df = 55, p<.001, r = -.65 for the Cantonese data; t = -8.6991, df = 38, p<.001, r = -.82
for the English data) as well as perceived degree of bodily-based motivation
(t = -13.045, df = 55, p<.001, r = -.87 for the Cantonese data; t = -13.464, df = 38,
p<0.001, r = -.91 for the English data). This means that participants selected “don’t
associate” for pairs that were not perceived as being bodily-based and which were
not among the most frequent pairings identified in the first study. This finding pro-
vides further support for the hypothesis that bodily-based word-colour associations
are more universal. We can give two potential explanations for the frequent selec-
tion of the “don’t associate” option: First, the association might be more cultural
than bodily-based, thus “don’t associate” was selected; and, second, in the first
study, participants may have been pushed to select a colour because there was no
“don’t associate” option which led to the higher number of “don’t associate” an-
swers in the second study for these word-colour pairs.
We also analysed whether perceived embodiment ratings differed signifi-
cantly between English and Cantonese (Figure 3). Overall, perceived embodi-
ment ratings for shared associations between Cantonese and English (N = 23)
were correlated (t = 3.1079, df = 21, p<.01, r = .56). There were only three cases
where the mean ratings differed significantly (circled in Figure 3): (1) ‘dirty is
brown’ (t = -4.0264, df = 76.89, p<.001, two-sided, independent samples), (2)

8 In order to compare standard deviations, they were normalised by calculating the variation
coefficient (Gries 2013: 125).
9 Figures regarding the frequency of the selection of the “don’t associate” option can be ac-
cessed via the raw data in the online repository.
 124

80
Dangerous_red

Furious_red
New_white
70
Fresh_green
Formal_black

Calm_blue
Safe_green Depressed_grey
Fearful_black
60
Jeannette Littlemore et al.

Bored_grey
Evil_black
Energetic_orange Truthful_white
Sour_yellow Old_grey
50 Mysterious_purple

Mean embodiment ratings Cantonese

Extrovert_orange Dirty_brown

Corrupt_black
40 Disgusted_brown Intelligent_blue
Passionate_red
Sweet_orange

40 50 60 70 80
Mean embodiment ratings English

Figure 3: Comparison of perceived embodiment ratings for associations that are shared between English and Cantonese.
 Is Comfort Purple or Green? Word-colour Associations 125

‘old is grey’ (t = -3.3371, df = 76.994, p<.01, two-sided, independent samples),

and (3) ‘passionate is red’ (t = -5.8091, df = 38.72, p<.001, two-sided, indepen-
dent samples). Our first study showed that in Cantonese, both brown as well as
black are almost equally strongly associated with dirty (cf. Table 2). It might be
because Cantonese has these two competing colours that dirty and brown were
perceived as less bodily-based. With respect to the association between ‘old’
and ‘grey’, this was significantly less likely to be perceived as bodily-based in
Cantonese than in English. The mean degree of perceived embodiment for ‘old is
grey’ was 51 in Cantonese, and 70 in English. However, the perceived embodiment
rating for ‘old’ and ‘grey’ in Cantonese was still relatively high. Finally, although
‘passionate’ and ‘red’ were strongly associated in Cantonese (actually more so
than in English, see Table 2), they were perceived as less bodily-based compared
to English. The exceptional character of this word pairing is also clearly notice-
able in Figure 3. The reason why the association between ‘passionate’ and ‘red’
received such a low degree of perceived embodiment in Cantonese in contrast to
its high strength of association requires further research.
In general, this second finding supports the hypothesis that universally shared
associations also tend to be more bodily-based, thus confirming Hypothesis 3.

4.4 RQ4. Are Bodily-Based L2 Word-Colour Associations More

Likely to Be Acquired than Culturally-Based Ones?
In order to find out whether L2 behaviour was driven by degree of perceived em-
bodiment, we considered cases where English and Cantonese differed in their as-
sociations (Table 5) and compared these to the perceived embodiment ratings in
English and Cantonese. The respective cases were colour associations for ‘bitter’,
‘comfortable’, ‘erotic’, ‘fearful’, ‘grown’, ‘intelligent’, ‘jealous’, ‘salty’, ‘shy’, and
‘stable’. See below Figure 4 for exclusive culture-specific associations in English.
We can see that ‘grown is green’ and ‘salty is white’ were actually more
bodily-based than expected given their frequency of selection. It was for these
two pairs that the Cantonese / English L2 group converged towards English.
However, even though ‘fearful’ and ‘black’ were not significantly associated in
English in the first study (cf. Table 5), the association was still perceived as hav-
ing a relatively high physical basis, as can be seen in Figure 4. The reason why
‘fearful’ and ‘black’ were not significantly associated with one another even
though there was a great degree of perceived embodiment supporting that con-
nection requires further research. The same applies to ‘intelligent is blue’, an
association which was reported as having a fairly strong physical basis, but
which was not found to occur with a high frequency in the first questionnaire.
126 Jeannette Littlemore et al.

80 70

Erotic_red
Perceived embodiment rating

Grown_green
60

Sallty_white

Jealous_green
50

Fearful_black

Intelligent_blue Stable_brown
40

Bitter_yellow

Shy_blue
30

Comfortable_purple

0 20 40 60 80
Strength of association (% frequency of selection)

Figure 4: Degree of perceived embodiment for culture-specific associations (English).

The perceived embodiment ratings for associations for ‘jealous’, ‘stable’, ‘bit-
ter’, ‘shy’ and ‘comfortable’ were lower than expected given their relatively high
frequency of selection. This may explain why we found that the Cantonese /
English L2 group did not converge to these specific associations (cf. Table 5).
We now turn our attention to the culture-specific associations for Cantonese
(see Figure 5 below). Interestingly, these findings mirror the trends observed for
English above. ‘Grown is blue’ and ‘salty is brown’ received very low perceived
embodiment ratings, and it was for these associations where the Cantonese /
English L2 group converged to English. In turn, ‘bitter’, ‘comfortable’, ‘fearful’,
‘jealous’, and ‘stable’ received bodily-basis ratings in line with the ratings for the
strength of the association (or were even higher), and thus it was in these cases
where the Cantonese / English L2 group retained the Cantonese associations.
Two interesting cases that deserve further consideration are ‘erotic’ and ‘shy’.
Recall that, for ‘erotic’, we saw in Table 5 that the Cantonese / English L2 group
diverged from the Cantonese associations. This is supported by the fact that
‘erotic is yellow’ received a comparably low bodily-basis rating. Interestingly, the
L2 group still did not converge to the English association, and remained some-
where in between the native and the target language. For ‘shy’, we can see that
in terms of bodily-basis the respective associations received relatively low bodily-
basis ratings in both English and Cantonese. This may explain why the L2 did not
exhibit any strong preference for shy with any colour.
 Is Comfort Purple or Green? Word-colour Associations 127

80 70
Perceived embodiment rating

Comfortable_green
Fearful_black
60

Stable_blue Jealous_red
50

Erotic_yellow
Bitter_brown
Intelligent_blue
40

Shy_red
30

Grown_blue
Salty_brown

0 20 40 60 80
Strength of association (% frequency of selection)

Figure 5: Perceived degree of bodily-basis for culture-specific associations (Cantonese).

Overall, these findings suggest that associations that are formed in the L2
tend to converge to the more bodily-based associations that are present either in
the L1 or the L2, much in line with our expectations in Hypothesis 4. They also
suggest that the less bodily-based an association is in the L1, the less likely it is
that a person will remain ‘loyal’ to that association when they are using their L2.

5 Conclusion
In this chapter we have provided an overview of the most prominent word-colour
choices in English and Cantonese. Our data reveals that there is a great degree of
convergence between the languages, but more interestingly, a myriad of diverg-
ing choices, the study of which has helped to further our knowledge of these two
cultures. Whereas some of the diverging choices can be accounted for in terms of
cultural differences, the findings from our study suggest that the level of per-
ceived embodiment in the motivation for a particular word-colour association
plays a crucial role in determining the level of agreement both within and across
languages, as well as the extent to which the association is likely to be adopted
by an L2 speaker of the language. These findings provide empirical evidence for
128 Jeannette Littlemore et al.

the role of perceived embodiment in motivating linguistic and conceptual associ-

ations in both the first and second language.
More importantly, our findings have implications for cross-cultural commu-
nication, and by extension, the ways in which language teachers deal with im-
plicit meaning, as they suggest that bodily-based associations that have been
established in the L1 are more likely to be entrenched than culturally-based asso-
ciations, and are therefore likely to be difficult to modify when speaking the L2.
It would be interesting to investigate whether this pattern also extends to other
forms of bodily-based language, such as collocations that are motivated by
bodily-based metaphor. For example, Akpinar and Berger (2015) explored the de-
velopment of collocations over the course of 200 years in a corpus of 5 million
books and found that metaphoric collocations that reflect an underlying sensory
relationship are significantly more likely to remain in the language than their
non-sensory counterparts. Their study revealed that sensory metaphoric colloca-
tions were thus ‘more culturally successful’ than their non-sensory counterparts.
For example, the term ‘sharp increase’ was more likely to remain in the language
once it had entered it than the ‘severe increase’ and the term ‘bright future’ was
more likely to remain than the term ‘promising future’. Akpinar and Berger also
tested 365 participants on their ability to recall the different terms and found that
they were significantly better at recalling the bodily-based metaphorical colloca-
tions than their literal equivalents. If these collocations behave in the same way
as the bodily-based word-colour associations in our study, then one would ex-
pect them to be more entrenched and therefore less resistant to change in cases
where the L2 uses a different collocation.
However, a note of caution needs to be sounded as our study was based on
correlational analyses and we cannot therefore be sure about the role of causal-
ity. Although we have assumed that degree of perceived embodiment explains
the frequency of association, it could be that the relationship works in the oppo-
site direction (i.e. where frequent associations lead one to assume that the rela-
tionship is bodily-based) or it could be that both frequency of association and
perceived embodiment are influenced by a third, as yet unidentified variable.
Other aspects of our data which merit further investigation are those cases
where perceived embodiment and strength of association diverge. In other
words, it would be useful to conduct an investigation (possibly making use of
qualitative methods) into all the cases that appear far from the correlation line
in Figures 1 and 2. For example, it would be interesting to explore the idea that
‘passionate is red’ in Cantonese. This word pair exhibited a strong association
in study 1 but received low perceived embodiment scores in study 2. Similarly,
the idea that ‘fearful is black’ was rated as being relatively bodily-based in
English and Cantonese but in English it was not at all associated, and in the L2
 Is Comfort Purple or Green? Word-colour Associations 129

it was less strongly associated than in Cantonese. Findings such as these are
difficult to explain without recourse to interview data.
Finally, other questions that could usefully be explored relate to the range of
associations that people make in their first and second language, the reasons that
they provide for making these associations, and the role of the participants’ idiom
knowledge in the first or second language in shaping their responses. In our
study, we did not allow for the fact that participants may associate more than one
colour with each of the prompts. This is particularly relevant for bilinguals (all of
the participants in our study were bilingual speakers of English and Cantonese),
who may produce more associations than monolinguals, due to their increased
cognitive flexibility and associative networks (Bialystok 2001a, 2001b; Bialystok
et al. 2014). Furthermore, we did not ask the participants to provide explanations
for their answers. We therefore do not know what motivated their choices.
Another consideration is the fact that some L2-type (English) associations may
have appeared when participants were performing the task in their first language
(Cantonese). More pronounced differences might be found if the study were repli-
cated with inclusion of respondents who were monolingual speakers of a language
distant from English. And finally, although we hypothesised that knowledge of idi-
oms may have affected responses to some of the items, we did not explore this
systematically. The role played by their explicit knowledge of idioms could be ex-
plored through a qualitative examination of the reasons given for their choices
and the role played by their implicit knowledge of idioms could be examined
through corpus-based frequency studies of the colour-based idioms in each of the
two languages. We now intend to conduct a follow-up study in which participants
are permitted to provide multiple associations and to motivate their responses.
This will be accompanied by a corpus-based analysis of colours and their associ-
ated idioms in English and Cantonese, whose findings we will use to help explain
some of the variation.

References
Akpinar, Ezgi & Jonah Berger. 2015. Drivers of cultural success: The case of sensory
metaphors. Journal of Personality and Social Psychology 109(1). 20–34.
Barchard, Kimberley, Kelly Grob, & Matthew Roe. 2017. Is sadness blue? The problem of using
figurative language for emotions on psychological tests, Behavior Research Methods 49
(2).443–456.
Bialystok, Ellen. 2001a. Bilingualism in development, language, literacy and cognition.
Cambridge: Cambridge University Press.
Bialystok, Ellen. 2001b. Metalinguistic aspects of bilingual processing. Annual Review of
Applied Linguistics 21. 139–181.
130 Jeannette Littlemore et al.

Bialystok, Ellen, Fergus Craik, & Gigi Luk. 2014. Bilingualism: Consequences for mind and
brain, Trends in Cognitive Science 16. 240–250.
Chen, J., Natalie Kacinik, Yingjun Chen, & Nianyang Wu, N. (2014). Metaphorical color
representations of emotional concepts in English and Chinese speakers: Evidence from
behavioral data. Poster session presented at the Society for Personality and Social
Psychology conference, Long Beach, CA.
Gries, Stefan. 2013. Statistics for linguistics with R. A practical introduction. 2nd edition.
Berlin, Boston: Walter De Gruyter.
Kövecses, Zoltan. 2005. Metaphor in culture: Universality and variation. Cambridge, UK:
Cambridge University Press.
Kövecses, Zoltan. 2015. Where metaphors come from: Reconsidering context in metaphor.
Oxford: Oxford University Press.
Matsuki, Keiko. 1995. Metaphors of anger in Japanese. In John R. Taylor & Robert E. MacLaury
(Eds.), Language and the cognitive construal of the world, 137–151. Berlin: de Gruyter.
Mikolajczuk, Agnieszka. 1998. The metonymic and metaphoric conceptualization of ‘anger’ in
Polish. In Angeliki Athanasjadou & Elzbieta Tabakowska (Eds.), Speaking of emotions:
Conceptualization and expression, 153–191. Berlin: Mouton.
R Core Team 2017. R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Sherman, Gary & Gerald Clore. 2009. The color of sin: White and black are perceptual symbols
of moral purity and pollution, Psychological Science 20(8). 1019–1025.
Waggoner, John & David Palermo. 1989. Betty is a bouncing bubble: Children’s
comprehension of emotion-descriptive metaphors. Developmental Psychology 25.
152–163.
Xing, Zhiqun 2008. Semantics and pragmatics of color terms in Chinese. Journal Of Historical
Pragmatics 9(2). 315–319.
Gill Philip
Metaphorical Reasoning in Comprehension
and Translation: An Analysis of Metaphor
in Multiple Translations
1 Introduction
Metaphor is routinely said to present a problem in translation (or rather, for trans-
lators; see Philip 2016). The evidence to support this hypothesis, however, is pat-
chy. Over recent decades, several scholars have proposed guidelines on how to
tackle the translation problems that metaphor poses, Newmark’s (1981) set of pro-
cedures being probably the best known (see also van den Broeck 1981; Dobrzyńska
1995; Dickins 2005; Prandi 2010). Such guidelines may suggest ways in which a
translator can deal with metaphor, but remain primarily theoretical. Empirical stu-
dies of how translators actually deal with metaphors draw on these, sometimes re-
fining them significantly in the light of their analyses (e.g. Papadoudi 2010;
Shuttleworth 2017). Yet all deal with published or otherwise officially-sanctioned
translations. By the time a translation has reached its definitive version, the stra-
tegies used by a translator, and the reasons for doing so, remain opaque. Trans-
lation process research offers some insights (see Schäffner and Shuttleworth
2013); equally, we may infer that time constraints or language transfer have in-
fluenced translation choices. Ultimately, however, “we cannot retrace the actual
pathways of the translator’s decision-making procedures” (Schäffner 2017: 257).
Studying the work of trainee translators offers a different perspective. While it
remains largely impossible to investigate the decision-making process of each indi-
vidual, the analysis of multiple translations of a given source text offers ample evi-
dence of the strategies that are actually used in translation. It additionally allows
researchers to incorporate aspects of language proficiency into their analysis, a
matter largely taken for granted with professional translations. Both of these issues
are discussed in this chapter, which presents initial findings from ongoing re-
search into trainees’ translations, with a particular emphasis on the translation of
figurative language. Section 2 offers an overview of existing research in the field,
and is followed in section 3 with a description of the data and methods used. The
remainder of the chapter is dedicated to matters arising in the translations. Sec-
tion 4 deals with metaphor comprehension and the role of context in determining

Gill Philip, Department of Humanities, University of Macerata, Macerata, Italy

e-mail: [email protected]

https://doi.org/10.1515/9783110630367-007
132 Gill Philip

translation choices. Section 5 investigates the difficulties posed by creative variants

of a conventional metaphor. Finally, Section 6 looks at the interplay between
source language proficiency and translation in its examination of a short extract
involving metonymy and irony.

2 Student Translation Research

The present research is positioned within a relatively under-explored area of trans-
lation studies, one in which novice translator activity is examined by comparing
multiple translations. The first attempts to investigate learners’ translations began
around twenty years ago, Bowker and Bennison’s (2003) Student Translation Ar-
chive (STA) probably being the first systematic attempt to compile a computer-
readable corpus of learners’ translations. The MeLLANGE project (Kübler 2008,
Castagnoli et al. 2011), drew on this work, expanding the number of languages and
including the source texts and some professional translations. An error-tagging in-
ventory was also developed, which forms the backbone of the nascent Multilingual
Student Translation project (MUST, Granger and Lefer 2017), which also features a
new query interface designed for the viewing of multiple translations and the origi-
nal source text.
Other learner translation corpora initiatives have focused on L2 competence
rather than translation competence per se. These include the Polish-English PEL-
CRA (Uzar and Waliński 2001) and the Norwegian-English NEST (Graedler 2013),
which comprise translations into the L2 as a means of examining L2 production.
Nacey (2017) investigated aspects of metaphor use in NEST, matching translated
terms to Newmark’s (1981) guidelines for the translation of metaphor and other fi-
gurative language. She observes that although different metaphor types seem to
incur different strategies, students overwhelmingly prefer to translate (L1) source-
language metaphors as metaphors in the (L2) target text rather than paraphrasing
or deleting them. However, she does not discuss the quality of the translations, the
role of source text interference or the degree to which L2 proficiency per se might
affect translation choices.
Nacey’s (2017) investigation has some points of contact with research conduc-
ted into learners’ production of figurative language in the L2. Some general trends
have emerged and can be summarised as follows. First of all, Nacey (2013) has
found that metaphor is at least as frequent in learner writing as it is in native-
speaker written production. This is remarkable since many scholars have noted
that learners often appear to lack the language awareness necessary to recognize
figurative meanings in text (Littlemore 2009) and tend to stick to the core literal
 Metaphorical Reasoning in Comprehension and Translation 133

meaning of familiar words (Danesi 1994; Littlemore and Low 2006; Littlemore
2009). It is common to find calquing of metaphorical words and phrases (Philip
2006, 2010, 2011a), suggesting that learners assume a connection between word
forms and meanings that transcends individual languages. While it would be sim-
plistic to lay the blame entirely on syllabus design, there is a distinct lack of atten-
tion paid to metaphorical meaning in the language learning process (see overview
in Nacey 2013: 43–61). As a result, learners are ill-equipped to cope with most of
the non-literal meanings that they encounter in their foreign language study, with
the possible exception of idioms which, despite their apparent prominence, are ac-
tually low-frequency vocabulary items (Moon 1998) and are not typical of figura-
tive language in general (Philip 2011b). This chapter intends to investigate whether
any of these observations hold true for metaphor comprehension and production
in translation.

3 Data and Methods

The data used in this study comprises multiple translations into Italian (L1) of a
range of English source texts. The translations were produced by two consecutive
cohorts of students studying English and another modern language to advanced
(C1–C2) level (mean age 25, native language Italian). The translations were part
of a final (5th) year module on metaphor in translation in the Modern Languages
and Literatures department of an Italian university, and for this reason the
source texts were metaphorically rich, featuring conventional linguistic meta-
phors, metonymies, metaphorical idioms, analogies, as well as exploitations and
creative variants of conventional forms, metaphor clustering, and conceptual me-
taphor. The initial selection of the source texts was based on their metaphorical
content, using an adapted version of MIPVU (Steen et al. 2010) to identify meta-
phors and other figurative expressions. This analysis informed the sequencing of
the texts during the module. Thus texts whose use of metaphor was limited to
conventional linguistic and conceptual metaphors were introduced before texts
containing creative variants, metaphor clusters and text-structuring metaphor.
Corpus studies of translated texts typically rely on parallel concordances,
in which the source text and the target text are aligned at sentence level in
order to facilitate querying. By querying a word in the source text, it is possible
to view the translation in the equivalent area of the target text. Commercially-
available parallel concordancers cannot process multiple target texts, and al-
though some proprietary concordancers can do so, they are restricted to use
with corpora which have already been processed and uploaded to the remote
134 Gill Philip

server.1 The current research used AntConc (Anthony 2017) to display the concor-
dances via carefully-chosen search terms. These were not hypothetical transla-
tions of the source text metaphors, which are expected to vary from translation
to translation. Instead, they were unequivocal translation equivalents (e.g. nume-
rals, proper names, technical terms, or words/phrases with only one possible
translation) lying in close proximity to the metaphors of interest. By adjusting
the viewing settings to show as much context left and right as the software per-
mits, the translations of the metaphors can be viewed by scrolling left or right;
they can also be sorted alphabetically to bring out the patterns that emerge. Al-
though the output is not perfectly aligned, it is more than satisfactory for class-
room purposes. For more detailed examination, the concordances are extracted
and manually aligned, as will be seen below.

4 Comprehension, Context and Translation

Choices
Mid-way through Religion for Atheists, De Botton deals with the topic of ‘Tender-
ness’. He builds up his argument through a metaphorical framing of Catholicism
and Atheism which casts each in stark contrast to the other. The Catholic world-
view attracts metaphors of warmth, softness, comfort (particularly with reference
to the maternal figure of Mary) as well as negative references to childishness and
emotional display; the atheist standpoint is characterised though metaphors of
coldness, hardness, sharpness, as well as positively-framed maturity and rationa-
lity. This text was presented to cohort 1 in the final lesson, who collectively pre-
pared a definitive version. Cohort 2 tackled the text at an earlier stage (lesson 6).
As a complex persuasive text, many metaphors are present, often occurring
in brief bursts. These further reinforce the overriding metaphors mentioned
above, and their translation requires not just surface comprehension but also
an appreciation of the evaluative stance being conveyed. Here, we will examine
the first sentence of paragraph two.

At its most withering and intellectually pugnacious, atheism has attacked religion for
blinding itself to its own motives, for being unwilling to acknowledge that it is, at base,
nothing more than a glorified response to childhood longings (. . .)
(De Botton 2012: 173, emphasis added)

1 Both NEST (http://clu.uni.no/humfak/nest/index-e.html) and MUST (registered users only)

feature such concordancers.
 Metaphorical Reasoning in Comprehension and Translation 135

At first blush, the highlighted (italicised) metaphors might be viewed as a sort of

mixed metaphor, drawing on three different conceptual source domains. On clo-
ser inspection, they can be seen to function together, building up and reinforcing
the argument that atheist criticism of Catholicism is powerful and destructive.
Withering inverts the polarity of the LIFE IS GROWTH metaphor to imply deterio-
ration with little hope of survival. Pugnacious (from the Latin word for ‘fist’,
‘punch’) and attacked appear to exemplify ARGUMENT IS WAR, while blinding
suggests an extreme interpretation of PHYSICAL INJURY AS IMPEDIMENT, the
injury and impediment in this case being permanent and irreversible.
The question arises, with regard to students’ translations, as to whether or
not this general impression of powerful destructive argument is appreciated
(when reading the source text) and adequately conveyed (in the translations). By
this stage of the module, students have acquired an initial grounding in the diffe-
rence between linguistic and conceptual metaphor and are aware that metaphor
is not only a literary trope. However, they are only beginning to become familiar
with the general principles of the MIP metaphor identification procedure (Praggle-
jaz group 2007) and its refinement in MIPVU (Steen et al. 2010), and with the dif-
ferent classifications of metaphor in terms of vividness or figurative intensity
(Goatly 2007, Hanks 2006). Specific guidelines on how to deal with metaphor in
translation, particularly with regard to their deliberate or non-deliberate use
(Steen 2014), their conflictual or consistent textual meanings (Prandi 2010), and
their power to evoke sensory experience including imagery (Newmark 1981, Shutt-
leworth 2014, 2017), have yet to be introduced.
Tables 1a–c show the concordances for the first, second, and third parts of
the sentence respectively, with the translations of the metaphors highlighted in
bold face (duplicates have been removed).

Table 1a: Translations of “At its most withering and intellectually pugnacious,”.

S Al più sferzante ed intellettualmente combattivo, . . .

S Al suo massimo feroce e intellettualmente battagliero, . . .
S Nella veste più spaventosa e intellettualmente combattiva, . . .
S Al suo più spaventoso e intellettualmente pugnante, . . .
S All’apice della propria ferocia e intellettualmente aggressivo, . . .
S essendo per la maggior parte critico e intellettualmente aggressivo, . . .
S Nel suo più profondo, e intellettualmente combattivo, . . .
S Al suo massimo declino e intellettualmente aggressivo, . . .
S All’apice della sua decadenza e aggressività intellettuale, . . .
S Al suo massimo avvizzimento e intellettualmente combattivo, . . .
136 Gill Philip

Table 1b: Translations of “atheism has attacked religion . . . ”.

S, S, S, S, S, S . . . l’ateismo ha attaccato la religione . . .

S . . . l’ateismo ha criticato la religione . . .
S, S . . . l’ateismo ha aggredito la religione . . .
S . . . l’ateismo ha accusato la religione . . .

Table 1c: Translations of “ . . . for blinding itself to its own motives”.

S . . . per rendere sé stessa cieca ai suoi motivi . . .

S . . . per eclissarsi alle sue cause . . .
S . . . per accecare le proprie ragioni . . .
S . . . per il legame stesso alle sue motivazioni . . .
S . . . poiché questa nasconde a sé stessa i suoi argomenti . . .
S . . . per essersi lasciata abbagliare dalle sue stesse motivazioni . . .
S . . . per eclissarsi alle proprie motivazioni . . .
S . . . per aver accecato se stessa verso le sue proprie motivazioni . . .
S . . . di negare l’evidenza . . .
S . . . per abbagliare i propri motivi . . .

The ARGUMENT IS WAR metaphors have, on the whole, been translated

well. Pugnacious, (Table 1a) for an Italian speaker, bears a clear etymological
connection with pugno (‘fist’) which is implied in the cognate pugnante and, by
extension to the generic domain of fighting, in combattivo (‘combative’) and ag-
gressivo (‘aggressive’) and their derived forms; battagliero (‘battling’) is more ex-
plicitly bellicose. Attacked (Table 1b) was also translated appropriately, although
many of the students fell victim to source-text influence, six choosing the cognate
attaccare instead of the more appropriate aggredire. Although these words are
near-synonyms, dictionary definitions indicate that attaccare is associated with
physical contact (attaching things, and attacking in sports and military attacks)
whereas aggredire has a well-established figurative sense for verbal attacks (Ox-
ford-Paravia). In classroom discussion, it was plain that the students were not
aware of these nuanced differences in meaning, yet were it not for the formal si-
milarity with attack, it is doubtful that so many of them would have preferred
attaccare over aggredire.
The other two metaphors in the sentence are much more problematic, testi-
fied by the far greater variety found in the translations. Withering (Table 1a) is
conventional but not frequent and is therefore unlikely to have a place in lear-
ners’ working vocabulary. This makes it an interesting case study as far as
 Metaphorical Reasoning in Comprehension and Translation 137

translation strategies are concerned, since unfamiliar words prompt students to

use dictionaries. However, locating the appropriate meaning in a dictionary is
not always straightforward, and much rests on understanding how senses are
arranged. The metaphorical sense of withering cannot be transferred directly
into Italian via the use of the literal equivalent avvizzire because it maps onto a
different target domain: its metaphorical meaning is “to fade or languish from
lack of attention or love” (Oxford Paravia).
Of the ten students who produced translations for this text, one (student 10)
used the inappropriate literal translation, avvizzimento. Two (students 8–9)
chose words that are compatible with the Italian metaphorical meanings sugges-
ted by avvizzire mentioned above, but which are contextually incongruous in this
extract, namely declino (decline) and decadenza (decadence). In other words, the
students appreciated that the meaning was metaphorical but failed to recognize
the metaphorical meanings of wither and avvizzire do not correspond. A further
individual (student 7) appears to have bypassed the use of dictionaries altogether
and guessed the meaning from context alone – no other explanation can account
for her translating At its most withering with Nel suo più profondo (‘At its dee-
pest’/ ‘at heart’). In all four of these translations (students 7–10), the term chosen
is contextually inappropriate and should have been rejected as a result.
Context is all-important, and should be an aid to the translator in en-
suring that the choices he or she makes are consistent with the tone as well
as the overall meanings being expressed in a text. What withering means in
this context is reinforced in the lexicogrammar, most obviously through
the syntactic coordination of withering and pugnacious (coordination with
and normally implies an additional example of the same general idea).
More subtle, but essential to the complete understanding of the metaphor,
is the point that withering is agentive and causative; that is to say, the
speaker’s words make the hearer wither. Six of the translations manage
to convey the intended meaning fully, mainly drawing on the source do-
main of danger: feroce (‘ferocious’), spaventoso (‘fearsome’), and sferzante
(‘scathing’). Critico (‘critical’) is the only non-metaphorical translation
choice, and is an example of one of Newmark’s suggested strategies for
dealing with metaphor in translation, namely conversion of metaphor to
sense (Newmark 1981: 90). Due to the inappropriateness of using avvizzire
metaphorically, we can surmise that for the successful translations (stu-
dents 1–6) students not only recognized that the intended meaning was
metaphorical, but also consulted a bilingual dictionary, were able to locate
the figurative meaning successfully, and selected words whose meaning
were contextually appropriate.
138 Gill Philip

The last metaphor in the sentence, blinding (Table 1c) also proved a chal-
lenge. In one case (student 4), the metaphor was deleted due to misreading blind
as bind (‘legame’). Two students (5 and 9) opted for a literal paraphrase (conver-
sion of metaphor to sense), with nasconde a sé stesso (‘hide from oneself’), and
the more accurate negare l’evidenza (‘to deny the evidence’). Three other transla-
tion options were used, all drawing on the source domain of (lack of) VISION.
The first involves derivatives of the verb accecare (‘to blind’) (students 1, 3, and
8). The other two solutions proposed depart from the English word form but are
more idiomatically appropriate and convey the intended meaning successfully.
They are: eclissar(si) (literally, ‘to [be] eclipse[d]’) chosen by students 2 and 7,
and abbagliare (literally, ‘dazzle’) chosen by students 6 and 10. Although the
most successful versions (students 2, 3, 7, 9 and 10) adopt different strategies and
solutions, they manage to capture the meaning in just a few words. In contrast,
the translations that adhere closely to the source text phraseology sound clumsy.
The discussion of the ten translations of De Botton’s sentence allows some
interim claims to be made regarding the ‘problem’ of metaphor and how it rela-
tes to source text comprehension and target text solutions. In the first place, a
source text metaphor has to be recognized as such, because if it is read as lite-
ral, the resulting translation will not convey the intended meaning adequately.
In some cases the meaning may be radically changed, to the detriment of the
translation as a whole. Secondly, it is not enough just to recognise the meta-
phor, because the translator needs to be able to comprehend its meaning(s)
fully. Context helps, but equally important is the lexical environment –the
cotext– particularly collocations, grammar, and syntax. Metaphor recognition
and comprehension are also identified by Gibbs (1994: 116) as essential for un-
derstanding metaphor, although their sequencing appears to be reversed when
the stimulus text is in a foreign language (i.e. recognition before comprehen-
sion). Finally, conventional linguistic metaphors which co-exist in both source
and target languages are on the whole unproblematic in translation, provided
the metaphor does not contribute significantly to lexical cohesion, and/or is
not involved in word-play. This observation reinforces similar statements made
by Steen (2014: 16–17) and Prandi (2010: 318–320) with reference to “non deli-
berate” and “consistent” metaphor use respectively. Source text influence may,
however, lead translators to prefer cognates over more appropriate synonyms.
 Metaphorical Reasoning in Comprehension and Translation 139

5 Figurative Thinking and Unfamiliar Lexis

in Translation
The translated metaphors in this section feature in an extract from Vance Pac-
kard’s (1957) The Hidden Persuaders, an early text on marketing strategies. The
text was presented to two consecutive cohorts of students (a total of 28 transla-
tions were submitted). The focus here is on the ways in which the student trans-
lators dealt with the author’s exploitation of upper-/ middle- /low-brow and
brow-wise in the final sentence of the extract. The exploitations are not codified
in monolingual or bilingual dictionaries, making it necessary for students to re-
cognize the metaphors, comprehend them, and then interpret them – this third
phase also being listed by Gibbs (1994: 116–117) as essential to the processing
of metaphor.

[Lloyd Warner] defined his social classes not only in terms of wealth and power but in
terms of people’s consumption and sociability habits. This broader approach to differen-
tiation has received support from other perceptive observers of American society. Russell
Lynes, the Harper’s editor and writer, in his famous dissection of upper brows, lower
brows, and middle brows, used the tossed salad as a more reliable indicator of a person’s
status brow-wise than the size of his bank account. (Packard 1957: 99, emphasis mine)

Translating anything to do with social class can pose problems due to sociocultu-
ral differences in how social status is construed. An extract like this is therefore a
real test of comprehension and ability to express an alien concept in a way that
is meaningful in the target language. Eleven students translated upper brows,
lower brows, and middle brows adequately, using words found elsewhere in the
extract to bolster their interpretations: seven opted for (social) class, three focu-
sed on educational background, and one individual used strati (‘layers’). The ma-
jority, however, found the expression problematic. Five left the English text.
Eight used a literal translation equivalent: fronte (‘brow’, ‘forehead’), recognising
the metonymical connection between high foreheads and intelligence, which is
also present in Italian. Four miscued and opted for sopracciglia (‘eyebrows’).
These translations aroused some hilarity in class. One, delle sopracciglia alzate,
delle sopracciglia abbassate, delle sopracciglia al centro (‘of the raised eyebrows,
lowered eyebrows and eyebrows in the middle’), evokes eyebrow movement ex-
pressing surprise or puzzlement. Another student, having decided upon eyebrow
as the central translation, translated the expression as sopracciglia, ciglia inferiori
e superiori (‘eyebrows, and upper and lower eyelashes’).
These choices had repercussions for the later reference to brow-wise. Stu-
dents who failed to translate brow earlier also failed to translate brow-wise. Those
who had opted for social class or intelligence reiterated this interpretation, as did
140 Gill Philip

those who had made reference to foreheads. The ‘eyebrow’ group produced inter-
esting elaborations on the themes already introduced, adding to the comic relief
provided earlier, including a neologism intelligenza sopracciliare (‘eyebrow intel-
ligence’) and a version that suggested that a person’s status was determined in
base all’arcata sopraccigliare (‘by their eyebrow shape’).
Although the “brow” metaphor was one of the main reasons for selecting this
text for inclusion in the course, one of the most problematic features encountered
was not a metaphor at all, but a cross-reference to “people’s consumption . . . ha-
bits” (Packard 1957: 99): tossed salad. To the native speaker of English, the refe-
rence is transparent enough, though its truthfulness might be questioned. The
implication is that a preference for salad consisting of iceberg lettuce alone is
seen as conservative and culturally impoverished, and is indicative of lower social
class. Including more “exotic” leaves such as rocket or lamb’s lettuce suggests
greater cultural sophistication. The concept was alien to these Italian students,
for whom no such cultural stereotype exists: salad is, by definition, mixed.
Twenty students stuck close to the source text’s “tossed salad”, although
several indicated their uncertainty with question marks or set the word apart
from the body text with scare quotes. In class discussion, it transpired that they
assumed that they were dealing with a metaphorical expression, but since they
did not know its intended meaning they decided against deviating from the
source text term. This same explanation was given by four students who left a
blank at this point in the text. Two students made their figurative interpretation
explicit: l’immagine dell’insalata mista (‘the image of the mixed salad’ and la
metafora dell’insalatona (‘the metaphor of the large mixed salad’). A further
two students interpreted tossed salad metaphorically: as un modello misto (‘a
mixed model’) and as pinzimonio sociale (‘social crutités’2), thus preserving the
image of the salad within a novel metaphor.
This error –interpreting a literal expression as metaphorical– illustrates a
misapplication of figurative thinking, defined as “the use of a query routine
which assumes that an unknown expression might be figurative” (Littlemore
and Low 2006: 6). Here the expression is not unknown, but for these students
the literal meaning could not be accommodated into their perceived frame of
discourse regarding social class. As we would hope metaphor-aware students
might do, they tried to find a way to interpret the intended meaning by match-
ing up their knowledge of salads with plausible contextually-appropriate

2 Pinzimonio is a selection of raw vegetable sticks to dip in seasoned olive oil or marie-rose
sauce.
 Metaphorical Reasoning in Comprehension and Translation 141

meanings, as if they were dealing with a metaphor vehicle term. Their efforts
were not entirely unsuccessful.
This sentence from Packard’s Hidden Persuaders raises some additional
considerations about the ‘problem’ of metaphor. As before, it is clear that
source text metaphors must first be recognized as being metaphorical, and the
meaning in context has to be appreciated in order to find an appropriate candi-
date term for translation. Source text influence is again seen to lead translators
to prefer cognates over more appropriate synonyms. This time, we have seen
that this can lead to the coining of new linguistic metaphors (e.g. a new meta-
phorical use of fronte to mirror the English brow); it is not clear whether such
coinings are a deliberate attempt to adhere closely to the source text or whether
the unfamiliar expression creates favourable conditions for an equally unfami-
liar expression in the target text. This may therefore be either a deliberate trans-
lation strategy (i.e. foreignizing, Venuti 1995) or an interlanguage error, and
deserves further investigation. Finally, these students, who are being taught
how to identify and interpret metaphorical meanings in text, seem to apply fi-
gurative thinking as a general translation strategy for unfamiliar/ incongruous
expressions, e.g. in the translations for ‘tossed salad’. While the results proved
inaccurate in this case, that so many sought a solution outside the realm of the
literal can be seen as a vindication of the aims of the module.

6 L2 Language Proficiency and the Successful

Translation of Metaphor
During classroom time, when the students’ translations were projected on the
screen for discussion, it started to become clear that a considerable amount of
source text influence was present. This emerged primarily through (over-) adhe-
rence to the source text phraseology, syntax, collocations, and lexical choices,
and was particularly evident in the second cohort’s work. This prompted me to
look closer at the grades each of the students had achieved in their C1 language
test at the end of the previous academic year and to start matching features in
the translations with the five grade bands, A–E. In doing so, what emerged was
an immediate confirmation of my suspicions that the first cohort was not sim-
ply a ‘better’ group in terms of translation; their grade point average was in-
142 Gill Philip

deed a full band higher3than that of the second cohort and their Reading/use of
English was two bands higher.4 This suggests that a significant cause of
the second cohort’s translation problems was, in fact, their lower overall profi-
ciency in English, particularly their ability to fully understand and interpret the
source texts.
An overview of each cohort’s L2 language proficiency is shown in Table 2.
What the grade bands of these students does not immediately suggest is that
students’ English language proficiency should affect the quality of translations
into the L1, Italian. The translations prepared by students with lower L2 profi-
ciency were consistently less sophisticated than those of their peers. In addition
to calqued words, collocates and phrases, it is noticeable that the target text
lexis often fails to reflect the nuances present in the source text. While it may
be misplaced to assume that all students should be equally proficient in their
L1 writing skills, there does appear to be a connection between lower L2
comprehension and shortcomings in L1 translation.

Table 2: English language proficiency at level C1,

cohorts 1 and 2.

Grade band Cohort  Cohort 

A (/  e lode)  
B (–)  
C (–)  
D (–)  
E (–)  
Total  

To illustrate how lower overall language proficiency can impact translation,

this section focuses on part of a short essay entitled “What Contemporary Art
Means to Me”, in which the art critic David Lee lambasts the pretentiousness of
young contemporary artists. The text was chosen so that students could try
their hand at translating irony, contempt and incremental use of invective, and
was translated by 28 students in total – 14 from each cohort. The extract discus-
sed here opens the essay.

3 Cohort 1’s mean overall C1 language proficiency score was 27/30 [B], compared to 24/30 [C]
for Cohort 2.
4 Cohort 1’s mean reading comprehension and use of English score was 26/30 [C], compared
to 20/30 [E] for Cohort 2.
 Metaphorical Reasoning in Comprehension and Translation 143

What does ‘contemporary art’ mean to me, as distinct from plain old ‘art’?
First of all, it means an income. Not a very good one, but editing a magazine like Art
Review beats working for a living, especially if the alternative requires kowtowing to a suit
with a business studies degree from the University of South Bacup.
(Lee 2000: 197, emphasis mine)

Kowtow, a term borrowed from Chinese, is defined in Collins COBUILD as fol-

lows: “If you say that someone kowtows to someone else, you are criticizing
them because they are too eager to obey or be polite to someone in authority”.
Two bilingual dictionaries (Il Ragazzini, Oxford Paravia) paraphrase the term as
prostrarsi (davanti a) (‘prostrate oneself [before]’) but do not offer an alternative
which fully captures the metaphorical and pragmatic nuances. The students
overwhelmingly opted for two translations – prostrarsi as found in the dictio-
nary (13) and the somewhat more suitable inchinarsi (‘to bow’) (10), or other si-
milar images (piegarsi ‘bend down/over’, and inginocchiarsi ‘kneel’), as can be
seen in Table 3.

Table 3: Translations of kowtow and their distribution over the five English proficiency
grade bands.

Grade band bow prostrate other

A (/  e lode) – prostrarsi () scodinzolare ()

B (–) inchinarsi () prostrarsi () –
C (–) inchinarsi () prostrarsi () convertire ()
prostrarsi ()
D (–) piegarsi () prostrarsi () fare salamelecchi ()
E (–) – prostrarsi () –
Total   

The students who adapted the dictionary definition, interpreting it as more likely
to be ‘bow’ than ‘prostrate’ can be seen to belong to the grade bands B and C;
those who took the dictionary definition at face value come from all levels but
there is a concentration at grades C–D. No difference was found between students
studying Chinese as their second foreign language (L3) and the others. What did
come as a surprise, however, was that only two students –neither of whom knew
Chinese– provided idiomatic and pragmatically complete translations. One (L3
French) used fare salamelecchi, a borrowing from Arabic which translates literally
as ‘to do Salaam Aleikums’ which captures the deferential attitude in the action
and the observer’s disapproval; another (L3 Russian) opted for scodinzolare (‘to
wag one’s tail’), recalling the behaviour of a dog eager to please its master. The
144 Gill Philip

animal metaphor used of a human’s behaviour conveys a negative evaluation.

There was one anomalous translation (convertire – to convert) which looks like a
guess based on superficial phonemic similarities.
Kowtow collocates with the metonymy suit, which also produced a range of
translations, shown in Table 4. In class discussion of the students’ translations,
we reached the consensus that the best translation was simply cravatta (‘tie’),
which none of the students suggested in their written translation tasks: the ele-
ven students who used cravatta (mainly grade bands B and C) did so in
conjunction with giacca (‘suit jacket’), and all but one added uomo (or syno-
nym) to explicitate the metonymy as ‘a man in a suit and tie’. A further ten stu-
dents (examples from all grade bands, with a concentration at C and below)
translated the item of clothing, using a range of possible synonyms, none of
which adequately preserved the figurative meaning in the Italian text. Three
students adapted the metonymy, demonstrating that they understood that suit
referred to the person wearing a suit, e.g. un uomo d’affari (‘a businessman’) or
a manager, or qualche elegantone (‘some sharp-dresser’, maintaining the tone
of contempt in the ST). Four students (grades C and D) interpreted suit as causa
(‘law-suit’), despite the absence of a supportive context for this meaning.

Table 4: Translations of suit and their distribution over the five English proficiency grade
bands.

Grade band giacca e cravatta literal translation person wrong meaning

A (/  e lode) – divisa () elegantone () –

B (–)  completo () – –
C (–)  completo () un uomo d’affari causa ()
abito ()
tailleur ()
vestito ()
D (–)  completo () un manager causa ()
E (–) – costume () – –
Total    

As the data in Table 5 demonstrates, the fictitious placename South Bacup genera-
ted little in the way of variation, probably because language learners are encoura-
ged early on not to translate proper names unless there is an existing translation
in the TL. That said, from the tone of the earlier part of the sentence in which it
occurs, a reader ought to understand that the place being referred to is insignifi-
cant, making South Bacup either a real place that is held as emblematic of
 Metaphorical Reasoning in Comprehension and Translation 145

Table 5: Translations of South Bacup and their distribution over the five English
proficiency grade bands.

Grade band No translation Literal translation Creative translation

A (/  e lode)  – –
B (–)   –
C (–)   Nulla ()
Ripiegoland ()
D (–)   Merendine ()
E (–) – – Stocavolo ()
Total   

insignificance, or an invented place. Most students did not alter the place name,
although one did put scare quotes around it. Six students Italianised the location
by translating South as sud (5) or meridionale (1). This effectively reinforces the
literal meaning, while the English term allows some ambiguity to remain.
Of the four creative translations, there are two ‘real’ words: Nulla (‘noth-
ing[ness]’) and Merendine (‘children’s snacks’) plus two creative forms: Stoca-
volo, derived from ’sto cavolo (a euphemistic slang expression used to express
sarcastic disaffection) and Ripiegoland, which draws on ripiego (‘I fold/ bend
over’) attached to the suffix –land, used for the names of entertainment parks
throughout Italy. All four variants work exceedingly well, conveying contempt,
sarcasm and belittlement, and they were much appreciated by the students’ col-
leagues – several of whom confessed that they had been unsure how much li-
cence they could take in preparing their translations and had, as a consequence,
chosen the safe option of non-translation. The higher-achieving students were, at
least in this case, more conservative than their peers in preparing their work. It
can be hoped that the class discussions of the examples give them more confi-
dence in the future to attempt interpretative translations which depart from the
source text wording.
This final extract adds a further nuance to the ‘problem’ of metaphor in trans-
lation: L2 competence. While it is impossible to make conclusive claims on the
basis of one source language term, there seems to be enough evidence to support
the idea that lower proficiency is linked to a tendency to interpret metaphorical
language as literal, even when the context suggests otherwise. The higher the stu-
dents’ proficiency, the more likely they are to adapt and contextualise the trans-
lations proposed by bilingual dictionaries. Weaker students are less inclined to
adopt translation solutions which deviate from the ST wording and may even fail
to identify the correct meaning of figurative expressions. They appear to be more
146 Gill Philip

concerned with translating individual words than they are of translating texts,
and as a result their efforts may feature inexplicable incongruities. That said,
there are some aspects of learner behaviour which are not ascribable to profi-
ciency level alone. Some students –including high-achievers– may be surprisin-
gly conservative in their work, conscious as they are of having to fulfil the
expectations of the academic environment in which they are operating. In the
example discussed in this section, we have seen that students at the upper bands
of level C1 almost always recognize metaphors and attempt to translate them as
such, but they are inclined to explicitate, as in the translation of suit as uomo in
giacca e cravatta (‘a man in a suit’). Conversely, despite sometimes mistranslating
metaphors (especially if they have interpreted them as literal), lower-achieving
students also have their moments of glory, suggesting effective translations that
their ‘better’ classmates can learn from. Clearly, a translation is evaluated as a
whole and not for occasional flashes of brilliance, but the point here is that expli-
cit training in how to deal with figurative language in translation does bear fruit.

7 Conclusions
In this chapter I have offered an overview of an ongoing research and teaching
project dealing with metaphor in translation. It has emerged over the three years
that I have been collecting and analysing data that metaphor is problematic in
translation, mainly because learners have not acquired the necessary tools for
identifying figurative language (idioms excepted), have not been encouraged to
investigate polysemy, and at a certain point in their language learning they get
stuck in the rut of literal meaning. By encouraging advanced learners to notice
metaphors in non-fiction, to observe how they contribute to written texts and to
appreciate how their meanings can be interpreted, we arm them with a set of es-
sential skills. These help them to move forward both in their language learning
and in their translator training by showing them how to deal with words and
meanings that are not necessarily codified in dictionaries.

References
Dictionaries
Collins COBUILD on CD ROM v. 1.0. Glasgow: HarperCollins.
Oxford-Paravia Italian Dictionary, 2nd edn. 2006. Oxford: Oxford University Press.
Il Ragazzini Dizionario Inglese Italiano – Italiano Inglese 3rd edn. 1995. Bologna: Zanichelli.2
 Metaphorical Reasoning in Comprehension and Translation 147

Other References
Anthony, Laurence. 2017. AntConc version 4.3.4. Waseda University: Tokyo http://www.lauren
ceanthony.net/software/antconc/ (accessed 23 February 2018).
Bowker, Lynne & Peter Bennison. 2003. Student translation archive and student tracking
system. Design, development and application. In Federico Zanettin, Silvia Bernardini &
Dominic Stewart (eds.), Corpora in translator education, 103–117. Manchester: St Jerome.
Castagnoli, Sara, Dragos Ciubanu, Kerstin Kunz, Natalie Kübler & Alexandra Volanchi. 2011.
Designing a learner translator corpus or training purposes. In Natalie Kübler (ed.), Corpora,
language, teaching, and resources: From theory to practice, 221–247. Berlin: Peter Lang.
de Botton, Alain. 2012. Religion for atheists. London: Penguin.
van den Broeck, Raymond. 1981. The limits of translatability exemplified by metaphor
translation. Poetics Today 2(4). 73–87.
Danesi, Marcel. 1994. Recent research on metaphor and the teaching of Italian. Italica 71.
453–464.
Dickins, James. 2005. Two models for metaphor translation. Target 17(2). 227–273.
Dobrzyńska, Teresa. 1995. Translating metaphor: Problems of meaning. Journal of Pragmatics
24(6). 595–604.
Gibbs, Raymond W. Jr. 1994. The poetics of mind. Cambridge: Cambridge University Press.
Goatly, Andrew. 2007. The language of metaphors. London: Routledge.
Graedler, Anne-Line. 2013. NEST – a corpus in the brooding box. Corpus linguistics and
variation in English: Focus on non-native Englishes. http://www.helsinki.fi/varieng/se
ries/volumes/13/graedler/ (accessed 6 April 2018)
Granger, Sylviane & Marie-Aude Lefer. 2017. General report of the MUST kickoff meeting, 2–3
December 2016. Centre for English Corpus Linguistics – Université catholique de Louvain.
Louvain-la-Neuve.
Hanks, Patrick. 2006. Metaphoricity is gradable. In Anatol Stefanowitsch & Stephan Th. Gries
(eds.), Corpus-based approaches to metaphor and metonymy, 17–35. Berlin: Mouton De
Gruyter.
Kübler, Natalie. 2008. A comparable learner translator corpus: Creation and use.” LREC 2008
workshop on comparable corpora, 73–78.
Lee, David. 2000. What contemporary art means to me. In Ivo Mosley (ed.), Dumbing down:
Culture, politics and the mass media, 197–199. Bowling Green (OH): Imprint Academic.
Littlemore, Jeannette. 2009. Applying cognitive linguistics to second language learning and
teaching. Basingstoke: Palgrave MacMillan.
Littlemore, Jeannette & Graham Low. 2006. Figurative thinking and foreign language teaching.
Basingstoke: Palgrave MacMillan.
Moon, Rosamund. 1998. Fixed expressions and idioms in English. Oxford: Clarendon.
Nacey, Susan. 2013. Metaphors in learner English. Amsterdam & Philadelphia: John Benjamins.
Nacey, Susan. 2017. Metaphor in multiple learner corpus translations. In Pieter de Haan,
Sanne van Vuuren & Rina de Vries (eds.), Language, learners and levels: Progression and
variation. Corpus and language use – proceedings 3, 227–252. Louvain-la-Neuve:
Presses Universitaires de Louvain.
Newmark, Peter. 1981. Approaches to translation. Oxford: Pergamon.
Packard, Vance. 1957. The hidden persuaders. London: Penguin.
148 Gill Philip

Papadoudi, Dafni. 2010. Conceptual metaphor in English popular technology and Greek
translation. Manchester: University of Manchester. PhD thesis. https://core.ac.uk/down
load/pdf/40036079.pdf (accessed 12/ 09/2018)
Philip, Gill. 2006. Metaphor, the dictionary, and the advanced learner. In Elisa Corino, Carla
Marello & Cristina Onesti (eds.), Proceedings XII EURALEX international congress,
895–905. Alessandria: Edizioni dell’Orso.
Philip, Gill. 2010. “Drugs, traffic, and many other dirty interests”: Metaphor and the language
learner. In Graham Low, Lynne Cameron, Alice Deignan & Zazie Todd (eds.), Researching
and applying metaphor in the real world, 63–79. Amsterdam & Philadelphia: John
Benjamins.
Philip, Gill. 2011a. “ . . . and I dropped my jaw with fear”: The role of corpora in teaching
phraseology.” In Natalie Kübler (ed.), Corpora, language, teaching and resources,
49–68. Frankfurt: Peter Lang.
Philip, Gill. 2011b. Colouring Meaning. Collocation and connotation in figurative language.
Amsterdam: John Benjamins.
Philip, Gill. 2016. Review of Miller, D. & Monti, E. (eds.). (2014). Tradurre figure/ Translating
figurative language. Metaphor and the Social World 6(1). 156–168.
Pragglejaz group. 2007. MIP: A method for identifying metaphorically used words in
discourse. Metaphor and Symbol 22(1). 1–39.
Prandi, Michele. 2010. Typology of metaphors: Implications for translation. Mutatis Mutandis
3(2). 304–332.
Schäffner, Christina. 2017. Metaphor in Translation. In Elena Semino & Szofia Demjén (eds.),
Routledge handbook of metaphor and language, 247–257. London: Routledge.
Schäffner, Christina & Mark Shuttleworth. 2013. Metaphor in translation: Possibilities for
process research. Target 25(1). 93–106.
Shuttleworth, Mark. 2014. Translation studies and metaphor studies: possible paths of
interaction between two well-established disciplines. In Donna Miller & Enrico Monti
(eds.), Tradurre figure / Translating figurative language, 53–65. Bologna: Bononia
University Press.
Shuttleworth, Mark. 2017. Studying scientific metaphor in translation. London: Routledge.
Steen, Gerard. 2014. Translating metaphor: What’s the problem? In Donna Miller & Enrico
Monti (eds.), Tradurre figure / Translating figurative language, 11–24. Bologna: Bononia
University Press.
Steen, Gerard, Aletta Dorst, Berenicke Herrmann, Anna Kaal, Tina Krennmayr & Trintje Pasma.
2010. A method for linguistic metaphor identification: from MIP to MIPVU. Amsterdam &
Philadelphia: John Benjamins.
Uzar, Rafał & Jacek Waliński 2001. Analysing the fluency of translators. International Journal
of Corpus Linguistics 6. 155–166.
Venuti, Lawrence. 1995. The translator’s invisibility: A history of translation. London:
Routledge.
Tina Krennmayr
Metaphorical Alignment in Cross-Cultural
Office Hours’ Consultations
1 Introduction
When people engage in conversations, about 7.7 per cent of their speech is meta-
phorical (Kaal, 2012). Metaphors in talk are largely conventional (Cameron, 2003;
Kaal, 2012). They display dynamic patterns that develop as speakers take turns
speaking. They may be repeated, relexicalized, contrasted, expanded or elaborated
on (Cameron and Maslen 2010), thereby creating systematicity in metaphor use.
People may be “bouncing them back and forth, co-creatively recycling, extending,
fine tuning and retuning each other’s metaphors” (Carter 2004: 121). When a topic
is particularly difficult, metaphors tend to occur in clusters (Cameron and Stelma
2004). “People use metaphor to think with, to explain themselves to others, to or-
ganise their talk, and their choice of metaphor often reveals – not only their con-
ceptualisations – but also (. . .) their attitudes and values” (Cameron 2008: 197).
Metaphors in conversation thus fulfill a number of important roles.
In educational contexts, metaphor can be a powerful tool for lecturers to help
their students understand new and abstract concepts (Beger 2011). Lecturers fur-
thermore use metaphors to organize their lecture, to frame problems or to change
topic (Corts and Pollio 1999, Low 2010, Low, Littlemore and Köster 2008). For in-
ternational students who follow their education in a non-native language, this
has important consequences. They need to be able to produce and understand
metaphorical language. This is not easy. Knowing how to navigate metaphor in
talk may contribute to a learner’s ability to understand as well as to successfully
convey meanings.
Conceptual and linguistic differences in metaphor may lead to miscommuni-
cations and misunderstandings in intercultural contexts. In their analysis of uni-
versity lectures, Littlemore (2001) and Littlemore, Chen, Barnden, and Koester
(2011) found that overseas students tended not to understand or to misinterpret
metaphors. Misinterpreting metaphors means that the students failed to under-
stand the meanings the lecturers intended and, at the same time, were not aware
that they misunderstood. An example of such a misunderstanding is the meta-
phorical use of the conventional metaphor point, in ‘some point over the next

Tina Krennmayr, Vrije Universiteit Amsterdam, Amsterdam, Netherlands,

e-mail: [email protected]

https://doi.org/10.1515/9783110630367-008
150 Tina Krennmayr

week’ where a student interpreted the word to mean ‘interesting topic’, rather
than a reference to a particular moment in time (Littlemore et al., 2011: 15).
Learning to navigate metaphors in talk seems to be key to a learner’s ability to
understand as well as to successfully convey meanings.
In lectures, there is limited opportunity for engaging with the instructor,
and failures to understand can be difficult to address. May face-to-face conver-
sations between lecturers and students provide an opportunity to minimize
non- and misunderstandings? Do they allow students and lecturers to commu-
nicative more effectively and may they thus provide better learning experiences
for international students?
This study investigates metaphorical alignment of metaphor use in a teach-
ing context where speakers use English as a lingua franca. It explores a new
and relatively quick way of uncovering linguistic alignment (or the lack thereof)
between speakers in the context of office hours’ consultations at university: the
software tool Wmatrix (Rayson 2008).
Students whose first language is not English tend to misinterpret English lan-
guage metaphors used by lecturers (Littlemore 2001, 2003; Littlemore et al. 2011).
This is because metaphors can differ across linguistic and cultural communities
in several ways. Metaphors in one language can be rare or may not exist in an-
other (e.g. Deignan 2003); they may be used in both but may differ in their fre-
quency of use (e.g. Boers and Demecheleer 1997) or may differ in details. For
example, anger metaphors in Chinese conceptualize heat as a hot gas – not as a
liquid, as in English (e.g. Yu 1995).
The metaphors most likely to be culture-dependent are complex metaphors.
For instance, the source domain of the THE MIND IS A COMPUTER metaphor (e.g.
human memory storage) is probably unavailable in cultures not exposed to that
technology. Primary metaphors (Grady 1997), on the other hand, are more closely
rooted in physical experiences, and thus less likely to vary across cultures, since
the physical experiences are shared (e.g. conceptual metaphors such as LESS IS
DOWN; his results are below average).
Yet even metaphors rooted in physical experience have been shown to be in-
stantiated in different ways across languages (see Ibarretxe-Antuñano 2013).
While knowledge in the Western world is often talked about in terms of sight
(e.g. I see what you mean; you can also view this from a different perspective (e.g.
Danesi 1990; Sweetser 1991), this is not the case in many Australian languages
where knowledge is talked about in terms of hearing (Evans and Wilkins 2000).
Another example of a metaphor that is regarded as universal, namely anger as a
pressurized container, has been shown to exhibit differences across languages.
Differences pertain, for example, to the kind of container involved (the heart in
Zulu, the belly in Japanese etc.), what kind of substance fills the container (a gas
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 151

in Chinese, a liquid in English) and whether the substance is heated or not

(Kövecses 2005).
While Zulu, Chinese and English are quite dissimilar, more closely related lan-
guages also display differences. For example, the metaphorical sense of see (un-
derstand) is used twice as often than the literal one in the British National Corpus.
This difference is less pronounced with the Spanish (veo) and the Basque (ikusten
dut) equivalents (Ibarretxe-Antuñano 2013). Clearly there are not just differences in
metaphorical conceptualizations and linguistic form, but also differences in use.
Littlemore, MacArthur, Cienki and Holloway’s (2012) data consisting of one-
on-one sessions between lecturers and students reveal that native speakers share
metaphors when talking to each other. They repeat them and toss them back and
forth, thereby generating discourse coherence and shared understanding. This
observation is along the lines of what Niederhoffer and Pennebaker (2002) found
for other word categories than metaphor (e.g. prepositions or past tense verbs).
They discovered that people communicating with each other seem to coordinate
their interaction by “matching” their word use, both on a conversational and on a
turn-by-turn level. In other words, even at the basic word level, conversational
partners’ choice of words unrelated to content is very similar. While non-native
speakers participating in Littlemore et al.’s (2012) tutoring sessions are sometimes
able to share and further develop metaphors, misunderstandings are common.
The interlocutors are not always aware that one speaker used the words literally
while the other one meant them metaphorically, and the other way around. As a
consequence of the resulting misunderstandings, communicative goals may not
be achieved.
MacArthur, Krennmayr and Littlemore (2015) have also gone beyond re-
searching metaphors in the rather unidirectional discourse setting of the lecture,
in their case by investigating office hours’ consultations between lecturers and
students using English as a lingua franca. Besides some overlap, they found di-
vergent uses of the ways lecturers and students metaphorically talked about
knowledge and understanding. They detected substantial differences in terms of
frequency, forms and meanings of UNDERSTANDING IS SEEING metaphors. Successful
metaphor use seems to be dependent on “the way the conversational partners
enact their roles as collaborative participants” (MacArthur, 2016). Thus, what
seems crucial in cross-linguistic and cross-cultural research on metaphor use in
face-to-face conversations is to investigate in how far the conversational partners
‘align’ their use of metaphors. Do they take them up and repeat them? Do they
reformulate them or expand and elaborate on them?
This is, to my knowledge, the first study combining computational tools for
identifying metaphorical lexis and an investigation of metaphorical alignment
between speakers in a lingua franca situation. An analysis of semantic domains,
152 Tina Krennmayr

(which are treated as roughly corresponding to source domains) that are over- or
underrepresented in the lecturer versus the student sub-corpus, can provide in-
sight into the degree of alignment between speakers. It is important to note
though, that lack of alignment does not automatically mean that listeners did
not understand the metaphors introduced by their interlocutors. This research
can serve as a basis for further research that examines whether conversational
partners have actually understood the metaphors offered.

2 The Wmatrix Annotation Tool

Wmatrix (Rayson 2008), a web interface for corpus analysis, contains the USAS
(UCREL semantic analysis system) (Rayson, Archer, Piao and McEnery 2004), a
framework that automatically annotates each word of a running text semanti-
cally. It was not designed for the purposes of metaphor analysis. The system is
built on a large semantic lexicon. Each item in this lexicon has a syntactic tag
as well as one or multiple semantic tags assigned to it. The semantic tagset is
categorized into 21 main semantic fields, which are further subdivided into 232
more fine-grained semantic labels. The original tagset was loosely based on the
Longman Lexicon of Contemporary English (McArthur 1981) but has since been
revised (Archer, Wilson and Rayson 2002). An example of the semantic field
called sensory (X3) is presented below:

X3 Sensory
X3.1 Sensory: Taste
X3.1+ Tasty
X3.1− Not tasty
X3.2 Sensory: Sound
X3.2+ Sound: Loud
X3.2− Sound: Quiet
X3.3 Sensory: Touch
X3.4 Sensory: Sight
X3.4+ Seen
X3.4− Unseen
X3.5 Sensory: Smell
X3.5− No smell

The letters and the first digit designate a semantic field at its most general level.
Added digits after a dot indicate subdivisions; the finer the subdivision, the more
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 153

digits there are. The operators plus (+) and minus (−) indicate a positive or
negative position on a semantic scale. The semantic groupings represent senses
that “are related by virtue of their being connected at some level of generality
with the same mental concept” (Archer et al. 2002: 1). When a text is uploaded
and annotated by USAS, the output is a list in which each lexical unit is assigned
one or more semantic tags, as shown in Table 1.

Table 1: Sample excerpt of the European Corpus of

Academic Talk with semantic tags assigned.

POS-tag Text semantic tag

DD that Z Z
VBZ ’s A+ Z
APPGE your Z
NN focus X.+ X. A.+ A.+

The first column in Table 1 lists the part-of-speech tags (POS-tags), the second
column displays the words of the text and the final column indicates the se-
mantic tags for each lexical unit. When there are multiple semantic tags for a
word, the most likely tag is placed in initial position. The ranking is derived by
a combination of factors such as, for example, the POS tag of the word (e.g. if
spring has a noun tag, it filters out the jump sense), frequency (e.g. green as a
color is more likely than green as in inexperienced), the context a word is likely
to occur in (e.g. account of followed and preceded by a noun phrase most likely
refers to narration), or the surrounding words (for more details on methods of
disambiguation see Rayson (2003: 67–68) and Rayson et al. (2004). The codes
can be checked against a document provided on the Wmatrix platform listing
the labels for the semantic fields for each of the codes. In the example above,
the first tag of focus, X5.1+ stands for “attentive”, the second for “sensory:
sight”, the third for “detailed” and the last for “important”.

3 Uncovering Alignment with Wmatrix

Uncovering in how far interlocutors align their use of metaphors requires identi-
fying metaphors in the first place. Examples of metaphor identification proce-
dures are MIP (Pragglejaz Group 2007), or MIPVU (Steen et al. 2010), a more
detailed and refined variant of MIP. Both procedures analyze texts for metaphor
154 Tina Krennmayr

manually on a word-by-word basis. The core of these procedures entails compar-

ing and contrasting the contextual meaning with a more basic meaning of a
word. For example, the contextual meaning of the adjective high in “high status”
is “important compared to other people or things, especially in a particular sys-
tem or organization.” The lexical unit also has a more concrete, basic meaning,
namely “large in size from the top to the ground” (Macmillan). The two meanings
contrast, but the contextual meaning can be understood in comparison with the
more basic meaning. Therefore, in this context, high is metaphorically used.
Going through extensive stretches of discourse in this way is a time-
consuming task, primarily because it must be done manually. Current programs
for automatic metaphor identification do not yet match human coding abilities,
though there are some promising developments (e.g. Berber Sardinha, 2009;
Dodge, Hong and Stickles, 2015; Mason, 2004; Neuman et al., 2013; Patwardhan
and Pedersen, 2006; Shutova, Douwe and Jean, 2016).
In order to speed up identification and to process larger amounts of data,
several studies combine manual and computer assisted approaches. This can
be done by concordancing a predefined set of lexical items (e.g. Deignan, 2005;
Koller, 2004; Musolff, 2004) or by hand-coding a small corpus sample and then
performing a search in a bigger corpus based on what was found in the small
sample (e.g. Cameron and Deignan, 2003; Charteris-Black, 2004).
Rather than searching for pre-selected lexical items Koller, Hardie, Rayson
and Semino (2008) used a semantic approach to detect metaphorically used
words. They reasoned that the semantic fields that the annotation tool Wmatrix
automatically assigns to words of a text roughly correspond to metaphorical do-
mains, as suggested by conceptual metaphor theory. They can thus be exploited
for targeted metaphor analysis (Hardie, Koller, Rayson and Semino, 2007). While
the tool yields semantic fields that may correspond to source domains, it does
not identify metaphors. Concordances for all the lexical items in a semantic cate-
gory can be generated. Decisions as to which of the target words are actually meta-
phorically used need to be made by the analyst, e.g. by employing metaphor
identification procedures such as MIP or MIPVU. Wmatrix thus combines auto-
matic and manual analysis: automatic retrieval of potentially relevant lexis and
manual identification of metaphor.
A feature within Wmatrix that is particularly useful for uncovering alignment
between (groups of) speakers is the ‘keyness analysis’. The keyness analysis com-
pares a corpus to a reference corpus (a normative corpus such as samples from
the BNC or any other uploaded text). Semino, Hardie, Koller and Rayson (2009)
have successfully used the keyness analysis to compare metaphor across genres.
The comparison procedure produces those semantic categories that are signifi-
cantly more or less frequently used in the analyzed text than in a reference corpus.
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 155

For the analysis of conversations, the results of such a comparison can serve as a
first indication as to whether two (groups of) speakers are on the “same page.”
The following section will apply Wmatrix to a corpus of office hours’ con-
sultations (European Corpus of Academic Talk) (MacArthur et al. 2014) between
students and lecturers using English as a lingua franca to uncover alignment
(or the lack thereof) between students and teachers when using metaphors
from the semantic domain of SIGHT. Sight metaphors are expected to be promi-
nent in academic discourse because of the close connection between vision and
knowledge (Sweetser 1991) as noted above. For an overview of research on the
domain of sight to refer to knowledge and mental processes, including cross-
cultural variation, see MacArthur et al. (2015).

4 Method
4.1 The Corpus

The European Corpus of Academic Talk (EuroCoAT) consists of 27 conversa-

tions during office hours’ consultations between Spanish Erasmus students
and their lecturers at five European universities in Sweden, Ireland, the UK and
the Netherlands. The participants’ first languages are Spanish, English, Greek,
Dutch, German, Chinese, and Swedish. The recordings amount to a total of about
350 minutes. To investigate (lack of) metaphorical alignment when the interlocu-
tors draw on the domain of SIGHT, two sub-corpora are created (a lecturer and a
student corpus) and one of them is used as a reference corpus for the other. The
lecturer corpus consists of 40,432 and the student corpus of 17,317 words, total-
ing just under 60,000 words.

4.2 Detecting Over- or Underuse and Retrieving SIGHT

Metaphors

The Wmatrix dictionary contains three fields relating to sight: “sensory: sight”
(X3.4), “seen” (X3.4+) and “unseen” (X3.4−). As described above, when Wmatrix
assigns multiple semantic tags for a word, they are ordered according to likeli-
hood, placing the most likely tag in initial position. For example, focus in “that’s
your focus” is assigned the tag “attentive” first. It is the most likely tag and repre-
sents the contextual meaning of “the thing that people are concentrating on or
paying particular attention to” (Macmillan). The tag representing the basic
156 Tina Krennmayr

meaning of focus (the source domain tag), namely “the part of a camera, tele-
scope, microscope etc that you turn until you can see something clearly”
(Macmillan), is listed only second, indicating the high conventionality of the use
of focus in the present context. As a consequence, when examining the concor-
dances of the semantic field of “sight” they will not contain the lexical item
focus. This is because the concordance list only contains items that were as-
signed the “sight” semantic tag in the first position. This ordering is thus detri-
mental to metaphor analysis, because it means that relevant lexical items will be
missed. In order to retrieve a more comprehensive list, it is crucial that the tag
corresponding to the more basic meaning of a word is assigned as a first tag.
In order to force the program to assign the semantic tag that closely corre-
sponds to the most basic sense in first position, the lecturer corpus and student
corpus are uploaded using the ‘Domain Tag Wizard’, a feature that Koller et al.
(2008) added in order to facilitate metaphor analysis. It allows for privileging a
semantic field, placing it in the first position of a word’s tag list. This way the
concordance list includes words that have the sight tag in the first position of the
taglist and these are the words the metaphor researcher will be interested in.
Taking the example of focus above, the tag order using the Domain Tag Wizard is
X3.4 X5.1+ A4.2+ A11.1+ (“sensory: sight”, “attentive”, detailed”, “important”).
Using the Domain Tag Wizard (i.e. privileging the semantic field of sight) makes
focus part of the concordances listed under the “sensory: sight” semantic field.
To be maximally inclusive it is additionally possible to examine the so-
called broad list of each of the semantic fields related to “sight”. The broad list
includes all words that contain a certain semantic tag –regardless of its position
in the taglist. The researcher can examine this list manually and decide if any
of the words listed but not included in the concordances should be added to
the semantic field of interest.
Once the lecturer and student sub-corpora are tagged, they are compared
with each other performing the ‘keyness analysis’. In other words, one of the
two sub-corpora is used as a reference corpus to the other in order to give a first
indication of (mis)alignment: the analysis yields semantic fields that are used
more or less frequently in one corpus in comparison to the other (‘over-’ or
‘underuse’), as measured by LogLikelihood values. A value of 6.63 is used as a
cut-off point for 99% confidence of significance (Rayson 2003).

4.3 Coding Metaphors

While Wmatrix retrieves lexical units that belong to the domain of vision, it
does not automatically retrieve metaphors. Metaphors were identified using the
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 157

MIP procedure (Pragglejaz Group 2007). For each sight term in the concordance
line, it was determined if its contextual meaning could be compared, con-
trasted, and understood via a more basic meaning of the word. If so, it was
coded as metaphorically used. The data set was coded by three researchers,
each annotating a third of the complete corpus, cross-checking another third,
and making notes when disagreeing on coding decisions. Disagreements (23
cases) were discussed and resolved as a group.
Lexical units were put into one of three categories: non-metaphor, meta-
phor, and a hybrid category. There were three cases for which the contextual
meaning could not be established due to lack of context. They were discarded
from metaphor analysis. Examples (1) – (4) illustrate each category (transcript
identifier in parenthesis).

(1) let’s see if I (.) find one I marked recently (UE1, England; lecturer)
(2) I can see that it is a (.) difficult area (.) for almost everyone (UE6, England;
lecturer)
(3) I have some NOT so good responses to some questions and you will see
what people have done in the past (UE7, England; lecturer)
(4) We’ll we’ll we’ll look at you tell me what question (UI8, Ireland; lecturer)

The first example is a straightforward case of literal use of see: the lecturer is
searching on the computer screen. (2) is a straightforward case of metaphorical
use. See is used to mean understand. (3) is an example of an item in the hybrid
category. This category comprises items where both non-metaphorical and a
metaphorical meanings are at work at the same time. In this case, see is meta-
phorical because the student will understand what kind of answers are possible
when she is shown old exams but it is also literal because she will also look
(with her eyes) at the exam copies. (4) illustrates a case which was excluded
from metaphor analysis. The utterance is incomplete; the contextual meaning
of look at cannot be established.

5 Results
5.1 Keyness Analysis

There were three Wmatrix domains that clearly referred to ‘sight’. “Sensory:
sight,” “seen,” and “unseen”. Using a value of 6.63 as a cut-off point for 99%
confidence of significance (Rayson 2003), the semantic domain “sensory: sight”
158 Tina Krennmayr

was used significantly more frequently in the lecturer corpus than in the student
corpus, with a LogLikelihood value of 24.71 and a %DIFF of 46.57. The fields
“seen” and “unseen” were also used more frequently by the lecturers than by the
students but not significantly so. The number of lexical items in these latter fields
was small, however, (only four in “seen” and nine in “unseen” consisting of only
one lemma (notice) and two lemmas (miss and blind) respectively.) The signifi-
cant difference between lecturers and students with regard to the use of lexical
units from the semantic field of “sensory: sight”, may indicate a lack of align-
ment in the use of sight lexis and calls for closer investigation.

5.2 Sight Terms

The following lemmas were retrieved in the lecturer corpus: look, see, watch,
reflect, observe, spot, visible, focus, follow, view, viewpoint (semantic field “sen-
sory: sight”), blind, miss (semantic field “unseen”), and notice (semantic field
“seen”). In the student corpus the following items were found: look, see, watch,
visual, reflect, focus, view, follow (semantic field of “sensory: sight”), miss (se-
mantic field of “unseen”), and notice (semantic field of “seen”).
Two of the terms retrieved, namely follow and miss, are not part of the sight
domain. A more basic meaning of follow is “to walk, drive etc. behind someone,
when you are going in the same direction as them”, and a more basic meaning
of miss is “to fail to hit or reach something that you aim at.” These meanings
are movement-related, not sight-related, and were therefore excluded from the
analysis. This demonstrates that the concordances produced by Wmatrix still
need to be subjected to careful and critical inspection by the researcher to man-
ually filter out terms that do not actually belong to the sight semantic field.
Applying the Domain Tag Wizard catches most of the relevant lexis. In
order to check if any items that fit into one of the three lexical fields related to
sight have nevertheless been missed, the broad lists were checked for each of
them. As explained in section 4.2, a broad list contains all items that have a
sight-related field anywhere in the taglist, regardless of its position. The broad
list for the semantic field of “sensory: sight” yielded 31 hits that had a sight-tag
in their tagset but not in first position (see Table 2).
It is worthwhile to manually examine this list for potential sight lexis can-
didates. A number of items are clearly not related to sight. These are face- (part
of the compound Facebook), honor, internship, and note. The 23 cases of look
and the one case of see, however, are clearly sight terms and should be added
to the concordance list. See contained the tag “sensory: sight” in the taglist but,
even though the Domain Tag Wizard was used, it did not get pushed into first
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 159

Table 2: Broad list for the semantic field “sensory: sight” for the complete
EuroCoAT corpus.

lemma (frequency) first USAS tag semantic field

look like () A.+ comparing: similar

face- () B anatomy & physiology
honor () G.+ ethical
internship () M location and direction
look scholarly () P/A education in general/seem
note () Q. paper documents and writing
look to () S.- no power
look for () X. investigate, examine, test, search
see () X.+ understanding

position, which is why one case of see (“we are relatively lenient for plagiarism
as you see that you you just fail the essay and not the module”) did not show up
in the concordances. Wmatrix treated look like, look scholarly, look to, and look
for as single lexical units. While the lexical units look at and look up were as-
signed a sight tag in first position when uploading the corpus, look like, look
scholarly, look to, and look for were not and were thus only detected through the
broad list. The broadlists of “seen” and “unseen” did not reveal any new items.
Inspecting the broad list added 26 relevant lexical units. This yielded a total of
334 concordance lines: 76 for the student corpus and 258 for the lecturer corpus.
Figure 1 demonstrates that lecturers use a wider range of terms related to
sight than the students. The 258 tokens include twelve terms in the lecturer cor-
pus, whereas the 76 tokens in the student corpus include only eight terms.
Students thus make more limited use of sight terms, although this distinction
may be attributable to the smaller size of the student subcorpus.
The top three sight terms in the lecturer corpus are see (39.9% of all sight
terms), look (35.7 %) and focus (9.3%). These are also the top three in the stu-
dent corpus but they are ordered differently: focus (40.8%), look (23.7%) and
see (22.4%). Normed rates of all sight-related terms in both subcorpora are pre-
sented in Table 3.

5.3 Metaphorical Sight Terms

The top three sight terms in the corpus are also the top three metaphorically
used sight terms in both the lecturer and the student corpus: focus (54.4%),
look (21.1%) and see (15.8%) in the student corpus and see (40.4%), look
160 Tina Krennmayr

1,2% 0,8% 0,8% see

0,8% 1,3% 1,3%
1,9% look 1,3%
1,9% 2,6%
0,4% focus focus
2,7% 6,6%
4,79% view look
reflect see
9,3% 39,9% observe 40,8% reflect
spot 22,4%
watch
notice
35,7% view
watch
23,7% visual
visible
notice
viewpoints
blind

Figure 1: Sight terms in the lecturer (left) and the student sub-corpora (right) in per cent.

Table 3: Sight terms used by lecturers versus students

(normed per 10,000 words) (see also MacArthur et al. 2015).

sight term lecturers students

Focus . .

look . .
see . .
reflect . .
watch . .
view . .
visual . .
notice . .
observe . .
spot . .
visible . .
viewpoints . .
blind . .

(32.6%) and focus (12.4%) in the lecturer corpus (see Figure 2). As with sight
terms overall, the lecturers use a wider range of metaphorical sight terms (see,
look, focus, view, spot, reflect, notice, observe and viewpoints) than the students
(focus, look, see, reflect, view and notice). Spot, observe and viewpoints were
never metaphorically used by the students. Visual, watch and visible were never
metaphorically by any of the speakers. Normed rates of all sight-related terms
in both subcorpora are presented in Table 4.
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 161

1,6% 1,6% 1,8%

2,1% see 1,8%
2,6% 1,0% 5,3%
5,7% look focus

focus look
15,8%
12,4% 40,4% see
view
54,4% reflect
spot 21,1%
view
32,6%
reflect
notice
notice

observe

viewpoints

Figure 2: Metaphorical sight terms (clear cases and hybrid cases) in the lecturer (left) and the
student sub-corpora (right) in per cent.

Table 4: Metaphorical sight terms used by lecturers

versus students (normed per 10,000 words).

sight term lecturers students

focus . .

look . .
see . .
reflect . .
watch . .
view . .
visual . .
notice . .
observe . .
spot . .
visible . .
viewpoints . .
blind . .

6 Analysis and Discussion

The Wmatrix analysis reveals asymmetries in the use of sight metaphors used by
lecturers and students, suggesting that the conversational partners may not al-
ways align their metaphor use. Lecturers use a wider range of metaphorical sight
terms and largely make use of them more frequently than the students. While
power relationships, proficiency, personal style or lack of engagement may play
162 Tina Krennmayr

a role in this, the differences do not fully account for the lack of alignment. Note
that the use of sight metaphors by a Spanish-speaking lecturer in the corpus re-
sembled Spanish-speaking students’ use more closely than use by English-
speaking colleagues (MacArthur et al. 2015), which suggests that linguistic and
cultural differences may be at play. Moreover, examination of literal uses of un-
derstanding (semantic field X2.5+ “understanding” with the lexical items realize
and understand), reveals a significantly higher use of literal terms by students
than by lecturers.
A closer inspection of the transcripts reveals that students seem to struggle
with the lecturers’ use of sight metaphors. This is corroborated by the students’
short and restricted responses (e.g. yeah, okay, uhu and mhm.) Consider exam-
ple (5) from UE2 (England) in which a lecturer (NT) tries to give advice to a stu-
dent (MM) about how to best prepare for an exam:

(5) NT: but on the other hand you might look and see well actually when I look
at those topics a couple of them actually go together so it’s really only
one topic
MM: yeah yeah
NT: that it’s spread over two weeks
MM: mhm
NT: so you might look at that hh okay
MM: yeah
NT: so that that’s what you need to do so you KNOW exactly what you have
to do: so what i would say is the best thing to do is you you understand
that you’ve got the theory of the lectures so you’ve got to look through
your lecture
MM: yes
NT: notes

The lecturer frames the topic of exam preparation in terms of sight, drawing on
metaphorically used sight lexis on several occasions within a very short time-
frame: “you might look and see”, “look at those topics”, “look at that”, and
“look through your lecture”. The student does not pick up any of the sight meta-
phors, however. She neither repeats nor elaborates on them. Her restricted re-
sponses may indicate that she is having difficulties understanding what the
lecturer is trying to convey.
The student’s apparent inability to respond using metaphors that continue or
elaborate on those of the instructor, may seem surprising, given that this lecturer
repeatedly frames knowledge and mental activity in terms of vision throughout
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 163

the conversation. The lecturer used metaphors from the domain of sight 25 times
in total. Unlike in conversations between native speakers where interlocutors
share metaphors to create common ground and shared understanding (Cameron
2008; Littlemore et al. 2012), the student did not use a single sight term (neither
metaphorical nor non-metaphorical) during the whole office hour consultation.
There was complete lack of alignment. While the lecturer was heavily drawing on
the sight domain to talk about exam preparation, it seems the student was some-
what at a loss in her effort to understand the lecturer’s advice. This becomes
clearer as the conversation goes on, as illustrated in Example (6) – the student
still does not know “what to talk about” (see also MacArthur et al. 2015).

(6) NT: exactly and the other thing is if you look HARD at (.) the (.) topic and
you think (.) there is no debate here (.) it’s just facts it probably isn’t
going to be on the exam paper
MM: okay
NT: okay
MM: okay yeah
NT: so that can help you focus
MM: yeah
NT: a little bit more
MM: because (.) I’m I was really I really worried about that
NT: yeah
MM: because (.) sometimes you have a lot of theory but
NT: mm
MM: okay I can learn it but I don’t know what to talk about

The lecturer frames intense mental activity and careful thinking in terms of di-
recting one’s eyes at a physical object. Looking hard at an object means that one
can see it more clearly and detect its details and peculiarities. The metaphorical
use of looking hard at a topic, thus thinking about something very deeply and
carefully, can be understood via this more concrete meaning. The lecturer contin-
ues that looking hard can help the student focus, that is, concentrate on a set of
topics. Again, she uses a metaphor from the sight domain (focus), thereby con-
tinuing the frame she had set up. However, the student still does not take up any
of the metaphors from the visual domain. There is still no systematicity in the
use of sight metaphors. In short, the interlocutors do not align.
What may be particularly difficult for the student is that many sight meta-
phors tend to be simultaneously metaphorical and literal. For example, when
the lecturer in (5) says “you might look and see”, she refers to both the process
of physical looking, which is followed by the ‘product’ of understanding.
164 Tina Krennmayr

Earlier in the conversation both meanings are again at play at the same time
when the lecturer says “well actually when I look at those topics a couple of
them actually go together”. Look at here refers not only to mental activity (met-
aphorical) but also to reading (literal).
Students also seem to misinterpret metaphorical meanings as literal.
Consider example (7) from UE1 (England), where the student wants to get feed-
back on his essay.

(7) AC: okay yeah that’s fine you can also (.) ee as we saw in (.) as we saw in
the: (.) in the film fresa y chocolate no (.) the character cause
MM: ah yes i saw the film it was

In her explanations, the lecturer refers to the film “fresa y chololate”: see in “as
we saw in the film” refers to more than watching – it also expresses the idea
that the film was discussed. The student, however, seems to understand the
meaning purely literally, reacting with “ah yes I saw the film”. This mirrors
Littlemore et al.’s (2012) data, which also revealed that non-native speakers
may interpret metaphorically used words as literally used.
Examples (5) – (7) have illustrated a lack of alignment discovered by using
the Wmatrix semantic annotation tool. While Wmatrix is certainly useful to un-
cover (lack of) alignment, there are some caveats the researcher needs to be
aware of. Consider the following excerpt from US3 (Sweden) – a conversation
between a lecturer (FL) and a student (RH) in which the student gets feedback
on a written assignment.

(8) FL: you (.) need to correct me now because (.) when when I read your essay
it was sort of a little (.) difficult to follow your
RH: mhm
FL: train of thought (.) but I think that what you want to do (.) is to say that
he contradicts himself.
RH: yeah
FL: that is your main point
RH: mhm
FL: so this this ONE thing is something that you need to pick out from your
essay, lift out and make clearer (.) this is your main idea (.) you’re not
talking about (.) Oscar Wilde or (.) or or the (.) article you are talking
about how he contraDICTS himself in the article (.) this is your point (.)
and you need to lift it out
RH: yeah (.) I have to be more clear (.) I know
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 165

The lecturer frames advice on how to improve an essay in terms of concrete,

invisible things that need to be made visible (pick out, lift out, clear). The student
continues the metaphor by repeating the metaphorical use of clear. In this ex-
cerpt is seems that the student actually understood the lecturer’s advice – it is a
rare example of alignment between the student and the lecturer. What is interest-
ing to note is that the student chooses to repeat the most conventional of the lec-
turer’s metaphorical expressions.
However, pick out, lift out, and clear were not retrieved by Wmatrix. This is
due to the way the Wmatrix dictionary categorizes lexical items under specific se-
mantic field labels. The sight terms turned up by examining the Wmatrix seman-
tic fields that clearly reference “sight,” belong to those describing the physical
process of seeing (in their basic meanings) such as the ones in the metaphorically
used examples below:

(9) looking at the way you answer exam questions (UE7, England, lecturer)
(10) Alejo Carpentier sees the problems ee in the black ee people (UI1, Ireland,
lecturer)
(11) these people have this view (UE2, England, lecturer)

More peripherally related items were not found using Wmatrix. Danesi (1990:
223) established additional categories related to sight such as, for example
metaphors that refer to “differences in the physical conditions (. . .) which affect
the modalities of visual perception”:

(12) That was a brilliant idea

(13) I take a dim view on that whole affair.
(14) What you are saying is not very clear
(15) That is a transparent argument.
(16) Can you elucidate your idea?

and vision metaphors “which enlist the modalities involved in visual perspec-
tive”, such as:

(17) I have a different outlook than you do.

(18) Within hindsight, I would have done it in the same way.
(19) You have very little foresight on most issues.
(20) Please look your ideas over.
(21) Her speech threw light on the matter.
(22) Allow me to point out the weaknesses in your argument.
166 Tina Krennmayr

Also the lexical items from (23) – (25) that Sweetser (1991: 38) lists under the
KNOWING IS SEEING metaphor would go undetected.

(23) a clear presentation

(24) an opaque statement
(25) a transparent ploy

Pick out, lift out, and clear in Example (8) were not revealed by Wmatrix, because
none of them was categorized under any of the three sight-related semantic fields
of “sensory: sight”, “seen” and “unseen”. For example, clear was not detected be-
cause it received the following semantic tags: “likely”, “open; finding; showing”,
“colour and colour patterns”, “general appearance and physical properties”, “en-
tire; maximum”, and “weather”. None of these semantic tags pertains to one of
the three selected sight-related semantic fields. Instead, clear was assigned to the
semantic category “likely” – its first tag – and was thus not retrieved. Instead, its
15 uses are buried in a set of 699 concordance lines mainly containing modals
such as would, can, may, might, could and adverbs such as certainly and probably.
As a consequence, it is difficult for the researcher who has no detailed knowledge
of the complete corpus to detect this interesting section of discourse. In isolation,
pick out and lift out are movement-related metaphors. Yet in this excerpt they sur-
round to “make clearer” and together, they are “attracted” (Cameron and Low
2004) or drawn into the sight metaphor. This shows that conceptual domains are
flexible structures which are not automatically pulled out by applying the
Wmatrix tool. They require closer inspection of larger stretches of text surround-
ing metaphorical lexis.
A potential solution to these restrictions may be to run additional analyses
that assign domains that are related to sight only indirectly as a first tag. For
example, the three lexical units clear, transparent and brilliant share the tags
“colour and colour patterns”; one can additionally decide to examine the con-
cordances of that semantic field. While this is certainly possible, the following
questions arises: Where do we draw the line between the core semantic fields
describing the physical process of seeing and semantic fields that relate to
sight more indirectly? “Colour and colour patterns” is indirectly related to sight
because we see colors by distinguishing light of different wavelengths with our
eyes. Following Danesi, the semantic fields of “light” and “darkness,” which
are semantic fields within Wmatrix, could arguably be included into the analy-
sis as well. There are two instances of terms grouped under the semantic field
of “light” in the corpus: the non-metaphorical “enjoy the sunshine” and the
metaphorical “you need to think about these questions in that light.”
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 167

However, there are other, more vague semantic field labels included in the
Wmatrix tool such as “open/closed; hiding/hidden; finding; showing” (when
something is open we can see it, if something is hidden we cannot see it), “gen-
eral appearance and physical properties” or “judgement of appearance” (which
both refer to the way something looks). Whether or not these should also be
included seems a more difficult call.
Besides the challenge of determining where the border lies between fields
within and outside of the domain of SIGHT, checking every semantic field that
may be somehow related to sight, even if only remotely, and consequently sift-
ing through each of the concordance lines (most of which will not contain the
targeted word) defeats the purpose of searching a corpus relatively quickly with
reliance on the built-in semantic dictionary. A possible approach is that of
Demmen et al.’s (2015). They combine manual metaphor identification and se-
mantic tagging of a corpus subset with computer-assisted analysis of the com-
plete corpus to find lexical items in related semantic tags. This approach seems
to be especially helpful with big datasets.

7 Conclusion
Conversational partners have been shown to coordinate their choice of words
when talking to each other (e.g. Niederhoffer and Pennebaker 2002). Similarly,
this has also been observed for metaphorical language use. When native speak-
ers engage in conversations, they have been shown to repeat and expand on
each other’s metaphors in order to create knowledge and common understand-
ing (Cameron 2008; Littlemore et al. 2012). In conversations between partici-
pants using English as a lingua franca in office hours’ consultations, such
alignment is less pronounced. The lecturers’ metaphorical language use seems
to be a stumbling block to creating common ground and shared knowledge.
Investigating metaphorical alignment between speakers in lingua franca
contexts is crucial. Being able to understand and use metaphors in talk is key to
successful conversations. Part of this skill involves being able to create discourse
coherence and signaling understanding by repeating, expanding, changing, or
elaborating on the interlocutor’s metaphor use. Thus when participants in con-
versations align, they throw metaphors back and forth across turns. When meta-
phors are not systematically used (i.e. the conversational partner does not use
metaphors from the same or related source domains) this can be an indication of
lack of alignment.
168 Tina Krennmayr

This paper has shown that Wmatrix, through its semantic tagger and the
keyness analysis, is a valuable tool for investigating metaphorical alignment be-
tween (sets of) speakers. Its value goes beyond the case-study presented here
and promises to be useful for studies on different semantic domains that corre-
spond to potential source domains (e.g. domains M1 to M8 for source domains
related to movement or B1 “anatomy and physiology” for source domains related
to the human body) or with different groups of speakers. The tool provides a
quick way of detecting (lack of) alignment of metaphors from a pre-selected
source domain (in this case, SIGHT) by comparing two corpora (in this case, lec-
turers and students). When comparing the two corpora, asymmetric use of words
from semantic domains can be revealed. This case study has revealed an asym-
metric use of the domain of sight between lecturers and students. Words from
the semantic field of “sensory: sight” appeared far more often in the lecturer cor-
pus than in the student corpus, which was a first indication of lack of alignment
in student instructor interaction. Such a comparison can form the starting point
of closer investigation of the discourse. In the present dataset lack of alignment
was the norm. Students generally did not take up lecturers’ sight metaphors.
While not taking up metaphors does not automatically mean that they did not
understand them, a close look at the discourse shows that they did not seem to
understand the advice given and appeared to misinterpret metaphorical mean-
ings as literal.
Wmatrix partly automates retrieval of relevant lexis as it automatically as-
signs semantic fields to each lexical unit in the corpus. Tagging accuracy is
96–97% for POS tags and 91–92% for semantic tags (Rayson et al., 2004).
Wmatrix retrieves lexical items falling within a semantic field but it does not
retrieve metaphors. It is still left to the analyst to determine which of the lexical
items listed under a semantic field are indeed metaphorical. Nevertheless, the
tool can tremendously speed up analysis.
While the program is a valuable tool for uncovering alignment, the researcher
needs to be aware that the program may not detect all cases of (non)alignment in
the corpus. The analysis of sight metaphors has shown that a small number of
more peripherally related lexis will not be found. This is due to the way Wmatrix
groups lexis under specific semantic field labels. Some of these potentially rele-
vant sight-related terms may thus end up in semantic fields that do not explicitly
refer to the targeted domain. There is also a small number of tokens that may not
be listed under the pre-selected semantic fields (in this case “sensory: sight”,
“seen” and “unseen”.) These can be added by making use of the Broad List func-
tion. The researcher also must critically evaluate whether all lexical items listed
under the semantic fields of ‘sight’ indeed are a good fit, and those that are not
need to be excluded. Unlike a keyword approach, however, Wmatrix allows
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 169

searching for “open-ended sets” (see Demmen et al. 2015), meaning that the re-
searcher is not restricted to pre-selected lexical items.
As long as its caveats are kept in mind, Wmatrix is a valuable tool for inves-
tigating (metaphorical) alignment in talk. It speeds up analysis dramatically
and allows a quick way into the data. The analysis can provide a first step in
understanding challenges that may arise through the use of metaphor for
speakers communicating with each other using English as a lingua franca.
Future research may additionally look into the time dimension and address the
question of how alignment develops as the conversation unfolds.

References
Archer, Dawn Elisabeth, Andrew Wilson & Paul Rayson. 2002. Introduction to the USAS
category system. Benedict project report, October 2002.
Beger, Anke. 2011. Deliberate metaphors? An exploration of the choice and functions of
metaphors in US-American college lectures. metaphorik.de 20, 39–60.
Berber Sardinha, Tony. 2009. A tool for finding metaphors in corpora using lexical patterns.
Paper presented at the Corpus Linguistics 2009, Liverpool.
Boers, Frank & Murielle Demecheleer. 1997. A few metaphorical models in (western) economic
discourse. In Wolf A. Liebert, Gisela Redeker, & Linda R. Waugh (eds.), Discourse and
perspective in cognitive linguistics, 115–129. Amsterdam: John Benjamins.
Cameron, Lynne. 2003. Metaphor in educational discourse. London & New York: Continuum.
Cameron, Lynne. 2008. Metaphor and talk. In Raymond W. Gibbs (Ed.), The Cambridge
handbook of metaphor and thought, 197–211. Cambridge: Cambridge University Press.
Cameron, Lynne & Alice Deignan. 2003. Combining large and small corpora to investigate
tuning devices around metaphor in spoken discourse. Metaphor and Symbol 18(3).
149–160.
Cameron, Lynne & Graham D. Low. 2004. Figurative variation in episodes of educational talk
and text. European Journal of English Studies 8(3). 355–373.
Cameron, Lynne & Robert Maslen (Eds.). 2010. Metaphor Analysis: Research practice in
applied linguistics, social sciences and the humanities. London & Oakville: Equinox.
Cameron, Lynne & Juurd H. Stelma. 2004. Metaphor clusters in discourse. Journal of Applied
Linguistics 1(2). 107–136.
Carter, Ronald. 2004. Language and creativity: The art of common talk. London & New York:
Routledge.
Charteris-Black, Jonathan. 2004. Corpus approaches to critical metaphor analysis.
Houndmills, Basingstoke, Hampshire & New York: Palgrave Macmillan.
Corts, Daniel P. & Howard R. Pollio. 1999. Spontaneous production of figurative language and
gesture in college lectures. Metaphor and Symbol 14(2). 81–100.
Danesi, Marcel. 1990. Thinking is seeing: Visual metaphors and the nature of abstract
thought. Semiotica 80(3–4). 221–238.
Deignan, Alice. 2003. Metaphorical expressions and culture: An indirect link. Metaphor and
Symbol, 18(4). 255–271.
170 Tina Krennmayr

Deignan, Alice. 2005. Metaphor and corpus linguistics. Amsterdam & Philadelphia: John
Benjamins.
Demmen, Jane, Elena Semino, Zsófia Demjén, Veronika Koller, Andrew Hardie, Paul Rayson &
Sheila Payne. 2015. A computer-assisted study of the use of Violence metaphors for
cancer and end of life by patients, family carers and health professionals. International
Journal of Corpus Linguistics, 20(2). 205–231.
Dodge, Ellen, Jisup Hong & Elise Stickles. 2015. MetaNet: Deep semantic automatic metaphor
analysis. Proceedings of the Third Workshop on Metaphor in NLP, 40–49.
Evans, Nicholas & David Wilkins. 2000. In the mind’s ear: The semantic extensions of
perception verbs in Australian languages. Language 7(3). 546–592.
Grady, Joseph E. 1997. THEORIES ARE BUILDINGS revisited. Cognitive Linguistics 8, 267–290.
Hardie, Andrew, Veronika Koller, Paul Rayson & Elena Semino. 2007. Exploiting a semantic
annotation tool for metaphor analysis. In Matthew Davies, Paul Rayson, Susan Hunston &
Pernilla Danielsson (Eds.), Proceedings of the Corpus Linguistics 2007 Conference.
Ibarretxe-Antuñano, Iraide. 2013. The power of the senses and the role of culture in metaphor
and language. In Rosario Caballero & Javier E. Díaz-Vera (Eds.), Sensuous cognition,
109 –133. Berlin, Germany & New York: Mouton de Gruyter.
Kaal, Anna A. 2012. Metaphor in conversation. Oisterwijk: Boxpress.
Koller, Veronika. 2004. Metaphor and gender in business media discourse: a critical cognitive
study. Basingstoke & New York: Palgrave Macmillan.
Koller, Veronika, Andrew Hardie, Paul Rayson & Elena Semino. 2008. Using a semantic
annotation tool for the analysis of metaphor in discourse. metaphorik.de 15. 141–160.
Kövecses, Zoltan. 2005. Metaphor in culture. Universality and variation. Cambridge:
Cambridge University Press.
Littlemore, Jeannette. 2001. The use of metaphor by university lecturers and the problems that
it causes for overseas students. Teaching in Higher Education 6(3). 335–351.
Littlemore, Jeannette. 2003. The effect of cultural background on metaphor interpretation.
Metaphor and Symbol 18(4). 273–288.
Littlemore, Jeannette, Phillys Chen, John Barnden & Almut Koester. 2011. Difficulties in
metaphor comprehension faced by international students whose first language is not
English. Applied Linguistics 32 (3). 408–29.
Littlemore, Jeannette, Fiona MacArthur, Alan Cienki & Joseph Holloway. 2012. How to make
yourself understood by international students: The role of metaphor in academic
tutorials. ELT Resarch Papers (12–06). 1–22.
Low, Graham. 2010. Wot no similes? The curious absence of simile in university lectures. In
Graham Low, Zazie Todd, Alice Deignan & Lynne Cameron (eds.), Researching and
applying metaphor in the real world, 291–308. Amsterdam, The Netherlands,
Philadelphia, PA: John Benjamins.
Low, Graham D., Jeannette Littlemore & Almut Koester. 2008. Metaphor use in three UK
university lectures. Applied Linguistics 29(3). 428–55.
MacArthur, Fiona. 2016. Overt and covert uses of metaphor in the academic mentoring in
English of Spanish undergraduate students at five European universities. Review of
Cognitive Linguistics, 14(1). 23–50.
MacArthur, Fiona, Rafael Alejo, Ana M. Piquer-Píriz, Carolina Amador-Moreno, Jeannette
Littlemore, Annelie Ädel Krennmayr, Tina & Elaine Vaughn. (2014). EuroCoAT. The
European Corpus of Academic Talk http://www.eurocoat.es.
 Metaphorical Alignment in Cross-Cultural Office Hours’ Consultations 171

MacArthur, Fiona, Tina Krennmayr & Jeannette Littlemore. 2015. How basic is “UNDERSTANDING IS
SEEING” when reasoning about knowledge? Asymmetric uses of sight metaphors in office
hours consultations in English as academic lingua franca. Metaphor and Symbol 30(3).
184–217.
Mason, Zachary J. 2004. CorMet: A computational, corpus-based conventional metaphor
extraction system. Computational Linguistics 30(1). 23–44.
McArthur, Tom. 1981. Longman lexicon of contemporary English. London: Longman.
Musolff, Andreas. 2004. Metaphor and political discourse: Analogical reasoning in debates
about Europe. Houndmills & Basingstoke: Palgrave Macmillan.
Neuman, Yair, Dan Assaf, Yohai Cohen, Mark Last, Shlomo Argamon, Newton Howard &
Frieder Ophir. 2013. Metaphor identification in large texts corpora. PLOS ONE 8(4),
e62343.
Niederhoffer, Kate G. & James W. Pennebaker. 2002. Linguistic style matching in social
interaction. Journal of Language and Social Psychology 21(4). 337–360.
Patwardhan, Siddharth & Ted Pedersen. 2006. Using WordNet based on context vectors to
estimate the semantic relatedness of concepts. Proceedings of the Twentieth National
Conference on Artificial Intelligence. 1692–1693.
Pragglejaz Group. 2007. MIP: A method for identifying metaphorically used words in
discourse. Metaphor and Symbol 22(1). 1–39.
Rayson, Paul. 2003. Wmatrix. A statistical method and software tool for linguistic analysis
through corpus comparison. (Ph.D. thesis), Lancaster University, Lancaster.
Rayson, Paul. 2008. Wmatrix: a web-based corpus processing environment. Computing
Department, Lancaster University http://ucrel.lancs.ac.uk/Wmatrix/.
Rayson, Paul, Dawn E. Archer, Scott Piao & Tony McEnery. 2004. The UCREL semantic analysis
system. Proceedings of the workshop on ‘Beyond Named Entity Recognition Semantic
labelling for NLP tasks’ in association with 4th International Conference on Language
Resources and Evaluation (LREC 2004), Lisbon, Portugal, 7–12.
Semino, Elena, Andrew Hardie, Veronika Koller & Paul Rayson. 2009. A computer-assisted
approach to the analysis of metaphor variation across genres. Paper presented at the
Corpus Linguistics Conference 2009, July 2009, University of Birmingham.
Shutova, Ekaterina, Kiela Douwe & Maillard Jean. 2016. Black holes and white rabbits:
Metaphor identification with visual features. Proceedings of the 2016 Conference of the
North American Chapter of the Association for Computational Linguistics: Human
Language Technologies, 160–170.
Steen, Gerard J., Aletta G. Dorst, Berenike Herrmann, Anna A. Kaal, Tina Krennmayr & Trijntje
Pasma. 2010. A Method for Linguistic Metaphor Identification: From MIP to MIPVU.
Amsterdam: John Benjamins.
Sweetser, Eve. 1991. From etymology to pragmatics: Metaphorical and cultural aspects of
semantic structure (Vol. 54). Cambridge: Cambridge University Press.
Yu, Ning. 1995. Metaphorical expressions of anger and happiness in English and Chinese.
Metaphor and Symbol 10(2). 59–92.
Susan Nacey
Development of L2 Metaphorical Production
1 Introduction
This chapter details a pseudo-longitudinal corpus-based exploratory study into the
development of metaphorical production of foreign/second language (L2) learners
of English as they progress through their school career. The particular focus here is
on the written production of linguistic metaphors in the L2 English of pupils from
the ages of 10–19 in Norway, where the subject of English is obligatory from the
first grade (age six). The empirical data is retrieved from a longitudinal corpus cur-
rently under compilation called “Tracking Written Learner Language” (TraWL),
consisting of authentic L2 texts written by Norwegian pupils from grades five
through the final year in upper secondary school. All metaphors in 45 texts (5 per
grade level) were identified using the Metaphor Identification Procedure Vrije
Universteit (see Steen et al. 2010).
The overall objective is to shed light on how metaphorical production devel-
ops as proficiency increases. To do so, this study adapts the methods and many of
the research questions from Littlemore et al.’s (2014) investigation into metaphor
use at different levels of L2 writing. One main aim is thus to measure how overall
metaphor density varies per grade level. A related aim is to uncover any patterns
in the metaphor density for open-class versus closed-class metaphor, to identify
any particular level at which the use of the former overtakes the latter, as has
been observed in previous research. Moreover, this study investigates how the dis-
tribution of metaphor clusters varies across levels, since clusters have been found
to serve important communicative functions and might therefore be expected to
increase with improved proficiency (see Cameron and Low 2004). Finally, this
study examines the discourse function of the observed metaphors (including met-
aphor clusters) in texts from the different grade levels, to explore potential differ-
ences with respect to the role metaphor plays as L2 proficiency increases.

2 Background
This section presents the background for the study. Section 2.1 first outlines the
theoretical underpinnings for the view of metaphor adopted in this study, as

Susan Nacey, Faculty of Education, Inland Norway University of Applied Sciences, Hamar,
Norway, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-009
174 Susan Nacey

well as the links between metaphor and the L2 providing the motivation for the
research presented here. Section 2.2 then briefly outlines previous research in
the field, focusing on research that has contributed towards the development of
the particular research questions addressed.

2.1 Why Metaphor Production in the L2?

My approach is framed by Conceptual Metaphor Theory (CMT), which holds that

metaphor in language reflects metaphor structures in thought, i.e. “metaphors
as linguistic expressions are possible precisely because there are metaphors in a
person’s conceptual system” (Lakoff and Johnson 1980: 6). These structures may
or may not be consciously active in the mind of the producers or recipients of
utterances. This means that people may not necessarily be aware of “doing met-
aphor” even when they are in fact doing so, partially because metaphor is so
intrinsic to our understanding of the world around us that we do not always per-
ceive of it as something extraordinary. Metaphor pervades our thought (i.e. con-
ceptual metaphor) and is represented in language by conventional, and usually
codified, ways of expressing something (i.e. linguistic metaphor).
The primary focus of this investigation is on the level of language, in that all
linguistic metaphors produced in the pupil texts have been identified. Previous
empirical studies have found that linguistic metaphors are ubiquitous in both the
first language (L1) and in the L2. When it comes to the L1, Steen et al. (2010: 189,
195) identified all metaphors in roughly 200,000 words from the British National
Corpus, and found that an average of one in every seven and a half words was
related to metaphor. They explain, however, that the distribution of linguistic
metaphors varies considerably according to register, with academic texts having
the most metaphor (17.5%), followed by news (15.3%), fiction (10.8%), and fi-
nally conversation (6.8%). Nacey (2013: 139) provides further evidence of the
ubiquity of metaphor in L1 English, reporting a metaphor frequency of 13.3% in
almost 20,000 words of text written by British A-level pupils. Pasma (2011) re-
ports similar findings for L1 Dutch, providing empirical evidence that the ubiq-
uity of metaphor is not restricted to English alone.
When it comes to metaphor in the L2, Nacey (2013: 138) also investigates
metaphor frequency in English argumentative texts written by Norwegian col-
lege students, and finds an overall metaphor frequency of 15.5%. Because meta-
phor is a frequent feature of both spoken and written discourse regardless of
register or language, it clearly has important consequences for language learn-
ers. Production of metaphor is one facet of metaphorical competence, which in
 Development of L2 Metaphorical Production 175

turn is an important component of learners’ overall communicative competence

(see e.g. Littlemore and Low 2006).

2.2 Previous Research into L2 Metaphorical Production

Previous research about metaphor and language learners has focused primarily
on metaphor comprehension, interpretation and/or appreciation, with L2 stud-
ies frequently concentrating on metaphor in connection with vocabulary acqui-
sition and retention. Much less work has been carried out on production of
metaphor in the L2; those studies that have been conducted tend to offer “snap-
shot” views of L2 competence, through investigating metaphor production of a
particular population at a particular point in time. In most studies, informants
are college-age students, perhaps because they are the most readily accessible
to researchers (see Nacey 2017 for a more comprehensive overview).
How metaphorical production develops as L2 proficiency increases is an
under-researched area. One important exception is Littlemore et al. (2014), who
investigated metaphorical use in 200 exam texts collected in the Cambridge
Learner Corpus, half written by German learners of English and half written by
Greek learners of English. Each set of 100 texts consisted of 20 texts graded at
each of five of the Common European Framework of Reference for Languages
(CEFR) levels, from A2 to C2 (elementary English to proficiency English). They
find that the metaphor competence of L2 writers improves as proficiency in-
creases, and use these findings to propose CEFR descriptors for metaphor use,
something that had been missing from the framework.1 The present study con-
solidates four of Littlemore et al.’s (2014: 120) research questions into three sub-
ordinate aims, asking them of the Norwegian learner texts. More specifically:

1. Does the amount of metaphor produced vary across grade levels, both over-
all and as a factor of word class (open versus closed)?
2. Does the distribution of metaphor clusters vary across the grade levels?
3. What function does metaphor play across grade levels?

Note that Littlemore et al. (2014) had access to texts that were already graded
for CEFR proficiency levels. In the present study, by contrast, grade level and
age serve as a proxy for proficiency level. An underlying assumption here is
thus that L2 language proficiency generally rises with additional instruction,
learning opportunity and exposure to a language.

1 Their findings were also reported in Littlemore et al. (2012).

176 Susan Nacey

The present chapter builds upon the work of Littlemore et al. (2014) by ex-
ploring metaphorical production in the context of English language learning in
the Norwegian educational system, and by including young language learners
as informants. This type of expansion follows in the footsteps of Hoang and
Boers (2018), who also built upon Littlemore et al. (2014), albeit from a different
perspective: their goal was to examine whether the patterns uncovered by
Littlemore et al. (2014) hold equally true for texts written by a more homoge-
nous group of L2 writers (Vietnamese university students).

3 Primary Material and Methods

This section discusses the material and methods employed in the present study.
Section 3.1 provides details about the texts analyzed for metaphor use, together
with information about the informants who produced those texts. Section 3.2 then
explains the procedure employed for metaphor identification, including how
learner errors were dealt with in the analysis.2 Finally, section 3.3 discusses the
procedure employed for identifying metaphor clusters.

3.1 TraWL corpus

The data for this study comes from “Tracking Written Learner Language”
(TraWL), a longitudinal corpus whose compilation began in 2016. The corpus in-
cludes English texts written by Norwegian pupils from Grades 5 through the
final year of upper secondary school – that is, pupils aged 10–19. These texts were
collected from schools in five different Norwegian cities and towns, anonymized,
computerized (necessary for the texts by youngest pupils, who write mainly by
hand). Important to note is that in Norway, English is defined as a core subject
already from the first grade, at the age of six. As a result, ten-year-old Norwegian
pupils – the youngest included in the corpus – are already expected to be able to
produce simple texts in English, about themselves and their surroundings.
The TraWL corpus contains all pupil-produced texts written in the participa-
tory classes during the year(s) of collection, contingent upon consent from both
the pupil and his/her parents or guardians. In addition, most collected texts are
linked to some metadata about task type, instructions, writing aids, etc. No

2 Although the term “error” is employed here, this chapter is agnostic about whether the
anomalies were errors (due to a lack of knowledge) or mistakes (due to oversight).
 Development of L2 Metaphorical Production 177

attempt was made by the corpus compilers to influence text production. The
TraWL corpus therefore represents the type of authentic English texts that are “nat-
urally” produced in the English language classroom in different grades in Norway.
Because text collection began in 2016, few texts had been through the en-
tire preparation process by the start of the present exploratory study, so few
were available for research. As a consequence, the material included here com-
prises only 5 texts per each of the 9 grade levels, the work of 45 different pupils;
see Table 1 for an overview. Topics necessarily vary by grade, with the younger
pupils writing about subjects that are familiar in their everyday lives, and fre-
quently concrete in nature. Writing conditions also varied, with the texts from
Grades 8–11 having been produced under examination conditions.

Table 1: Overview of data: Grade, ages, numbers of lexical units and topics.

Grade Ages Lexical units Topic

 –  My family

 –  A fairy tale
 –  Drawing / dream
 –  My favorite artist
 –  The most interesting thing I know
 –  How to make newcomers feel welcome
 –  Why is it important to learn English?
 –  Problems immigrants face / prejudice
 –  A personal statement

A total of 14,546 lexical units3 were analyzed for metaphorical use. The number
of lexical units varies per grade level, with the fifth and sixth graders, perhaps
unsurprisingly, writing the shortest texts. The boxplot in Figure 1 presents a
more nuanced picture by showing the minimum and maximum text length per
grade, along with the median and mean (indicated by the bold line in the boxes
and the plus signs, respectively).4 The shortest text analyzed is a 20-word-long
text written by a fifth grader, while the longest text is 871 words long and writ-
ten by a twelfth grader.
Variation in text length may be considered a challenge when investigating
metaphor production. As an example, Hoang and Boers included text length as

3 The term “lexical unit” is defined in section 3.2.

4 The R software environment for statistical computing and graphics (2014) was employed to
produce all plots in this chapter.
178 Susan Nacey

800
Total = 14546 +
600
Lexical units

+
400

+
+ +
+
200

+
+
+
0

5 6 7 8 9 10 11 12 13
Grade

Figure 1: Numbers of lexical units per text and grade.

a factor for exclusion of texts from their investigation into metaphor production
in L2 English essays written by Vietnamese university students. They chose to
omit any text comprising fewer than 250 words, reasoning that “counts of meta-
phorical language in very brief essays with little content risk not being repre-
sentative”. As a consequence, Hoang and Boers felt compelled to exclude all
essays written by first-year university students, because what remained after
imposition of the minimum word count was considered too few to compare
with the essays from the three higher year levels (Hoang and Boers 2018: 3).
The word length variation of the texts documented in Figure 1 is hardly unex-
pected, due to the difference in ages between the youngest and oldest learners.
Text length was not considered as a factor for exclusion from the present analy-
sis, unlike in Hoang and Boers (2018). Indeed, their assumption that metaphor
production in the shortest texts would then be unrepresentative because of their
short length begs the question of exactly what the texts are intended to represent.
Imposing similar strict minimum word-length restrictions on the texts in this
study would have entailed omitting all texts written by fifth and sixth graders (at
a minimum), making it impossible to draw any conclusions about metaphor pro-
duction at these levels. Such a decision would have been unfortunate.

3.2 Metaphor Identification

The Metaphor Identification Procedure Vrije Universiteit (MIPVU) was em-

ployed to identify all linguistic metaphors in the 45 texts. This procedure
 Development of L2 Metaphorical Production 179

identifies both “indirect” and “direct” metaphor.5 Indirect metaphors consist of

those words where there is a contrast between the basic and contextual senses,
where that contrast may be attributed to a relationship of comparison. As an
example, consider the italicized word step in (1), found in a ninth-grade text
about “the most interesting thing I know”.

(1) I can take the next step and push myself even further . . .6

The basic meaning of step (that is, the most concrete, specific and human-
oriented sense in the dictionary) is “a short movement made by putting one foot
in front of the other”, the first sense entry for the noun in the online version of
the Macmillan Dictionary (MM).7 By contrast, its contextual sense is MM’s sixth
sense entry for the noun: “one of the stages in a process, or one of the levels on a
scale”. These two senses are sufficiently distinct (i.e. represented by different
sense entries), and are related through comparison whereby we understand an
advance in an abstract process in terms of physical forward movement. In similar
fashion, the words take, push, and further, in bold script in (1), are also indirect
metaphors.
In direct metaphors, an underlying cross-domain metaphorical comparison is
triggered through “direct” language use, i.e. by words where there is no contrast
between the contextual meaning and a more basic meaning. An example is found
in (2), from a twelfth-grade text.

(2) Do not judge a book by its cover.

In the cases of the verb judge and the nouns book and cover, there is no distinc-
tion between their more basic and contextual senses, even though there is an
underlying conceptual metaphor because the topic under discussion is immi-
gration, rather than books. Interpretation of this sentence, which directly
evokes an “alien” physical source domain unrelated to the topic at hand, re-
quires the addressee to set up a cross-domain comparison between the referents
of the words in discourse.

5 MIPVU also identifies “implicit” metaphors, consisting of elements where the metaphorical
relationship is due to an underlying cohesive grammatical link referring to metaphorical text,
such as with anaphoric pronouns or determiners. Although implicit metaphors were identified
during the analysis, they are arguably of lesser importance for the present research and are
not discussed.
6 All quotations from pupil texts are reproduced here as written, including any mistakes/errors.
7 The Macmillan Dictionary is available at https://www.macmillandictionary.com/.
180 Susan Nacey

The basic unit of analysis for MIPVU is the lexical unit. In most cases, the
lexical unit corresponds to the orthographic word, so the terms lexical unit and
word are henceforth used synonymously in this chapter, for ease of reference.
Note, however, that MIPVU also considers phrasal verbs, compounds, some
proper nouns, and polywords (e.g. of course, fait accompli) as single lexical
units even though they consist of more than one orthographic word; the identi-
fication procedure employed for this study followed the MIPVU guidelines
when demarcating such multiword lexical units.

3.2.1 Learner Errors

Many of the texts contain errors, as is only to be expected in texts written by

language learners at varying proficiency levels. For the purposes of the present
investigation, errors that were of no consequence for either metaphor identifi-
cation or word count were not adjusted. By way of example, consider the words
her self, now, and princesse in the “Cinderella” text written by a sixth grader,
shown in its entirety in Figure 2 below:

Figure 2: Sixth-grade “Cinderella” text.

 Development of L2 Metaphorical Production 181

When it comes to metaphor identification, the word princesse is unneces-

sary to alter. Its metaphorical status may be determined through consultation
of MM’s entry for the English noun princess, the only obvious intended target.
By contrast, her self (written as two words) was corrected to herself (one
word), because this type of error would have affected the calculation of both
the overall number of lexical units analyzed and the overall metaphor density
per text (that is, number of metaphorical words per number of total words).
Finally, the adverb now was corrected to the verb know. Strict application of
MIPVU would have required metaphor identification based on the word as
written in the text, rather than on the basis of the analyst’s interpretation of
the intended word. This type of application was deemed overly strict for the
present study, however, as it would have unduly skewed the numbers of po-
tential metaphors. Whereas more comprehensive error tagging may be tricky
(see e.g. Granger et al. 2009: 13), this type of minor adjustment of low-level
errors proved unproblematic.

3.3 Metaphor Clusters

Identification of metaphor clusters was accomplished through a series of time

analyses, one per text, following Littlemore et al. (2014: 123):

A span size of 20 words was selected, and the metaphoric density across the words in this
span was calculated. The result was placed at the mid-point (the 10th word). The span
was then shifted one word down, and the metaphoric density calculated for the next 20-
word span. The result was placed at the mid-point (the 11th word), and so on until the
end of the text was reached.

Littlemore et al. adopted the span size of 20 words from Turner (2010: 65), who
explains that while the choice was arbitrary, “choosing a different span length
will retain the general shape of the moving metaphor density chart”. The same
20-word span was adhered to in the present study, to more easily allow for the
possibility of comparison between studies. By coincidence, the shortest text in
the TraWL data under examination was exactly 20 words long, making the
span meaningful for all texts.
This time analysis procedure allows for the production of charts illustrating
the moving metaphor density of each text, such as that shown in Figure 3,
which was created based on one of the longest texts in the material. The cut-off
point for what is considered a metaphoric cluster is 30% metaphor density, indi-
cated by the bold horizontal line in Figure 3. The choice of 30% also follows
Littlemore et al. (2014: 124), a density they found to include “visible metaphor use
182 Susan Nacey

7
50
8
Moving metaphor density (%)
40

1 2 5
3 4 6
10 20 30
0

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
Word number

Figure 3: Moving metaphor density chart: 12th grade text, 828 words long.

above and beyond the sorts of highly conventionalized metaphorical uses of prep-
ositions”. A subsequent important procedural issue concerns identifying bound-
aries between clusters in those cases when there are many data points above the
30% threshold in close vicinity, an issue not explicitly dealt with in Littlemore
et al. (2014). In the present study, consecutive peaks were considered part of a
single cluster if the relevant 20-word spans overlapped.
In this way, eight metaphor clusters were identified in the text providing
the data for Figure 3: these are numbered above each peak. In each case, we
find a period of a relatively low level of metaphor use, followed by a burst of
metaphor use indicated by a peak in the chart. As an example, the text in (3)
forming Cluster 1 deals with problems facing second-generation immigrants,
and runs as follows (with metaphor-related words in bold script):

(3) Their parents are often eager to get them into a school, give them educa-
tion to get a job, and to do well in life. This could lead to a huge amount
of stress and cause obstacles in their children’s life.

In essence, each moving metaphor density chart provides individual metaphor

portraits for each text. This approach allows for relatively simple visual inspec-
tion and comparison of how metaphor usage varies in each of the 45 texts
 Development of L2 Metaphorical Production 183

under investigation.8 High localized metaphor density may prove informative

for subsequent qualitative analysis, by providing indications of where analysts
could focus attention.

4 Findings
The following subsections discuss findings for each of the study’s three re-
search questions. Section 4.1 concerns the overall metaphor density per grade
level, as well as the variation in metaphor density in open-class versus closed-
class words across grade levels. The variation in the distribution of metaphor
clusters across grade level is addressed in section 4.2, while section 4.3 dis-
cusses the functions of metaphor across grade levels.

4.1 Metaphor Density per Grade Level and Word Class

The scatterplot in Figure 4, where each dot represents one text, indicates that
there is a strong positive association between metaphor density and grade level.
As the predictor variable of grade level increases, so too does the criterion vari-
able of metaphoric density. The strength of this association is significant (rs =
0.85, p < 0.0001).
Metaphor density starts off fairly low in the earliest grades, including one
fifth-grade text and one sixth-grade text that contain no metaphor at all. In gen-
eral, the metaphor density progressively increases by an average of 1.8%
per year for each grade (indicated by the slope of the regression line), with only
minor discrepancies. For instance, a ninth-grade text contains the highest over-
all metaphor density of 19.1%, but this is closely followed by a text from Grade
13 with a metaphor density of 19%.
The black horizontal line in Figure 4 represents the previously cited find-
ings of Nacey (2013) for the 15.5% metaphor density in roughly 20,000 words
of argumentative texts written in English by Norwegian college students. Some
TraWL texts begin to approach the same metaphor density already in the ninth
grade, but it is only in the upper secondary years (Grades 11–13) that most of
the texts in this sample are on par with college students in this respect.

8 All 45 metaphor density charts, along with the data and R-code for this chapter, is available
at https://susannacey.inn.no
184 Susan Nacey

15
Overall metaphor density (%)
10
5
0

5 6 7 8 9 10 11 12 13
Grade

Figure 4: Metaphor density per grade level.

Figure 5 shows the identical scatterplot as in the previous figure, but

where the regression lines indicate the trends for variation of the metaphor
density per grade level of open-class words (dashed line) versus closed-class
words (solid line). The strength of these associations is significant, rs=0.76
and rs=0.69 respectively (p < 0.0001 in both cases). Here we see that the
metaphor density for open-class words grows more quickly than that of
closed-class words. More specifically, this model shows us that we may ex-
pect an average growth of 2.2% metaphor density for the former, and 1.4%
for the latter.
In their investigation, Littlemore et al. (2014: 128) found that the frequency of
open-class metaphors overtook that of closed-class metaphors at the B2 (upper
intermediate) level; they suggest that this development may indicate “a qualita-
tive type of metaphor that learners need to use at this level”. However, as we see
by the crossing regression lines in Figure 5, this transition point appears much
earlier in the TraWL texts. While the average metaphor density in the fifth-grade
texts is higher for closed-class than for open-class words, this relationship begins
 Development of L2 Metaphorical Production 185

30
Dashed grey = open-class words
Solid black = closed-class words
25
Metaphor density (%)
10 15 5
0 20

5 6 7 8 9 10 11 12 13
Grade

Figure 5: Metaphor density per grade level (regression lines for open-class and closed-class
words).

to change already by the next grade – far below the B2 proficiency level – after
which the difference steadily widens with each successive grade.9
The clear and strong correlation between the two factors of metaphor density
and grade level, however, does not necessarily result from a causal relationship.
Studies investigating metaphor use are subject to the confounding factor of topic.
Indeed, Littlemore et al. (2014: 121) described their efforts to minimize the impact
of topic type by extracting a selection of essays with some of the same words (e.g.
politician, politics, government, etc.). Despite such mitigation attempts, however,
there may still be minor variations in topic that inevitably effect the metaphorical
language produced. By way of possible explanation for the increased use of open-
class metaphor at B2 level and above, Littlemore et al. (2014: 128) acknowledge
that “this is likely to be the result of tasks set”.
In the present study, topic most probably has an even greater effect upon met-
aphorical production than in the Littlemore et al. (2014) study. This is partially due

9 The Norwegian Ministry and Research has determined successful completion of 11th grade
English as the minimum threshold to satisfy university admission requirements, a level that is
generally recognized as the equivalent of the CEFR B2 level (Hellekjær 2008: 13–14).
186 Susan Nacey

to the nature of the TraWL corpus, as it consists of assignments and examination

texts that naturally occur in the pupils’ classrooms. The corpus thus mirrors a gen-
eral progression in language learning, where younger, less proficient learners first
write about very concrete topics where metaphor often plays a subordinate role,
appearing most often in function words such as prepositions and determiners. A
typical such “concrete” topic was addressed in the fifth grade texts under study:
“My family”, exemplified in (4) below, presenting the shortest text in the data in
its entirety. This text has only a single metaphorical word, the adjective cool.

(4) My family is crasy

My Family is cool
My broder cries alot
My grandmother is smart
My grandfather is friendly

More complex themes, such as the tenth-grade and twelfth-grade topic of immi-
gration, involve more abstract issues. These naturally trigger the use of meta-
phorical language. Indeed, one of the main premises of CMT is that abstract
concepts are structured and mentally represented in our thoughts in terms of
concrete entities; these would then be expressed through linguistic metaphors in
written texts. The increase in metaphor density across grades shown in Figures 4
and 5 is due to the increasingly abstract nature of the topics discussed, indicat-
ing a correlation between grade level and abstraction level. Ensuring uniformity
of topic is thus a challenge in pseudo-longitudinal or longitudinal studies of
metaphor development, in cases when the researcher is unable to influence
topic choice and where the age of the informants greatly varies.

4.2 Does the Distribution of Metaphor Clusters Vary across

Grade Levels?

The scatterplot in Figure 6 shows the observed occurrences of metaphor clus-

ters per grade. Each dot represents one text, and a “jitter” function has been
utilized to slightly separate otherwise overlapping dots that represent texts
from the same grade containing the same number metaphor clusters.
We see here that no metaphor clusters were produced in texts from Grades
5–8, with the exception of a single outlier in a seventh-grade text. This testifies to
the relatively low level of metaphor in texts written by the youngest children, per-
haps due to task topic (as discussed above), and/or due to short text length. Other
than that, the first clusters appear in the ninth grade, a trend that continues in the
 Development of L2 Metaphorical Production 187

Metaphor clusters (observed occurrences)

8
6
4
2
0

5 6 7 8 9 10 11 12 13
Grade

Figure 6: Observed occurrences of metaphor clusters per grade.

later grades. All texts in the upper secondary school grades 11–13 contain meta-
phor clusters, with one twelfth-grade text containing nine of them – the maximum
number of metaphor clusters in a single text in my data. Pupils in the higher
grades are thus using metaphor at a more sustained rate than earlier. The non-
parametric regression curve in Figure 6 predicts that, after a certain point, the
number of metaphor clusters will increase as grade increases. In brief, once pupils
have begun to use metaphors in clusters, they will then continue to do so in in-
creasing numbers (at least up to a certain, as yet undefined point).

4.3 What Function Does Metaphor Play across Grade Levels?

Addressing this question called for a manual search of all 45 texts to determine
the main function of the identified metaphorically used words, both metaphors
that appeared in clusters and those that did not. This analysis was informed by
the functions of metaphors identified in the Littlemore et al. (2014) study:

– to serve an evaluative effect;

– to create a dramatic effect;
– to create dramatic contrasts;
188 Susan Nacey

– to organize discourse;
– to convey sarcasm;
– for humor; and
– to sum up/round off an argument.

The metaphorical language identified in the texts was also considered in light
of the stated aims for writing English that are outlined in the Norwegian na-
tional curriculum, where being able to express oneself in writing is described
as a basic skill in English (along with oral skills, being able to read, numeracy
and digital skills). The curriculum provides competence aims for the conclu-
sion of Grades 2, 4, 7, 10 (the final year of mandatory schooling), as well com-
petence aims for the upper secondary years of Years 11–13.10 While the
competence aims for Grades 2 and 4 fall outside the scope of this study, the
remaining descriptors are relevant for the present study.

4.3.1 Grades 5–7

The competence aims related to writing in English for the end of Grade 7 are to
enable pupils to do the following:

– use reading and writing strategies

– understand and use a vocabulary related to familiar topics
– write coherent texts that narrate, retell and describe experiences and express own
opinions
– use basic patterns for orthography, word inflection, sentence and text construction to
produce texts

As touched on in section 4.1, metaphor appears to have little clear role at these
levels, other than in incidental use with function words and with some highly
frequent lexical verbs. Consider (5) and (6), where all metaphor-related words
are written in bold text.

(5) I have a brother, a mother and a father.

(6) Once upon a time, there was three sister a dragon in a gingerbread house
and a magic helper Mr Rock.

10 The Norwegian subject curriculum for English is available (in English) at https://www.
udir.no/kl06/ENG1-03?lplang=http://data.udir.no/kl06/eng.
 Development of L2 Metaphorical Production 189

Example (5) is taken from a fifth-grade text where pupils were asked to write
about their families, a topic which naturally triggers the metaphorical use of
the verb have when discussing family members. Following MIPVU, the basic
meaning of the verb have is that of showing possession (MM’s third sense entry
for the verb), while its contextual meaning is that of stating a relationship be-
tween someone and their family members (MM’s fifth sense entry). These two
senses are sufficiently distinct, and the contrast between them may be under-
stood through comparison whereby we understand the relationship between
people in terms of possession.
Example (6) is found in a sixth-grade text where pupils were asked to write a
fairy tale. Here we find the metaphorical use of the preposition upon, whose basic
sense corresponds to that of physically touching a surface or object. Metaphorical
use of prepositions is hardly surprising, as previous empirical studies have found
prepositions to be the most metaphorical word class of them all: “no other word
class comes close to matching their metaphoricity” (Nacey 2013: 146). What is no-
table about this particular case, however, is that the preposition is contained in
the fixed expression once upon a time, functioning as a discourse marker and
thereby increasing cohesion. In this way, metaphor may begin to play a (minor)
role in these earliest years towards achievement of the Grade 7 competence aim of
writing coherent texts.
Despite a general tendency towards relatively little use of metaphor in the
lower grades, individual differences can nevertheless be observed. Looking
again at Figure 6 in section 4.2 showing the number of metaphor clusters per
grade, we find a single seventh-grade text with three clusters. The moving met-
aphor density chart for this outlying text is presented in Figure 7. The text frag-
ments forming the three identified metaphor clusters appear in sequential
order immediately beneath the figure.
As shown in Table 1 in section 3.1, the topic of the seventh grade texts in-
cluded in the present study was Drawing / dream. This broad topic resulted in
rather varied written material, depending on the drawing or dream in question.
The pupil responsible for the text from Figure 7 wrote about a crush he once
had on a girl in school. Here he uses direct metaphor to explain the more ab-
stract experience of love. Note that by the MIPVU protocol, all lexical words in-
volved in a direct metaphor are coded as metaphor-related words (e.g. heart,
beating, and peices [sic] in the first sentence of Cluster 2); in addition, direct
metaphors may also contain function words that are also metaphorical in use
(e.g. in to [sic], in the same sentence). This method for coding metaphor natu-
rally results in a high concentration of words that are marked as metaphor.
They thereby contribute to an increased metaphor density, and appear as meta-
phor clusters in metaphor density charts.
190 Susan Nacey

3
60

2
50
Moving metaphor density (%)
40

1
30
20
10
0

0 50 100 150 200 250 300 350 400 450 500

Word number

Figure 7: Moving metaphor density chart: 7th grade text, 508 words long.

Cluster = I remember that my heart beating and that I had butterflies in my

stomach.
Cluster  = It felt like my heart breaking in to a thousend peices. It felt like I
got punched in the gut. It hurt like knives stabbing me.
Cluster = It felt like someone punched my balls and gut. My heart broke in
to a thousend peices and started burning. I’m heart broken and angry.

Bold black script = metaphor-related words; Italics = metaphorical flags;

Bold grey script = WIDLII

In Cluster 1, the pupil introduces an apparent shift in topic – from love to

butterflies – that only makes sense when interpreted metaphorically. Added to
this is the remembrance of his heart beating, which may be interpreted literally
or metaphorically (categorized as the MIPVU “WIDLII” category [When in doubt,
leave it in], and hence written here in bold grey rather than black script). Cluster
2 contains a series of metaphorical similes introduced by the italicized metaphor
flags felt like and like, whereas a metaphorical simile in Cluster 3 is used to intro-
duce an extended metaphor that continues to the end of the text. While all three
 Development of L2 Metaphorical Production 191

clusters function as dramatic effect, this third cluster also functions as a conclu-
sion.11 In short, this student uses metaphor to “narrate, retell and describe experi-
ences”, Grade 7 competence aims that build upon those from earlier years.

4.3.2 Grades 8–10

The competence aims related to writing in English for the end of Grade 10 are to
enable pupils to do the following:

– choose and use different reading and writing strategies that are suitable for the purpose
– understand and use a general vocabulary related to different topics
– demonstrate the ability to distinguish positively and negatively loaded expressions refer-
ring to individuals and groups
– write different types of texts with structure and coherence
– use central patterns for orthography, word inflection, sentence and text construction to
produce texts

At these levels, we find more metaphors used with an evaluative function than in
earlier texts. For instance, in eighth-grade texts, Selena Gomez is described as
sweet, Gerard Way’s music has a dark aura, Justin Bieber is a big star, and Austin
Mahomed is characterized as cool – all (in these cases) positively loaded expres-
sions about individuals that pupils are expected to learn to distinguish. Despite
such examples, however, these pupils display a stronger tendency towards em-
ploying non-metaphorical language to describe individuals:

(7) Selena Gomez have black/brown hair, brown eyes and are not that tall.

(8) [Austin Mahomed] is also cut,12 strong, hot, nice, fantastic, happy, and
peaceful.

The pupil in (7) provides a physical, and hence non-metaphorical, description

of the artist, reminiscent of texts from younger pupils who described their

11 Note that while metaphor clusters provide indications of where analysts could focus atten-
tion, not all clusters have a readily identifiable discourse function such as was observed in
this seventh-grade text. For instance, the cluster documented as example (3) in section 3.3
does not appear to serve any particular unified role, despite containing a high concentration
of metaphor.
12 Analyzed as the adjective cute.
192 Susan Nacey

family members almost exclusively in terms of their physical attributes. Even

descriptions of other less concrete facets of artists, however, do not necessarily
lead to great amounts of metaphorical production. For instance, only one of the
seven adjectives used to characterize Austin Mahomed in (8) is a metaphor-
related word.
Metaphor at these levels also plays additional roles beyond evaluation, ap-
parent in the lower grades to a lesser extent. Examples are found in a text
about ninth-grade handball, the assigned topic being “the most exciting thing I
know”. The author of this text writes about development of prowess in the
game variously in terms of a battle, in terms of physical elevation, and in terms
of a difficult journey; see examples (9), (10) and (11) respectively.13

(9) personally I struggled with having confidence

(10) coaches build up players and boost their confidence

(11) I can take the next step and push myself even further and show every-
one that I can manage anything I want to.

Example (11) is the closing sentence in the text, and serves to effectively round
off the pupil’s arguments in a strong and evaluative way, a function that
Littlemore et al. (2014: 134) record observing at the CEFR C1 level. Other ex-
pressions involving metaphorical movement or journeys are evident in the
tenth-grade texts. As an example, one pupil writes about adjusting to different
cultures in terms of a path immigrants travel after they “come into my local
community”:

(12) To learn new and difficult languages and traditions in another community
is difficult. Especially when you don’t get help along the way.

Such movement metaphors arguably act as a cohesive element in texts, and

hence contribute towards achievement of the Grade 10 competence aims of
writing with structure and coherence.

13 Example (11) was previously cited as (1), but is reproduced here for the sake of conve-
nience, with co-text added for clarity.
 Development of L2 Metaphorical Production 193

4.3.3 Grades 11–13

The competence aims related to writing in English for the end of Grade 13 are to
enable pupils to do the following:

– evaluate and use suitable reading and writing strategies adapted for the purpose and
type of text
– understand and use an extensive general vocabulary and an academic vocabulary related
to one’s education programme
– write different types of texts with structure and coherence suited to the purpose and
situation
– use patterns for orthography, word inflection and varied sentence and text construction
to produce texts

At these levels, pupils show some adeptness at using metaphor with appropri-
ate phraseology and collocations, to a degree that is less evident in texts from
Grades 10 or below. Examples include the following:

11th grade: pursue a scientific career, consume information, X allows you

to tap into far more of the world’s intellectual resources, X will give you
the best return for your efforts
12th grade: spend time, a cold relationship [between children and parents],
confronted with questions, it all comes down to X, take the initiative
13th grade: make a mark on this world, make X see things in a different
light, a dream come true
Greater variation of discourse markers involving metaphorically used
words are also evident in the Grades 11–13 texts: on one side/on the other
side, after all, to sum up, in this way, in conclusion.
Note that not all collocations involving metaphorical expressions are
completely successful, in the sense that they do not all adhere to conventionally
established colligational or collocation patterns. Consider examples (13) and (14).

(13) [English is] an especially complex language to get grips with [11th grade]

(14) the kids are keeping secrets for their parents about their social lives and
who they are hanging out with [12th grade]

In (13), we see an unconventional variation of the expression to get to grips with,

where the pupil has omitted the preposition to. However, this omission does not
interfere with successful communication; the text is still comprehensible and
194 Susan Nacey

unlikely to be misunderstood. Such unconventional phrasing may be compared

to that in (14), where the pupil substitutes the preposition for for the preposition
from that normally appears in this collocation.14 With the preposition as written,
the sentence implies that the youth are protecting secrets that belong to the pa-
rents. The pupil arguably wanted to express the opposite – namely that the
youth are protecting their own secrets, rather than those of their parents.
Personification first appears in the TraWL texts at these grade levels, as in
(15) and (16), retrieved from two different eleventh-grade texts.

(15) . . . your message reaches out to a great deal of more people . . .

(16) Imagine if every country spoke only their own language . . .

While the phrasal verb reaches out, whose basis sense denotes that of a person
stretching out an arm, represents a clear case of personification, the sentence in
(16) can be analyzed in more than one way. One could interpret country as a me-
tonymy standing for the people within that country, in which case the verb spoke
would not be metaphorical in use. Alternatively, this instance can be understood
as personification where country is personified through the verb spoke, which
normally has a human agent. Rather than viewing the distinction between meta-
phor and metonymy as an “either/or” dichotomy, however, it is more useful to
view the two as a continuum between with two clear outer poles; between these
poles, metaphor and metonymy interact in different ways (see e.g. Deignan 2005:
60–61). The co-occurrence of personification and metonymy was also observed
in Littlemore et al. (2014: 132), and has been noted in other types of data and dis-
cussed in the relevant literature (e.g. Dorst 2011; Low 1999; Nacey 2013: 18)
Texts at these grade levels also evidence greater ability to use metaphor to
create textual cohesion. As an example, one twelfth-grade text about chal-
lenges immigrants face in Norway begins with the metaphorical idiom “Do not
judge a book by its cover” (previously cited as example (2)). The pupil employs
this idiom to establish an overall cohesive framework for the text, whose main
themes concern how immigrants may be unfairly judged due to their foreign-
looking appearance. Further, some pupils at these levels display an ability to
manipulate metaphor, as in (17) where the pupil alters the otherwise frozen
metaphorical expression “needle in a haystack”.

14 Following the MIPVU protocol, which does not consider the prepositions for or of for meta-
phoricity, the word for in this example is coded ‘not metaphor’. By contrast, the preposition
from, appearing in this same context, would be coded as in indirect metaphor. Its basic mean-
ing is MM’s 3a sense entry: ‘starting at a particular place and moving away’.
 Development of L2 Metaphorical Production 195

(17) From the big perspective I’m no significance, not even the slightest bit
of a needle in a haystack.

Such creative appropriation of an established English expression provides pos-

sible evidence of increasing metaphorical competence as the pupil’s “feel” for
the language grows, facilitating experimentation with wording (see MacArthur
2010: 164). It should be noted, however, that collocational deviation is a risky
strategy for L2 learners, whose potential linguistic creativity may be taken for
linguistic error, even when their production is understandable in context (see
e.g. Nacey 2013: 163–168).
Finally, the use of metaphor to conclude arguments, while evident in ear-
lier grades, may be observed more often in texts written in Grades 11–13. An
example is provided in (18), serving as the conclusion of an eleventh-grade text
about the importance of learning English.

(18) When a language barrier is lifted, there is more space for opportunities
understanding and for common ground.

Here we find both space and ground from a source domain of some physical
location, with barrier and lifted adding a sense of physical obstacle. Such close
proximity of several metaphors from the same or related source domains con-
tributes towards reawakening the perceived metaphorical nature of the words.
This use represents an instance of potentially deliberate metaphor, inten-
tionally produced as metaphor through the introduction of an alien element to
change the reader’s perspective (see Reijnierse et al. 2017). Littlemore et al.
(2014: 137, 142) document parallel examples from their data, and argue that
such production belongs in the CEFR descriptors for the C2 level.

5 Concluding Thoughts
The overall objective of the present explorative study has been to shed further
light on the development and function of metaphorical production as profi-
ciency increases. The investigative approach adopted here has featured the ex-
ploration of both quantitative and qualitative facets of English metaphorical
production in L2 learner texts from the Norwegian Grades 5–13, where pupils
range from ages 10–19.
When it comes to quantitative measures, we have first seen that the meta-
phor density of texts increases steadily as grade level increases (section 4.1). In
196 Susan Nacey

addition, while the metaphor density of function words is greater than that of
open-class words in texts written by the very youngest students (ages 10–11),
this relationship quickly changes. Pupils begin to produce relatively more meta-
phors in the open-word classes than in the closed-word classes already by the
next grade, when pupils are aged 11–12 – a trend that continues unabated as
the grades and ages increase. Further, metaphor clusters, defined here as a
30% metaphor frequency over a minimum 20-word span, hardly appear in texts
written in the lowest grades, arguably resulting from a short text length and
relatively concrete topics. However, metaphor clusters do appear in texts from
the higher grades, in increasing frequency as grade level rises.
When it comes to more qualitative analysis, this study indicates that the
role of metaphor in discourse evolves as the grades increase. In the lowest
grades (5–7), metaphor has no clear function, although there are individual ex-
ceptions. Texts in later grades (8–10) evidence functions such as evaluation
and conclusion. In the uppermost grades (11–13), metaphor contributes to a
wider variety of textual features: (mostly) appropriate phraseology and colloca-
tions, more varied discourse markers, greater textual cohesion, personification,
along with summing up and concluding.
All texts at these highest levels in Grades 11–13 contain metaphor clusters.
The occurrence of such concentrated pockets of metaphor may thus be indicative
of a qualitative change in the type of metaphorical language required of more
proficient language users, in discussions that address more abstract topics than
those dealt with in lower grades. As discussed in section 4.1, Littlemore et al.
(2014) suggest that such a qualitative change may be indicated by the point at
which the metaphor density of open-class words becomes greater than that of
closed-class words, found at the B2 level in their material. That this transition
point appears so much earlier in the TraWL texts investigated here raises ques-
tions concerning the significance of this point, as well as the proficiency level at
which it generally occurs.
The present chapter details an initial study of metaphor production across
grade level, based on data from a longitudinal corpus that is currently under
compilation. As a consequence, the findings presented here should be inter-
preted with caution, particularly as they are based upon only five texts per
grade level. Moreover, future investigation into such issues should ideally con-
trol for the factors of topic uniformity and text length, considerable challenges
for investigations into the development of metaphorical production in either
longitudinal or pseudo-longitudinal investigations dealing with a wide range of
ages and/or including young learners. This study nevertheless represents a first
exploration of such data, paving the way for future research.
 Development of L2 Metaphorical Production 197

References
Cameron, Lynne & Graham Low. 2004. Figurative variation in episodes of education talk and
text. European Journal of English Studies 8. 355–373.
Deignan, Alice. 2005. Metaphor and corpus linguistics. Amsterdam: John Benjamins.
Dorst, Aletta G. 2011. Metaphor in fiction: Language, thought and
communication. Amsterdam: VU University Amsterdam dissertation. http://dare.
ubvu.vu.nl/bitstream/handle/1871/19629/dissertation.pdf?sequence=1 (accessed
23 August 2018).
Granger, Sylviane, Estelle Dagneaux, Fanny Meunier & Magali Paquot (eds.). 2009.
International corpus of learner English: Version 2. Louvain-la-Neuve: Presses
universitaires de Louvain.
Hellekjær, Glenn Ole. 2008. Lesing som grunnleggende ferdighet: En utfordring for
engelskfaget. [Reading as a basic skill: A challenge for the subject of English]. Språk og
Språkundervisning (4). 10–15.
Hoang, Ha & Frank Boers. 2018. Gauging the association of EFL learners' writing proficiency
and their use of metaphorical language. System 74. 1–8. doi:https://doi.org/10.1016/j.
system.2018.02.004 (accessed 23 August 2018).
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago, IL: University of Chicago
Press.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sarah Turner. 2012. Investigating
figurative proficiency at different levels of second language writing. Cambridge ESOL
(47). 14–26.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sarah Turner. 2014. An investigation in
metaphor use at different levels of second language writing. Applied Linguistics 35(2).
117–144.
Littlemore, Jeannette & Graham Low. 2006. Figurative thinking and foreign language learning.
Basingstoke: Palgrave Macmillan.
Low, Graham. 1999. “This paper thinks. . . ”: Investigating the acceptability of the
metaphor AN ESSAY IS A PERSON. In Lynne Cameron & Graham Low (eds.),
Researching and Applying Metaphor, 221–248. Cambridge: Cambridge University
Press.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where we are and where we should
be going? A view from the classroom. AILA Review 23. 155–173.
Nacey, Susan. 2013. Metaphors in learner English. Amsterdam: John Benjamins.
Nacey, Susan. 2017. Metaphor comprehension and production in a second language. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
503–516. London: Routledge.
Pasma, Tryntje. 2011. Metaphor and register variation: The personalization of Dutch news
discourse. Amsterdam: VU University Amsterdam dissertation. http://dare.ubvu.vu.nl/
handle/1871/19630 (accessed 23 August 2018).
R Core Team. 2014. R: A language and environment for statistical computing. Vienna, Austria:
R Foundation for Statistical Computing. http://www.R-project.org/ (accessed
23 August 2018).
198 Susan Nacey

Reijnierse, W. Gudrun, Christian Burgers, Tina Krennmayr & Gerard J. Steen. 2018. DMIP: A
method for identifying potentially deliberate metaphor in language use. Corpus
Pragmatics 2(2). 129–147.
Steen, Gerard J., Aletta G. Dorst, J. Berenike Herrmann, Anna A. Kaal, Tina Krennmayr & Tryntje
Pasma. 2010. A method for linguistic metaphor identification: From MIP to MIPVU.
Amsterdam: John Benjamins.
Turner, James 2010. Investigating figurative language in EFL learners' writing across levels of
proficiency. Birmingham: University of Birmingham MA thesis.
Irene Castellano-Risco and Ana M. Piquer-Píriz
Measuring Secondary-School L2 Learners
Vocabulary Knowledge:
Metaphorical Competence as Part of
General Lexical Competence
1 Introduction
Figurative language pervades our daily communicative exchanges although it
very often goes unnoticed by speakers, even in their native language (L1). In the
process of learning a foreign language (L2), the importance of dealing with non-
literal meanings has been widely recognised (Cameron and Low 1999; Littlemore
and Low 2006b; Low 1988). Two main reasons seem to justify paying attention to
figurative meanings in the L2 classroom: On the one hand, metaphor has been
shown to be ubiquitous in language (Lakoff and Johnson 1980; 1999) and figura-
tive uses are extensively employed in any communicative situation, including
the very basic ones, from early infancy (see Özçalışkan 2011 for a review) and,
therefore, L2 learners need to deal with them from the first stages of the L2 learn-
ing process and seem to be ready to do so (see Piquer-Píriz, this volume). On the
other hand, the misuse and misunderstanding of this type of language has been
reported as one of the reasons why non-native speakers are perceived as out-
siders in social interactions. Some scholars (Danesi 1992, 2008, 2016; Littlemore
2001; Littlemore and Low 2006a) have even pointed at the importance for L2
learners of developing what has been labelled as “metaphorical competence”. In
practice, however, neither defining what this competence consists in nor finding
ways to empirically measure it have proved to be easy tasks. Recently, some re-
search studies have examined learners’ production of metaphor (Hoang and
Boers 2018; Littlemore, et al. 2014; Nacey 2013; Turner 2014). In this chapter, we
focus on learners’ understanding of metaphor and propose a way of measuring
what could be a specific dimension of L2 learners’ metaphorical competence,
namely, their understanding of metaphorically-used words.
The chapter is structured as follows: first, the notion of metaphorical compe-
tence is analysed by considering the different theoretical approaches to this

Irene Castellano-Risco, English Philology Department, University of Extremadura, Badajoz,

Spain, e-mail: [email protected]
Ana M. Piquer-Píriz, English Philology (Faculty of Education), University of Extremadura,
Badajoz, Spain, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-010
200 Irene Castellano-Risco and Ana M. Piquer-Píriz

concept found in the literature as well as the main proposals that have been
made to explore it in L2 learners’ language. Here we outline our own proposal to
measure what could be considered one dimension of learners’ metaphorical
competence, specifically, their ability to understand figurative meanings. After
that, we present a study in which, following this proposal, we gauged the under-
standing and recognition of figurative meanings by a group of seventy-seven
third grade secondary school learners of English (aged 14-15), combining tools
that have been widely used and validated in vocabulary studies, i.e., the 2K and
the Academic Level versions of the Vocabulary Levels Tests (Schmitt, Schmitt
and Claphman 2001) and the Productive Vocabulary Levels Tests (Laufer and
Nation 1999), in conjunction with methods developed by metaphor scholars
(MIPVU, Steen et al. 2010) with the aim of highlighting how these two research
areas can be mutually enriching.

2 Metaphorical Competence
The notion of metaphorical competence can be traced back to the 1970s in the
work carried out in experimental psychology. In a seminal paper published in
1978, Gardner and Winner already attempted to establish the importance of this
notion for the humanities (philosophers, literary critics and linguists) and how
complementary research among psychologists and “humanistically oriented
scholars” would mean that “future treatments of figurative language should be
significantly enriched” (Garder and Winner 1978: 141). Over three decades later, it
can be said that the treatment of figurative language, particularly that brought
about by Cognitive Linguistics, has certainly been enriched by the complementary
work of applied linguists (e.g. Cameron 2003 or Low 1988), psychologists (Gibbs
1994, 2008; Robinson and Ellis 2008), and cognitive linguists themselves (Lakoff
and Johnson 1980, 1999; or Kövecses 2002, 2005, among others).
When considering specifically the notion of metaphorical competence, some
key contributions have been made, mostly by applied linguists (for a recent re-
view, see Nacey 2017: 504–505). In our view, two important insights have emerged
from this literature: On the one hand, the recognition of the importance for L2
learners to develop this ability at the different stages of their process of acquiring
the target language and, on the other, the idea that metaphorical competence is
not restricted to a particular dimension but, rather, aspects of metaphor in lexis or
discourse need to be addressed by learners in all four dimensions of communica-
tive competence, i.e., linguistic, sociolinguistic, discourse and strategic (c.f.
Littlemore and Low 2006a, 2006b).
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 201

A pioneering scholar in this field is Danesi, who has been working on meta-
phorical competence for the past two decades (1986, 1992, 2008, 2016). For this au-
thor, L2 learners’ control of figurative language, which is what he defines as
metaphorical competence, is essential for them to achieve real L2 proficiency
(Danesi 1986). In his view, the idea of metaphorical competence in Second
Language Acquisition (SLA) somehow overlaps with that of communicative com-
petence and he formulated the notion of ‘Conceptual Fluency’ (Danesi 1992) which
he defined as the learners’ mental capacity to use figurative discourse systemati-
cally. Behind this conception is the implication that the systematic nature of figu-
rative language makes it teachable in the same way as other aspects of the L2.
Low’s (1988: 129) definition of the construct, “a number of skills related to meta-
phor which native speakers are frequently expected to be good at, and which
learners need to develop to some degree if they hope to be seen as competent
users of the language”, also highlights the importance for L2 learners of develop-
ing this ability to be perceived as competent users of the target language.
Littlemore (2001: 461) went a step further in identifying these ‘skills’ and estab-
lished four main components that make up learners’ metaphorical competence:
“(a) originality of metaphor production, (b) fluency of metaphor interpretation, (c)
ability to find meaning in metaphor, and (d) speed in finding meaning in
metaphor.” These two authors together published a comprehensive monograph
(Littlemore and Low 2006b) devoted to figurative thinking and foreign language
instruction in which they highlight the importance of figurative language in the L2
learning process and show that learners have to deal with aspects of metaphor at
all stages of proficiency and in all the different dimensions of communicative com-
petence in the L2 (grammatical, textual, illocutionary and sociolinguistic). All
these accounts of metaphorical competence illustrate the intrinsic complexity of
the construct as well as applied linguists’ concerns for its applicability in the
process of learning a foreign language. These theoretical notions have recently
given rise to some proposals to measure empirically L2 learners’ metaphorical com-
petence. Although there has been an attempt to relate learners’ metaphorical
competence and their overall language proficiency (Teymouri and Dowlatabadi
2014), the bulk of research studies in this area consists of corpus work that has
focused on measuring learners’ production of metaphor (Littlemore’s [2001] first
component, i.e., ‘originality of metaphor production’) in their written outputs. For
example, Nacey (2013) reports a comparison between Norwegian, advanced learn-
ers’ texts extracted from the Norwegian subset of the International Corpus of
Learner English (NICLE) and those produced by British L1 novice writers from the
Louvain Corpus of Native English Essays (LOCNESS), and Turner (2014) analyses
the presence of metaphor in exam-based written production of French and
Japanese learners of English. To the best of our knowledge, one of the most
202 Irene Castellano-Risco and Ana M. Piquer-Píriz

comprehensive and ambitious study into learners’ production of metaphor so far is

Littlemore et al (2014). Using the Cambridge Learner Corpus of exam scripts, they
analysed two hundred essays produced by successful Greek and German learners
of English across five of the levels (A2 to C2) established by the Common European
Framework of Reference for Languages (CEFRL) in order to determine the amount
of metaphor employed at the different stages and the different functions for which
learners used those metaphors. The authors found that metaphor usage increased
as students progressed through the levels, and so did the sophistication of the
function for which learners used metaphors. However, error rates involving meta-
phor remained significantly higher than general error rates, even at the higher lev-
els, and L1 transfer was seen to have an effect on metaphor-related errors. Along
this line of research, Hoang and Boers (2018) have also explored the association of
advanced, EFL learners’ writing proficiency and their use of metaphorical lan-
guage in a more homogenous corpus consisting of two hundred and fifty-two es-
says produced by language majors from three year levels in an undergraduate
programme at a university in Viet Nam. Correct metaphor use was also found to be
strongly correlated with the students’ year levels and the grades awarded to the
essays.
In this chapter, we propose the analysis of a different dimension of metaphori-
cal competence, specifically, the understanding/recognition of metaphor (which
corresponds to Littlemore’s [2001] third dimension: ‘ability to find meaning in met-
aphor’), as we consider that L2 learners’ understanding of metaphorical meanings
is part of their general lexical competence which is defined as the ability to recog-
nize and use the words of an L2 in a native-like way (cf. López-Mezquita 2005).
Studies on lexical competence date back to the beginning of the twentieth century,
and, since early stages, researchers have attempted to identify the different types
of knowledge involved in vocabulary development. In most of the proposals, meta-
phorical competence per se has not been explicitly mentioned, but it has been im-
plicitly included in the aspects analysed since the very first attempts. For example,
Cronbach (1942) distinguished between five main vocabulary knowledge dimen-
sions, one of them being ‘breadth of meaning’ (here ‘breadth’ is not related to vo-
cabulary size but it refers the understanding and managing of those words with
more than one meaning), which made reference to the understanding and manag-
ing of polysemous words. Years later, Anderson and Freebody (1976) encapsulated
this idea of polysemy and full understanding of a word by giving it a new label
-‘depth’- and defining it as follows: “ . . . [There] is a second dimension of vocabu-
lary knowledge, namely the quality or “depth” of understanding. We shall assume
that, for most purposes, a person has a sufficiently deep understanding of a word
if it conveys to him or her all of the distinctions that would be understood by an
ordinary adult under normal circumstances” (1981: 92-93). Meara (1996) went a
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 203

step further and reflected on the importance of the interaction between both di-
mensions (‘breadth’ and ‘depth’), in his case labelled as ‘size’ and ‘organization’,
proposing that, as a learner’s lexicon grows, the importance of size decreases,
while the importance of depth of knowledge or organisation of the lexicon in-
creases. Recently, in a study about L2 learners’ vocabulary knowledge of primary
and peripheral meanings, Aizawa (2018: 31) acknowledged that: “We still do not
know whether the knowledge of words with multiple meanings should be catego-
rized as width or depth of vocabulary knowledge”. But, he concludes, following
Meara, that ‘breadth of knowledge’ is more important for lower-intermediate
learners while ‘depth of knowledge’ is more important at advanced stages.
Finally, Nation (2001) developed the most comprehensive categorization of
lexical competence up to that moment, by identifying three main categories of
word knowledge, namely, form, meaning and use; each of which containing spe-
cific aspects. In the case of the category meaning, he distinguished also between
three aspects of knowledge: form and meaning, concepts and references and the
different associations that learners may have when meeting the word. It is in these
two latter concepts that metaphorical competence could be considered. In addi-
tion, he included a second classification that had been developed in parallel to all
these proposals: the distinction between receptive and productive knowledge.
With this distinction, grounded in the experience of many teachers, and reflected,
among others, in Palmer’s (1921) and Henriksen’s (1999) proposals, the under-
standing of a term and the production of the same term are seen as different pro-
cesses, i.e., it is recognized that a learner may be able to understand or recognize a
word, but at the same time, may not have total control of it and may experience
difficulties in recalling or producing the word. This distinction is also relevant to
the conceptualization of metaphorical competence, because, as Littlemore and
Low (2006b: 46) pointed out “foreign language learners probably need to under-
stand metaphor more often than they need to produce it” and this is particularly
the case at basic levels of proficiency (A1-B1 CEFRL levels), which would corre-
spond to the most common levels of learners of English in mainstream education
in many countries around the world.
In our view, metaphorical competence is strongly linked to a more global ca-
pacity in the sense of Danesi’s ‘conceptual fluency’ or Littlemore and Low’s ‘figu-
rative thinking’ that consists in the ability to use our knowledge of concrete
things to understand abstract concepts. As is well known, this is the basis of the
Cognitive Linguistic paradigm (Lakoff and Johnson 1980, 1999). This natural abil-
ity is conceptual and, therefore, not restricted to the way we reason in our native
language but also applicable to other languages we may happen to speak and is
part of the human reasoning abilities from early infancy (cf. Ö̈zçalışkan 2011). In
204 Irene Castellano-Risco and Ana M. Piquer-Píriz

the early 1990s, Levorato and Cacciari (1992: 416-417) already highlighted the im-
portance, even for children, to develop a ‘figurative competence’. It has been
shown that even very young L2 learners are able to reason metaphorically and
metonymically when confronted with non-literal L2 meanings (Piquer-Píriz 2005,
2008a, 2008b, 2010). These studies show that children as young as five years of
age resort to analogical (metaphorical) or associative (metonymical) reasoning
when asked to explain some of the semantic extensions of specific words in
English whose core meanings they know.
At this point, it is also very important to distinguish between the mental
process and its linguistic products (cf. Gibbs’ [1999] for a seminal discussion on
this issue). Although human beings seem to be endowed with this natural abil-
ity to perceive the underlying motivation of non-literal meanings from early in-
fancy, a direct extrapolation of this strategy to lexical items in an L2 is not
possible because different languages may encode figurative meanings in differ-
ent ways. For instance, the long parts of a watch or clock that move around and
show the time are called ‘hands’ in English. This figurative meaning of the body
part term seems to be motivated by a metonymy-based metaphor where the
shared standard information about one specific function of the human hand,
i.e., pointing or indicating, and the same function of the device of a watch or
clock may have given rise to this semantic extension. In Spanish, the preferred
realisation for the ‘hands of a watch’ is ‘agujas’ (‘needles’) although ‘manecillas’
(literally, ‘little hands’) also exist. The Spanish realisation ‘agujas’ focuses on
shape rather than on function. Boers (2013) makes a fundamental point in rela-
tion to this aspect that may, perhaps, explain why the important findings of the
research literature into the applications of some of the main tenets from CL to
foreign language instruction (see Piquer-Píriz and Alejo-González 2016 for an
overview, and Boers 2011 for an insightful discussion of the contributions to the
field but also the main methodological limitations of these studies) have not
reached L2 classrooms as much as it would be expected: The fact that figurative
meanings are motivated does not mean that they are predictable. But this is,
precisely, according to Boers, why enhancing and promoting figurative aware-
ness can be of help for L2 learners:

It cannot be overemphasised that ‘motivated’ in CL jargon is not synonymous with ‘pre-

dictable’ [. . .] If there were no room for arbitrariness, then all natural languages would
look and sound identical. If the CL endeavour to describe linguistic phenomena as moti-
vated holds a certain promise for language pedagogy, it is precisely because in many
cases the motivation is not blatantly obvious to the learner. (Boers 2013: 211).
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 205

This author has suggested that rather than encouraging learners to generate L2
metaphors, it may be more successful to instil a ‘metaphor awareness’ in learn-
ers to “organize the steady stream of figurative language they are exposed to”
(Boers 2000: 564). In this same line is MacArthur’s (2010) seminal paper on
metaphorical competence in EFL where she argues for the importance of adding
‘metaphor awareness’ generally to the experience of learning English in in-
structed L2 setting:

In this approach, the specific pedagogical techniques employed [. . .] are seen to be less
significant than the general foregrounding of metaphor and the effect(s) this may have on
learners’ growing awareness of how metaphor permeates language (their own and the L2
to be learnt) as reflected in the growing felicity of the metaphorical language used in their
written work. (ibid.: 157)

In our view, any understanding of metaphorical competence as a strategy em-

ployed by L2 learners cannot be seen in isolation from the rest of strategies they
may resort to when confronted with new lexical items in the target language.
However, scant attention seems to have been paid to metaphorical competence in
vocabulary studies. Since the 1990s, there has been a soaring interest in the iden-
tification of the strategies learners use to develop their lexical competence. In this
regard, a large number of taxonomies have been proposed (Stöffer 1995; Gu and
Johnson 1996; Schmitt 1997; Nation 2001; Tseng, Dörnyei and Schmitt 2006; or
Siriwan 2007). Yet, these proposals show little consideration for the development
of metaphorical competence. For example, the two best-known taxonomies
(Schmitt 1997; Nation 2001) mention no strategies to develop learners’ capacity to
understand/produce metaphorical terms. Schmitt (1997) identified a total of fifty-
eight strategies, none of which seems to be specifically focused on aiding the un-
derstanding of figurative meanings. Similarly, Nation (2001) mentioned sixteen
general strategies, and these were all devoted to the expansion of other vocabu-
lary dimensions, neglecting, once again, the metaphorical dimension.
To be fair, it must be acknowledged that, conversely, few metaphor schol-
ars, with the exception of Boers (e.g. 2013), tend to take into consideration re-
search findings from vocabulary studies in mainstream applied linguistics.
This chapter is an attempt to connect two fields that can enrich one an-
other. In order to do so, it presents an exploratory study in which well-known
vocabulary tests that have been validated and are widely used with different
learners in many parts of the world (Schmitt 2010) were employed to measure
secondary-school learners’ general lexical competence as well as their under-
standing/recognition of figurative word meanings.
206 Irene Castellano-Risco and Ana M. Piquer-Píriz

3 The Study: Measuring L2 Secondary School

Learners’ Vocabulary Knowledge: Metaphorical
Competence within Lexical Competence
3.1 Objectives

Two main objectives were established for the study:

1. To measure L2 secondary school learners’ understanding of figurative
meanings in English using a validated test to measure lexical competence.
2. To compare their understanding of figurative meanings with their under-
standing of non-figurative senses.

3.2 Design of the Study

3.2.1 Participants

The study was conducted in a medium-size town of Extremadura (western Spain)

as part of a bigger project (Castellano-Risco, in preparation). A convenience sam-
ple of seventy-seven students (thirty-eight boys and thirty-nine girls) was ana-
lysed. Participants were asked to complete the Vocabulary Levels Tests (VLT) and
the Productive Vocabulary Levels Tests (PVLT) at two points within two different
academic years, the first time when they were in 3rd year (Year 9) and a year
after, when they were in 4th year of Compulsory Secondary Education (Year 10).

3.2.2 Instruments

Two instruments were used in the study. In order to measure their receptive vo-
cabulary size, students were asked to complete the 2K and academic bands of
the VLT (Schmitt, Schmitt and Claphman 2001). Developed by Nation (1983)
and refined by Schmitt et al. (2001), this test has been validated and is one of
the most widely used tests when measuring L2 vocabulary knowledge (Schmitt
2010). It consists of five levels and, in each level, thirty words, grouped in ten
clusters, are presented to the test-takers. In each cluster, there are three defini-
tions and six options, and test-takers are asked to match the definitions with
their corresponding word. The test is designed in a way that minimizes the
guessing aids and all the words, the tested ones and those presented in the def-
initions, belong to the level tested or to lower levels.
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 207

In order to measure productive vocabulary size, the 2K and academic bands

of the PVLT (Laufer and Nation 1999) were used. This set of tests measures the
productive knowledge of five main bands of vocabulary: four frequency levels
and an academic band. In each, eighteen sentences are presented. These senten-
ces include an incomplete word and test-takers are asked to complete the word,
bearing in mind the context provided with the sentence. In spite of being called a
“productive” test, Schmitt (2010) argues that it can be considered a form-recall
test, as test-takers are asked to recall some words, but they are not actually pro-
ducing any lexical item. Similarly, Read (2000) defines it as a “selective” test,
because the target language is pre-selected. From their point of view, the PVLT
cannot be considered a productive test properly, because it does not measure
what is produced by the learners, but how test-takers are able to create connec-
tions with other words.
The VLT and the PVLT are similar in the sense that they measure the same
bands that correspond to the number of word families considered sufficient to
engage in a daily conversation (2,000); and the academic band that measures
knowledge of the words included in the Academic Word List (Coxhead 2000)
and its precedessor, the University World List (Xue and Nation 1984). Moreover,
they also present the same distribution of word classes, following a 3 (noun): 2
(verb): 1 (adjective) ratio (Schmitt, Schmitt and Clapham 2001).

3.2.3 Procedure

Following the objectives that were established for the general project (Castellano-
Risco in preparation), the tests were administered at two different times within
two academic years: March, 2016 (T1) and February, 2017 (T2). At T1, the data col-
lection took one day, but at T2, two days were needed: the first day, the test-
takers answered the VLT and, on the second day, the PVLT was administered.
Students, thus, had to complete six different tests in total, as shown in Table 1.

Table 1: Data collection periods.

Data collection period Tests

Time  (March, ) – Version  of the K and academic bands of the VLT ( tests)

Time  (February, ) – Version  of the K and academic bands of the VLT ( tests)
– Version  of the K and academic bands of the PVLT ( tests)
208 Irene Castellano-Risco and Ana M. Piquer-Píriz

After their administration, the tests were scored and the data were proc-
essed and coded manually. In the scoring process of the PVLT, bearing in mind
that the aim was to explore the learners’ lexical knowledge and not grammati-
cal competence, certain decisions were made in relation to the ‘correctness’ of
the answers provided by the students: Spelling mistakes invalidated the answer
but the use of the wrong tense in the case of verbs or the incorrect choice of a
singular or plural form for nouns was not penalised.
In order to explore the participants’ understanding of figurative mean-
ings, an identification of the metaphorical uses in the tests was carried out.
To do so, the metaphor-identification procedure used was MIPVU (Steen et al.
2010) which has been extensively applied to identify metaphors in research
studies. As is well-known, this procedure is a refined version of the Metaphor
Identification Procedure (MIP) devised by the Pragglejaz group (2007). In
order to apply MIPVU, the analyst has, first of all, to identify the different lexi-
cal items, then, establish their meaning in that specific context and, finally,
decide whether it has a more basic contemporary meaning in other contexts.
If it does, that lexical item is considered to be metaphorical. Three raters ap-
plied the method to the vocabulary items included in the tests listed in
Table 1. A total of one hundred fifty-six items were examined (thirty words per
level and version in the VLT and eighteen words per level in the PVLT), and,
only five metaphorical uses were found: charm (2K band of the VLT, version
1), admire (2K band of the VLT, version 2), highlight (academic band of the
VLT, version 1), limit (2K band of the PVLT) and attain (academic band of the
PVLT). Then, these results were correlated with participants’ overall lexical com-
petence. Finally, the results obtained for these specific terms were compared
with the results obtained for other terms belonging to the same grammatical cat-
egory. Specific details of these analyses are provided in the following section.

4 Results
Starting first with an overall analysis of the global score differences between
the metaphorical and the non-metaphorical terms, we performed a Wilcoxon
test, which is the appropriate statistical measure when the data are not nor-
mally distributed. The test was statistically significant (z = ‒7.272, p =.000) and
showed that non-figurative terms obtained a higher mean score (52.77%) than
metaphorical ones (34.81%). In other words, participants were able to recognise
and produce literal meanings more frequently than metaphorical ones.
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 209

This result was confirmed when we separately analysed the terms in both
the VLT (z = ‒5.975, p = .000) and PVLT (z = ‒5.423, p = .000). With respect to
the results of the VLT, participants understood non-figurative terms (66.87%,
SD = 19.38) better than figurative terms (51.08%, SD = 25.79). As explained
above, in the VLT three items were identified as metaphorical. Two of them,
charm and admire, are found in the 2K band, whereas highlight belongs to the
academic band. To check whether each of these metaphorical terms individually
correlated with the non-figurative total score obtained in the corresponding vo-
cabulary band, a calculation of Spearman’s rho was carried out. The results
show that a correct recognition of the two metaphorical terms identified (charm
and admire) correlates (N= 77; p < .01) with a larger receptive knowledge of the
2K band (charm: r = .346, p = .002; and admire: r = .304, p = .007) and that the
understanding of highlight correlates with a greater recognition of the academic
band (highlight: r = .395, p = .001). In other words, it seems that the recognition
of the metaphorical terms is related to a better performance in the whole test.
Moving on to the metaphorical terms identified in the PVLT, there were only
two items identified as metaphorical: attain belonging to the academic band and
limit to the 2K band. In the analysis of the difference between the global scores
of the metaphorical and non-metaphorical terms, which again was statistically
significant (z = ‒5.423, p = .000), it was also found that the figurative terms
(10.39%, SD = 20.41) were significantly less recalled than the non-metaphorical
ones (31.60%, SD = 18.32).
However, in the detailed analysis of the relationship between each item
and the band to which it belonged, unlike in the previous items, a calculation
of Spearman’s rho only found a positive significant correlation (N= 77; p < .05)
between a correct production of the metaphorical term attain and the global
score of the 2K PVLT (r = .300, p = .008), whereas in the case of limit, no corre-
lation was found (r = .078, p = .499).
In sum, four out of the five metaphorical items correlated positively with
the overall results of the tests, which seems to indicate that recognition or un-
derstanding of figuratively-used words is related to students’ larger vocabulary
size. These results will be taken up in the discussion.
We also analysed the differences between the success rate of metaphorical
items and the rest of terms belonging to the same grammatical category. The
success rate for each of them was calculated by dividing the number of partici-
pants that provided the correct answer by the total number of participants. In
order to do so, each of the participants’ responses was given 0 or 1, where 0
meant that this item was not recognised or produced and 1 meant that the par-
ticipant got the correct answer. Then, the mean score for each item was calcu-
lated (see Table 2).
210 Irene Castellano-Risco and Ana M. Piquer-Píriz

Table 2: Descriptive analysis of learners’ success rate with

figurative terms.

N Mean Std. Deviation

Charm  . .

Admire  . .
Highlight  . .
Limit  . .
Attain  . .

Finally, a Wilcoxon test was carried out to analyse the significance of the differ-
ences. These results will be shown individually for each of the items. To start
with, the success rate of each metaphorical term found in the VLT (charm, admire
and highlight) in comparison to the rest of the words belonging to the same cate-
gory in the same test was examined. First, it was found that the success rate for
charm ðx = .30) was lower than that of the rest of nouns (see Table 3 below). The
analysis of the significance of this difference showed that all the differences be-
tween the success rates of charm and the rest of nouns were significant (p < .001)
with the exception of flesh (z= .000, p = 1.000).

Table 3: Descriptive analysis of the nouns found in the 2K band of the VLT.

Victory Birth Temperature Flesh Journey Education Scale

N       
x . . . . . . .
SD . . . . . . .

Salary Treasure Charm Lack Cream Wealth Pupil

N       
x . . . . . . .
SD . . . . . . .

This was not the case for admire. As can be seen in Table 4 below, it presented in
general a higher, and, in some cases, significantly higher, success rate ðx = .83)
in comparison to the rest of verbs of the test. In this case, it was found that the
success rate of admire was significantly higher than that of strengthen (z= ‒4.116,
p = .000), develop (z= ‒4.333, p = .000), arrange (z= ‒5.145, p = .000), manufac-
ture (z= ‒2.20, p = ‒.028) and elect (z=‒2.111, p = .035).
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 211

Table 4: Descriptive analysis of the verbs found in the VLT v.2.

Strengthen Introduce Admire Develop Arrange Prefer Manufacture Elect Melt

N         
x . . . . . . . . .
SD . . . . . . . . .

However, in the analysis of highlight, an academic metaphorical term also

found in the first version of the test, as in the case of charm, it was found again
that the success rate (x= .40) was lower than with the rest of the verbs. In fact,
the statistical analysis indicated that all the differences between highlight and
the rest of verb terms were statistically significant (p <.001) (see Table 5).

Table 5: Descriptive analysis of the verbs found in the academic band of the VLT v1.

Alter Deny Specify Retain Correspond Highlight Minimize Estimate Identify

N         
x . . . . . . . . .
SD . . . . . . . . .

Moving on to the analysis of the success rate of the metaphorical terms identified
in the productive tests, in the case of limit, the metaphorical term found in 2K
band of the PVLT, a lower rate in comparison to the rest of the items (x= .17) was
observed (Table 6). The significance of this difference between terms was exam-
ined and the Wilcoxon test results showed a significantly lower production of the
term limit in comparison to climb (Z= ‒4.743, p = .000), examine (Z= ‒3.053, p =
.002), and connect (Z= ‒3.402, p = .001).

Table 6: Descriptive analysis of the verbs found in the 2K band of the PVLT.

Climb Examine Surround Connect Wander Limit

N      
x . . . . . .
SD . . . . . .
212 Irene Castellano-Risco and Ana M. Piquer-Píriz

Finally, the analysis of the metaphorical term attain, an academic term in-
cluded in the PVLT, was carried out. Similar to previous results, the success
rate of attain (x = .04) was found to be lower in comparison to the success rate
of the rest of verb terms in the academic band of the PVLT. The statistical analy-
sis showed that the differences between the success rates of attain and the rest
of terms were statistically significant at the .01 level (p < .01) with the exception
of the difference between attain and rely, whose difference was significant at a
.05 level (z= ‒2.138, p = .033).

Table 7: Descriptive analysis of the verbs found in the academic band of the PVLT.

Inspect Accumulate Saturate Rely Evaluate Attain

N      
x . . . . . .
SD . . . . . .

In general, a pattern seems to be found in the differences between the success

rate for the metaphorical terms and the other terms belonging to the same word
class: in two out of the three terms identified in the receptive VLT, and in the
two metaphorical terms found in the PVLT, the metaphorical terms presented a
lower success rate in comparison to other items occurring in the same fre-
quency band and within the same grammatical category of the tests employed.
The only exception was found in the term limit and different reasons could ex-
plain this difference, as will be explained in the next section.
In order to facilitate the interpretation of the results, it is important to con-
sider the participants’ performance overall in the tests. Starting with the 2K
band of the VLT, in the first version, participants obtained a mean score of
58.90% of the words, whereas in the second version, administered one year
later, they got a mean score of 75.47% of the 2K band. In relation to the aca-
demic band, in the first version, participants recognised a total of 48.38% of
the academic terms, whereas in the second version, administered one year
after, they recognised 60.9% of the items. First of all, the mean scores of the
tests were explored. In the case of the 2K band of the PVLT test, learners suc-
cessfully produced 37.55% of the words, a percentage well below that found in
the receptive version of the 2K band, whereas in the academic band, partici-
pants were able to produce a mean of 23.12% of the academic words. This low
performance was also observed in the success rate of the different items. As
will be seen later, this may also explain some of the findings.
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 213

5 Discussion
This study has attempted to present a preliminary analysis of the use of stand-
ardised tools in second language acquisition for estimating L2 vocabulary size
with the objective of measuring L2 learners’ understanding of figurative
meanings.
The results show that, first of all, a general positive correlation between the
recognition or understanding of each metaphorical term and the global score of
the corresponding test was found. This seems to indicate that the mastery of
the metaphorical terms is clearly related to that of general vocabulary items.
The only exception is found in the term limit, and the lack of connection be-
tween the metaphorical term and the global score of, in this case, the 2K band
of the PVLT could be related to two factors: on the one hand, participants were
able to produce 30.07% of the words, which is well below the performance in
the receptive tests. This low performance is in line with other studies (Fan
2000; Laufer 1998; Laufer and Paribakht 1998; Melka 1997; Takala 1984) that
conclude that leaners’ productive mastery of vocabulary items lags behind their
receptive knowledge of them. On the other hand, if the metaphorical term limit
presents an unusual higher success rate in comparison to the other terms in the
test, this is probably due to its resemblance to its Spanish equivalent (lit. limi-
tar) and thus reflects a likely cognate effect. When dealing with this term, it is
possible that the learners resorted to their mother tongue aiming to look for
some kind of help. This fact may have produced a higher capacity of production
of the word, which, together with a low performance in general in the test,
could have led to this lack of difference with other terms.
Considering the results obtained from the analysis of a difference between
the success rate of each metaphorical item and other terms belonging to the
same grammatical category and test, it has been found that, in four out of the
five items examined, the success rate for the metaphorical term is significantly
lower in comparison to other terms in the same test. This finding is in line with
the outcomes obtained by Littlemore et al. (2014), who, as has been noted above,
also identified a greater proportion of errors in metaphorical items when com-
pared to non-metaphorical terms. This outcome seems to corroborate that partic-
ipants find it more challenging to understand the metaphorical items and this
leads to a lower rate of recognition or production. However, in this analysis, an
exception is found: admire, which, in fact, presented a greater success rate than
other words with the same category and test. This finding may be related, again,
to the particularities of the term: admire, like limit, is also a cognate in Spanish
(literally, admirar) and it could be expected that learners find this term easier to
214 Irene Castellano-Risco and Ana M. Piquer-Píriz

recognise. It seems that cognation seems to work as a facilitator, not only in gen-
eral acquisition of vocabulary, but also in the case of metaphorical items.

6 Conclusion
After over three decades of research into the notion of metaphorical competence,
it can be said that, despite the complexity of the construct, important progress
has been made into our understanding of this concept, particularly, from a theo-
retical perspective and in terms of identifying its importance in foreign language
instruction. As has been noted above, recent research studies in the field (Hoang
and Boers 2018; Littlemore et al. 2014; Nacey 2013; Turner 2014) have focused on
measuring learners’ production of metaphor providing insightful accounts of this
dimension of learners’ metaphorical competence. However, the receptive domain
has not received so much attention despite the fact that it would seem that L2
learners need to understand figurative meanings more often than they need to
produce them and this is particularly so at the basic levels of L2 proficiency (A1-
B1) which would correspond to the most common levels for most learners of
English in mainstream education worldwide.
In this chapter, we have focused on this somehow neglected dimension by
trying to bridge a gap between general vocabulary studies and metaphor re-
search. Especially relevant would be the design of instruments that would mea-
sure L2 learners’ capacity to understand and produce figurative meanings,
either with the inclusion of metaphorical items within the general vocabulary
tests, or with the development of tests specifically designed for this aim. Our
exploratory study is a first attempt in this regard but this area is in need of fur-
ther research that uses insights and tools from both fields that would enrich
our understanding of metaphorical competence as part of L2 learners’ general
lexical competence.

References
Aizawa, Kazumi. 2018. Exploring L2 learners’ vocabulary knowledge of primary and peripheral
meanings. Research Bulletin 77. 31–40
Anderson, Richard C. & Peter Freebody. 1981. Vocabulary knowledge. In John T. Guthrie (Ed.),
Comprehension and teaching: Research reviews, 77–117. Newark, D.E.: International
Reading Association.
Boers, Frank. 2000. Metaphor awareness and vocabulary retention. Applied Linguistics 21.
553–571.
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 215

Boers, Frank. 2008. How cognitive linguistics can foster effective vocabulary teaching. In
Frank Boers & Seth Lindstromberg (eds.), Cognitive linguistic approaches to teaching
vocabulary and phraseology, 1–61. Berlin: Mouton de Gruyter.
Boers, Frank. 2011. Cognitive semantic ways of teaching figurative phrases: An assessment.
Review of Cognitive Linguistics 9. 227–261.
Boers, Frank. 2013. Cognitive linguistic approaches to teaching vocabulary: Assessment and
integration. Language Teaching 46. 208-224.
Boers, Frank & Murielle Demecheleer. 2001. Measuring the impact of cross-cultural
differences on learners’ comprehension of imageable idioms. ELT Journal 55. 255–262.
Cameron, Lynne. 2003. Metaphor in educational discourse. New York: Bloomsbury Continuum.
Cameron, Lynne & Graham D. Low. 1999. Researching and applying metaphor. Cambridge:
Cambridge University Press.
Castellano-Risco, Irene. forthcoming. The impact of vocabulary learning strategies on
vocabulary size: a study of secondary learners in Content and Language Integrated
Learning programmes. Badajoz: University of Extremadura PhD dissertation.
Cronbach, Lee J. 1942. Measuring knowledge of precise word meaning. The Journal of
Educational Research 36(7), 528–534.
Coxhead, A. 2000. A new Academic Word List. TESOL Quarterly 34(2). 213–239.
Danesi, Marcel. 1986. The role of metaphor in second language pedagogy. Rassegna Italiana
di Linguistica Applicata 18. 1–10.
Danesi, Marcel. 1992. Metaphorical competence in second language acquisition and second
language teaching: The neglected dimension. In James E. Alatis (ed.), Language
communication and social meaning (Georgetown University Round Table on Languages
and Linguistics), 489–500. Washington: Georgetown University Press.
Danesi, Marcel. 2008. Conceptual errors in second-language learning. In Sabine De Knop &
Antoon De Rycker (eds.), Cognitive approaches to pedagogical grammar, 231–256.
Berlin: Mouton de Gruyter.
Danesi, Marcel. 2016. Conceptual fluency in second language teaching: An overview of
problems, issues, research findings and pedagogy. International Journal of Applied
Linguistics & English Literature 5(1). 145–153.
Fan, May. 2000. How big is the gap and how to narrow it? An investigation into the active and
passive vocabulary knowledge of L2 learners. RELC Journal 31. 105–119.
Gardner, Horward & Ellen Winner. 1978. The development of metaphoric competence:
Implications for Humanistic disciplines. Critical Inquiry 5(1). 123–141.
Gibbs, Raymond W. 1994. The poetics of mind: Figurative thought, language, and
understanding. Cambridge: Cambridge University Press.
Gibbs, Raymond W. 1999. Researching metaphor. In Lynne Cameron & Graham D. Low (eds.),
Researching and Applying Metaphor, 29–47. Cambridge: Cambridge University Press.
Gibbs, Raymond W. 2008. The Cambridge handbook of metaphor and thought. Cambridge:
Cambridge University Press.
Gu, Yongqi & Robert K. Johnson.1996. Vocabulary Learning Strategies and Language Learning
Outcome. Language Learning 46(4). 643–679.
Henriksen, Birgit. 1999. Three dimensions of vocabulary development. Studies in Second
Language Acquisition 21(2). 303–317.
Hoang, Ha & Frank Boers. 2018. Gauging the association of EFL learners’ writing proficiency
and their use of metaphorical language. System 74. 1–8.
216 Irene Castellano-Risco and Ana M. Piquer-Píriz

Kövecses, Zoltán. 2002. Cognitive-linguistic comments on metaphor identification. Language

and Literature 11(1). 74–78.
Kövecses, Zoltán. 2005. Metaphor in culture. Cambridge: Cambridge University Press.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago
Press.
Lakoff, George & Mark Johnson. 1999. Philosophy in the flesh: The embodied mind and its
challenge to western thought. New York: Basic Books.
Laufer, Batia. 1998. The development of passive and active vocabulary in a second language:
Same or different? Applied Linguistics 19(2). 255–271.
Laufer, Batia & Paul Nation. 1999. A vocabulary-size test of controlled productive ability.
Language Testing 16(1). 33–51.
Laufer, Batia & Taherek S. Paribakht. 1998. The relationship between passive and active
vocabularies: effects of Language Learning context. Language Learning 48(3). 365–391.
Levorato, Chiara & Cristina Cacciari. 1992. Children’s comprehension and production of
idioms: the role of context and familiarity. Journal of Child Language 19(2). 415–433.
Littlemore, Jeannette. 2001. Metaphoric competence: A possible language learning strength of
students with a holistic cognitive style? TESOL Quarterly 35(3). 459–491.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sara Turner. 2014. An investigation into
metaphor use at different levels of second language writing. Applied Linguistics 35(2).
117–144.
Littlemore, Jeannette & Graham D. Low. 2006a. Metaphoric competence and communicative
language abilities. Applied Linguistics 27 (2). 268–294.
Littlemore, Jeannette & Graham D. Low. 2006b. Figurative thinking and foreign language
learning. Basingstoke: Palgrave Macmillan.
López-Mezquita, María T. 2005. La evaluación de la competencia léxica: test de vocabulario,
su fiabilidad y validez. Universidad de Granada: Universidad de Granada PhD
dissertation.
Low, Graham D. 1988. On teaching metaphor. Applied Linguistics 9(2). 125–147.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where are we and where should we
be going? A view from the language classroom, Applied Cognitive Linguistics in Second
Language Learning and Teaching. AILA Review 23. 155–173.
MacArthur, Fiona & Ana M. Piquer-Píriz. 2007. Staging the introduction of figurative
extensions of familiar vocabulary items in EFL: Some preliminary considerations. Ilha de
Desterro: A Journal of English Language, Literatures in English and Cultural Studies 53.
123–134.
Meara, Paul. 1996. The dimensions of lexical competence. In Gillian Brown, Kristen Malmkjaer
& John Williams (eds.), Performance and competence in second language acquisition,
35–53. Cambridge: Cambridge University Press.
Melka, Francine. 1997. Receptive vs. productive aspects of vocabulary. In Norbert Schmitt &
Michael McCarthy (eds.), Vocabulary: Description, acquisition, and pedagogy, 84–102.
Cambridge: Cambridge University Press.
Nacey, Susan. 2013. Metaphors in learner English. Amsterdam: John Benjamins.
Nacey, Susan. 2017. Metaphor comprehension and production in a second language. In Elena
Semino & Zsófia Demjén (eds.), The Routledge handbook of metaphor and language,
503–515. Abingdon and New York: Taylor & Francis.
Nation, Paul. 1983. Testing and teaching vocabulary. Guidelines 5(1). 12–25.
 Measuring Secondary-School L2 Learners Vocabulary Knowledge 217

Nation, Paul. 2001. Learning vocabulary in another language. Cambridge: Cambridge

University Press.
Özçalışkan, Şeyda. 2011. Acquisition of metaphor. In Patrick Hogan (ed.), Cambridge
encyclopedia of language sciences, 486–488. Cambridge: Cambridge University Press.
Palmer, Harold E. 1921. The principles of language-study. London: Harrap.
Pragglejaz Group. 2007. MIP: A method for identifying metaphorically-used words in
discourse. Metaphor and Symbol 22, 1-40.
Piquer Píriz, Ana M. 2005. La comprensión de algunas extensiones semánticas de los lexemas
"hand", "mouth" y "head" en las primeras etapas del aprendizaje del inglés. Cáceres:
Servicio de Publicaciones de la Universidad de Extremadura.
Piquer-Píriz, Ana M. 2008a. Reasoning figuratively in early EFL: some implications for the
development of vocabulary. In Frank Boers & Seth Lindstromberg (eds.), Cognitive
linguistic approaches to teaching vocabulary and phraseology, 233–257. Berlin: Mouton
de Gruyter.
Piquer-Píriz, Ana M. 2008b. Young learners’ understanding of figurative language. In
Mara S. Zanotto, Lynne Cameron & Marilda C. Calvacanti (eds.), Confronting
metaphor in use: An applied linguistic approach, 183-198. Amsterdam: John
Benjamins.
Piquer-Píriz, Ana M. 2010. Can people be cold and warm? Developing understanding of
figurative meanings of temperature terms in early EFL. In Graham D. Low, Zazie Todd,
Alice Deignan & Lynne Cameron (eds.), Researching and applying metaphor in the real
world, 21–33. Amsterdam: John Benjamins.
Piquer-Píriz, Ana M. 2011. Motivated word meanings and vocabulary learning: The polysemy of
hand in the English for Young Learners classroom. Metaphor and the Social World 1(2).
154-173).
Piquer-Píriz, Ana M. & Rafael Alejo-González. 2016. Applying cognitive linguistics: Identifying
some current research foci (figurative language in use, constructions and typology).
Review of Cognitive Linguistics 14 (1).1-20.
Read, John. 2000. Assessing vocabulary. Cambridge: Cambridge University Press.
Robinson, Peter & Nick C. Ellis. 2008. Handbook of cognitive linguistics and second language
acquisition. New York: Routledge.
Schmitt, Norbert. 1997. Vocabulary learning strategies. In Norbert Schmitt & Michael
McCarthy (eds) Vocabulary: Description, acquisition and pedagogy, 199–227. Cambridge:
Cambridge University Press.
Schmitt, Norbert. 2010. Researching vocabulary: A vocabulary research manual. Basingstoke:
Palgrave Press.
Schmitt, Norbert, Diane Schmitt & Caroline Claphman. 2001. Developing and exploring the
behaviour of two new versions of the Vocabulary Levels Tests. Language Testing 18(1).
55–88.
Steen, Gerard J., Aletta G. Dorst, Berenike Herrmann, Anna A. Kaal, Tina Krennmayr & Trijntje
Pasma. 2010. A Method for linguistic metaphor identification: From MIP to MIPVU.
Amsterdam: John Benjamins.
Siriwan, Mayuranee. 2007. English Vocabulary learning strategies employed by Rajabhat
University Students. Nakhon Ratchasima: Suranaree University of Technology PhD
dissertation.
218 Irene Castellano-Risco and Ana M. Piquer-Píriz

Stöffer, Ilka. 1995. University foreign language students’ choice of vocabulary learning
strategies as related to individual difference variable. Alabama: University of Alabama
PhD dissertation.
Takala, Sauli. 1984. Evaluation of students’ knowledge of English vocabulary in the Finnish
comprehensive school. Jyväskylä, Finland: Institute of Educational Research.
Tseng, Wenta T., Zoltán Dörnyei, & Norbert Schmitt. 2006. A New approach to assessing
strategic learning: The case of self-regulation in vocabulary acquisition. Applied
Linguistics 27(1). 78–102.
Teymouri Aleshtar, Maryam & Hamidreza Dowlatabadi. 2014. Metaphoric competence and
language proficiency in the sameboat. Procedia - Social and Behavioral Sciences
(Proceedings of the International Conference on Current Trends in ELT) 98. 1895–1904.
Turner, Sarah. 2014. The development of metaphoric competence in French and Japanese
learners of English. Birmingham: University of Birmingham PhD dissertation.
Xue, Guo-yi & Paul Nation. 1984. A university word list. Language Learning and
Communication 3(2). 215–229.
Part II. 2. Fostering Knowledge of L2 Figurative
Language
Xinqing Wang, Frank Boers and Paul Warren
Using Literal Underpinnings to Help
Learners Remember Figurative Idioms:
Does the Connection Need to Be
Crystal Clear?
1 Introduction
Several idiom dictionaries for language learners include information on the
origins or literal underpinnings of idioms (e.g., American Heritage Dictionary of
Idioms 2003; Collins Cobuild Idioms Dictionary 2012; Oxford Idioms Dictionary for
Learners of English 2009 and https://idioms.thefreedictionary.com/). This sug-
gests that many dictionary makers assume that this kind of information must be
helpful or at least appealing to learners. Its actual benefits for language learners
have not yet been thoroughly evaluated, however. One proclaimed benefit
(e.g., Liontas 2017) is that information about the origins of idioms helps learn-
ers to appreciate the connection between metaphorical language and culture.
Another proclaimed benefit – and the focus of the present chapter – is that
awareness of their literal underpinning can aid learners’ retention of the idi-
oms (e.g., Boers, Demecheleer, and Eyckmans 2004). This expectation rests
on at least three theories.
One of these theories is that knowledge of the literal underpinning of idioms
renders their meaning more transparent. For example, taking a back seat in a vehi-
cle implies that someone else will be at the steering wheel and will thus likely be
in control. By analogy, then, the meaning of the idiom take a back seat to refer to
one’s non-determining role in a project or activity makes sense against the back-
drop of a more generic conceptual metaphor (Lakoff and Johnson 1980), according
to which projects or activities are likened to journeys. In other words, the idiomatic
meaning is motivated, to use a term from cognitive linguistics. If this indeed results
in a clearer link between the form and the idiomatic meaning of the expression, it
may be expected to facilitate acquisition (e.g., Steinel, Hulstijn, and Steinel 2007).

Xinqing Wang, School of Linguistics and Applied Language Studies, Victoria University of
Wellington, Wellington, New Zealand, e-mail: [email protected]
Frank Boers, Faculty of Education, The University of Western Ontario, London, Canada,
e-mail: [email protected]
Paul Warren, School of Linguistics and Applied Language Studies, Victoria University of
Wellington, Wellington, New Zealand, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-011
222 Xinqing Wang, Frank Boers and Paul Warren

The second theory is Paivio’s (1986) Dual Coding Theory, according to which
concrete concepts (or lexical items with concrete meaning) are easier to remember
than abstract ones owing to their imageability, i.e., their association with a mental
image of the referent. Although idioms have abstract meanings, resuscitating the
context in which they were originally used in a literal sense is likely to evoke im-
ages of concrete scenes, and this may thus render them memorable.
The third relevant theory is Levels of Processing theory (e.g., Cermak and Craik
1979), according to which “deep” processing creates stronger memories than “shal-
low” processing. Mental operations – called elaborations – that build rich semantic
associations around lexical items are considered deep in this model. Connecting
the meaning of an idiom to its literal underpinning qualifies as an example of this,
and the label coined by Boers et al. (2004) for this particular type of elaboration is
etymological elaboration. The term etymological should be interpreted here broadly
as the original context in which the expression was (and sometimes still is) used
literally. We shall call this the literal underpinning of an idiom.
One may justifiably wonder whether language learners truly experience infor-
mation about the literal underpinning of an idiom as helpful. There is evidence,
however, that many learners presented with L2 idioms spontaneously activate im-
ages related to the literal meaning of constituent words (e.g., Cieślicka 2006). This
inclination seems different from most native speakers, where the experimental ev-
idence suggests that idiomatic meaning is by default accessed directly (e.g.,
Siyanova-Chanturia, Conklin, and Schmitt 2011), although imagery can be trig-
gered also in L1 speakers of a language (e.g., Gibbs, Nayak, and Cutting 1989). If
it is true that many learners are inclined to activate imagery when they process L2
idioms, then pointing them to the “right” literal underpinning of an idiom (or, at
least, a literal reading that is congruent with the figurative meaning of the idiom)
may harness an inclination that is already present. That pointing learners to that
underpinning will often be necessary is illustrated in the following section.

2 Obstacles to Learners’ Autonomous

Recognition of the Literal-Figurative Link
Learners face diverse challenges should they try to establish a connection be-
tween the literal and figurative meanings of idioms. For starters, the idiom may
contain a key content word that is simply not yet known by the learner, as would
likely be the case for low-frequency words (e.g., in the doldrums; get short shrift;
at the end of one’s tether; bury the hatchet; pass on the baton). In addition, con-
tent words may look deceptively transparent owing to homonymy. For example,
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 223

learners may assume suit in follow suit refers to clothing (while it refers to playing
cards) and they may assume wake in the wake of relates to being awake. In a sim-
ilar vein, polysemy may cause misinterpretations. For instance, gun in jump the
gun could easily be misinterpreted as a weapon (rather than a starting pistol)
and wings in waiting in the wings could be mistaken for a bird’s wings (rather
than a theatre’s wings). It is also possible for learners to mistake new words for
already known ones because of formal resemblance (e.g., reign for rein in on a
tight rein, limp for limb in out on a limb, and tower for towel in throw in the towel).
Because the wording of idioms is often elliptic, learners may have insufficient
clues to work out a plausible literal reading. For example, it is unclear what thin
and thick refer to in through thick and thin (allegedly referring originally to mak-
ing one’s way through thick and thin bushes).
Even if a learner does recognize the literal meaning of an expression, this
does not guarantee an accurate connection between that literal reading and the
actual idiomatic meaning. For example, literally having a lot on one’s plate may
well be experienced as a good thing (especially if one is feeling hungry), and so
the more negative meaning of the idiom (‘being very busy; having a lot of work’)
seems not to follow logically from this. Besides, understanding the literal under-
pinning of idioms may require cultural knowledge that the learner may not yet
have. This regards culture-specific experiential domains. For instance, a learner
may be unfamiliar with sports such as cricket and golf, and so fail to recognize
the underpinning of hit someone for six and be par for the course. It also concerns
knowledge of historical events, myths, legends and fables behind expressions
such as cross the Rubicon, hang on a thread, a Trojan horse and a white elephant.
Moreover, even familiar-looking content words may prompt associations that are
not shared cross-culturally. For example, the heart is conventionally referred to
in western culture as the seat of emotions such as romantic love, while reason
supposedly resides in the mind. This duality seems absent from Chinese, how-
ever, where the heart can symbolize both reason and emotion. It is not surpris-
ing, then, that Chinese learners of English find idioms containing the words
heart or mind comparatively hard to interpret (Hu and Fong 2010).
It also needs to be acknowledged that the literal-figurative link can simply be
too obscure for the literal meaning to serve as a clue for interpretation. Examples
here include idioms such as a red herring, kick the bucket and break a leg. The
origins proposed in dictionaries (e.g., that the distracting scent of smoked herring
was used to train hunting dogs) may consequently be felt by learners to be rather
far-fetched.
The general question raised by these observations is whether all literal-
figurative links can be expected to be helpful. More specifically, the question
we attempt to answer in the study reported in the present chapter is whether
224 Xinqing Wang, Frank Boers and Paul Warren

the degree of transparency of the literal-figurative connection matters for the

mnemonic effect of etymological elaboration. First, however, we need to review
a small number of earlier experiments that have evaluated this instructional ap-
proach to idiom learning.

3 Some Earlier Experiments

A number of studies have already shown that raising learners’ awareness of
the literal senses of polysemous words (e.g., hurdles referring to obstacles for
athletes to jump over in track sports; soar in the sense of physical upward mo-
tion) benefits their comprehension and retention of the abstract or figurative
uses of said words (e.g., hurdles referring to problems to be overcome; soar to
denote upward trends in economics) (see Boers 2013, for a review). Focusing
more specifically on the use of literal underpinnings in teaching figurative idi-
oms, Boers et al. (2004) reported two experiments conducted with computer-
aided exercises, where each idiom was tackled in three exercise components.
In one component, learners were presented with the idiom in isolation and
asked to choose the most likely domain of origin. For example, when pre-
sented with jump the gun, they could choose between source domains such as
sports, war, jurisdiction, etc. Following this, the literal underpinning was dis-
played for learners (e.g., an athlete who jumps the gun in a racing contest sets
off before the starting pistol has been fired). The second component asked
learners to choose the correct figurative meaning of the expressions. For jump
the gun, for example, they were given the choice between (a) defend someone
at your own risk, (b) do something before the appropriate time, and (c) be
startled by an unexpected event. The correct choice was subsequently pointed
out to them. The third component presented learners with a completion task,
where the idiom was incorporated in a meaningful context and a missing con-
tent word needed to be supplied. For example, “Although we had agreed not
to tell anyone about my pregnancy yet until we were absolutely sure, my hus-
band jumped the _______ and told his parents straightaway.” This exercise
was used as a post-test to evaluate the effectiveness of the preceding steps.
The first experiment reported in Boers et al. (2004) aimed to assess the mne-
monic effect of presenting the literal underpinning of the idioms. In one condition
learners tackled the multiple-choice exercise on origins followed by feedback,
and they were then asked to do the completion exercise, i.e., they skipped the
exercise on the actual figurative meaning of the idioms. In the other condition,
the learners did the multiple-choice exercise on the figurative meaning of the
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 225

idioms and they were then asked to do the completion exercise. The former condi-
tion was found to yield the better scores in the completion exercise, which the
authors took as evidence of the mnemonic effectiveness of etymological elabora-
tion. Of particular relevance for the present chapter is the distinction which Boers
et al. then made between idioms whose source domains were identified readily in
the multiple-choice exercise and those which were not. In the case of idioms
whose source domains were hard to guess, recall turned out not to be better after
etymological elaboration than in the comparison condition.
In the second experiment reported in Boers et al. (2004), all participants
tackled the three exercise components of the program, in the order of (1) the
multiple-choice exercise on the figurative meaning of the idioms (followed by
feedback), (2) the multiple-choice exercise on the idioms’ origins, and (3) the
completion exercise, which served as a post-test again. This time, the average
post-test scores for idioms whose source domains were hard to guess were very
similar to those for idioms with better guessable source domains.
Leaving aside the inconsistent nature of the findings reported in Boers
et al.’s (2004) two experiments, what the findings do not tell us is whether it mat-
ters if learners find the proposed motivation of an idiom’s meaning transparent
after it is presented to them. For example, learners may understandably fail to
guess that jump the gun originates from track sports, but they might nonetheless
find that the latter explanation makes good sense when it is subsequently given
to them. By contrast, in a case such as follow suit they may not only fail to guess
that the expression originates from a card game, but perhaps also find the infor-
mation about this origin non-illuminating if they happen to be unfamiliar with
that card game. In the experiment reported below, we therefore asked learners to
evaluate the connection between the proposed literal underpinning of idioms
and their figurative meaning after these were both explained to them. A week
later, the participants were asked to recall the meaning of the idioms, and so we
could examine if there was an association between each learner’s ability to recall
an idiom and their appreciation of its proposed underpinning.
First, however, we need to point out that not all experimental evidence to
date has been favourable of etymological elaboration. Szczepaniak and Lew
(2011) asked upper-intermediate EFL learners to study booklets with informa-
tion about 18 idioms for ten minutes under one of four presentation condi-
tions, with differing amounts of information. The minimal condition consisted
of a definition of the idiomatic meaning and an example sentence. The richest
condition included, in addition, a picture of a literal interpretation of the ex-
pression and an etymological note. The learners were subsequently tested on
their recall of the form of the idioms (by means of a completion task where
they were required to supply a missing content word) and on their recognition
226 Xinqing Wang, Frank Boers and Paul Warren

of the meaning of the idioms (by means of a multiple-choice task). The test
was administered again three weeks later. While both the immediate and the
delayed post-tests furnished evidence in favour of adding pictures to the defi-
nitions of idiomatic meaning, neither produced evidence of the effectiveness
of etymological notes. It could be argued that the ten minutes of study time
allowed in Szczepaniak and Lew (2011) was perhaps insufficient for the partic-
ipants who were presented with the (longish) etymological notes. It takes
much less time and effort to take in a picture than a text passage, after all
(Boers, Warren, Grimshaw, and Siyanova-Chanturia 2017). Because this was
exclusively a pen-and-paper experiment, it is not known to what extent the
participants engaged with the etymological notes that were included in the
booklets. If they did not actually take in the information provided there, then
it would not be surprising that it left no impact. In our experiment, we there-
fore used a one-on-one interview procedure to make sure the participants did
consider the information given about the idioms’ literal underpinnings.
A second possible account for the lack of evidence in favour of the etymologi-
cal elaboration in Szczepaniak and Lew’s experiment could be that a fair num-
ber of the target idioms used (e.g. a white elephant, a loose cannon, a red
herring and have an axe to grind) were arguably ones where the learners
found it hard to follow the explanations for the connection between the origin
and the actual idiomatic meaning. Szczepaniak and Lew did not examine
(through an analysis at the level of individual items) whether etymological
notes perceived to be relatively transparent by the participants were the more
helpful ones. In the study we report next, we did examine this.
Something else that is largely missing from the available body of research on
L2 idiom learning is the potential role of individual learner traits. Exceptions are
explorations of the role of (young) age on L2 metaphor comprehension (Piquer-
Piriz, 2008) and explorations of so-called cognitive style differences among learn-
ers when they are taught idioms with the etymological elaboration technique
(Boers, Eyckmans, and Stengers 2006). It is conceivable that some learners expe-
rience etymological elaborations as more useful than other learners do and are
thus more inclined to appreciate the potential mnemonic potential of such elabo-
rations even in cases where they seem far-fetched. By contrast, for learners who
find it harder to appreciate the connection between an idiom’s meaning and its
proposed literal underpinning, the mnemonic benefits of etymological elabora-
tion may be confined mostly to relatively transparent connections. What seems
worth exploring, then, is whether such predispositions might be related to L2 stu-
dent profiles more generally. For example, one may wonder if high achievers
within a given student population reap the mnemonic benefits of etymological
elaboration more readily than their comparatively low-achieving peers. This
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 227

possibility appears compatible with findings from research on language-learning

aptitude (Robinson 2013; Skehan 2015), where one of the known predictors of
learning success is associative memory ability. High achievers are likely to have
an advantage in this regard, and, in theory, this could also apply to their remem-
bering the meaning of an idiom in association with additional information, such
as its literal underpinning. The possibility that high achievers might benefit the
most from etymological elaborations is also compatible with research on vocabu-
lary learning strategies (Gu 2003, 2013) which suggests that it is individuals who
welcome a wide range of strategies who tend to be relatively successful learners.
In addition, the high achievers within a student population are also likely to
have built a larger L2 vocabulary and to have developed greater familiarity with
the L2 in general. If so, they will be more likely to be familiar with the lexical
constituents of new idioms, and this could reduce the learning burden. It is now
relatively well established that the ability to learn new L2 lexical items tends to
increase as one becomes more proficient in the language (e.g., Elgort and Warren
2014), which is sometimes referred to as the Matthew effect (i.e., the rich get
richer faster). In sum, there are indeed grounds for hypothesizing that, within an
otherwise homogeneous population of L2 students, those who have been com-
paratively successful may be more inclined to engage with and remember the et-
ymological elaborations proposed by a teacher, regardless of whether some of
these elaborations seem far-fetched. An evaluation of this hypothesis will there-
fore be part of the present study as well.

4 The Present Study

4.1 Method

4.1.1 Participants

The participants (N = 25), 22 females and 3 males aged between 19 and 22 (medium
age = 20), were Chinese EFL learners majoring in English at a university in main-
land China. They were all in their third year of study at the university. They shared
very similar histories of EFL learning and were all considered to have intermedi-
ate-to-high proficiency in English. They thus made up a rather homogeneous pop-
ulation. They had all passed the Test for English Majors 4 (TEM-4) before the end
of their second year at the university. This criterion-referenced test is widely used
in China to gauge the English proficiency of university undergraduate English ma-
jors in accordance with the National College English Teaching Syllabus for English
228 Xinqing Wang, Frank Boers and Paul Warren

Majors. The mean grade of the participants on this test was 75.72 (SD 6.52), and the
grades ranged from 61 to 89 (and showed a normal distribution). According to the
TEM descriptors, a grade of 60–69 qualifies as a pass grade, one of 70–79 is con-
sidered a good grade, and one of 80 or above is an excellent grade. So, although
the sample of participants was homogeneous in terms of L1 background and EFL
learning history, the TEM-4 grades demonstrate different levels of EFL achieve-
ment. Whether this made a difference to their performance in the actual study will
be explored further below.
The participants were informed the study was about strategies for learning
English idioms, but the precise purpose was only explained to them after data
collection was completed. They all gave written consent for their data to be
used for this research project.

4.1.2 Instruments and procedure

The idioms presented to these participants were semi-randomly selected from dic-
tionaries such as Collins Cobuild Idioms Dictionary (2012), Oxford Idioms Dictionary
for Learners of English (2013), and American Heritage Dictionary of idioms (2003).
These were all expressions whose dictionary entries included notes about their ori-
gins (not all entries do) and/or for which origins are proposed in online resources
(e.g., the Phrase Finder on https://www.phrases.org.uk/ and the Free Dictionary on
https://idioms.thefreedictionary.com/) (retrieved in September 2016). As we were
interested in the role of item properties, we needed a substantial number of idioms.
Altogether, 80 idioms without close equivalents in the learners’ L1 (Mandarin)
were selected. Because interviewing each participant about so many idioms was
deemed unrealistic, the collection of idioms was divided into four equal sets of 20
idioms each. Each set was then used in interviews with separate groups of six (and
in one case, seven) participants. For motivational reasons, we also added a small
number of idioms that do have close equivalents in Mandarin and that were thus
easier to guess the meaning of, but these were excluded from the analysis.
The two interviews were conducted a week apart. During the first interview
(which took about an hour on average), the learner was informed about the mean-
ing and the literal underpinning of the idioms and was asked to evaluate the clar-
ity of the link between the two. The second, shorter, interview served to determine
how well the participants remembered the meaning of the idioms, and if they re-
membered this in conjunction with the literal underpinnings they had been told
about. The interaction in both interviews was in the participants’ L1 whenever
they preferred so. All the interviews were audio-recorded and transcribed. The
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 229

Mandarin excerpts were translated into English by the first author (who is a native
speaker of Mandarin) to facilitate inter-rater reliability procedures (see below).
The first interview proceeded along the following steps per idiom:
1. The idiom was presented in isolation (e.g., pull one’s weight) and the partici-
pant was asked if he or she was familiar with it, and, if so, to give its mean-
ing. The purpose of this step was merely to ascertain that the participant was
not yet familiar with the idiom. If a participant did give evidence he or she
already knew the idiom, the interview moved on to the next idiom on the list.
2. The idiom was presented in a short sentential context without revelatory se-
mantic clues (e.g., He needs to pull his weight), and the interviewee was asked
to hazard a guess at the meaning of the expression. Embedding the idiom in
a sentence served to illustrate at least the word class of its constituents and
thus made the interpretation task more realistic than guessing the meaning of
a completely decontextualized expression. At the same time, keeping the sen-
tential context minimal was a way of avoiding variability in the interpretabil-
ity of the idioms owing to a factor (i.e., the availability of contextual clues)
extraneous to the semantic transparency of the idioms as such.
3. The origin of the expression was explained and then the interviewee was
asked to make another guess at the figurative meaning of the idiom (or if the
new information confirmed their earlier interpretation). The information
about the origin of the idioms was taken from dictionaries or online resour-
ces. For example, according to American Heritage Dictionary of Idioms, pull
one’s weight comes from rowing, where all crew members are expected to
pull on their oar(s) to give momentum to the boat. In case more than one ori-
gin was proposed by different sources, the one that seemed the most plausi-
ble was chosen.
4. The researcher then explained the meaning of the idiom given in the dictio-
nary (or confirmed the interpretation proposed by the participant in the
previous step, if that happened to be correct). The idiom was then also pre-
sented in a slightly more elaborate context to illustrate its meaning and use
(e.g., If he doesn’t start pulling his weight, he’ll lose his job). These example
contexts were borrowed mostly from the dictionaries. After both the literal
underpinning and the actual idiomatic meaning of the expression were
clarified, the interview moved on to the next idiom in the set.

After their set of 20 idioms had been tackled following the above steps, the par-
ticipants were given a list of them, preceded by the following instruction:
You may feel that the figurative (idiomatic) meaning of some idioms follows
in a straightforward manner from the literal (source) meaning of the expressions.
For other idioms, you may find it much less obvious how the figurative (idiomatic)
230 Xinqing Wang, Frank Boers and Paul Warren

meaning is derived from the literal (source) meaning. You may even feel that
there is no clear link at all.
Now that you know the literal meaning as well as the figurative meaning of
the idioms, please circle the number on the scale below that best represents how
you feel about the relationship between the literal meaning and the figurative
meaning of the idiom.
5= The link between the literal and the figurative meaning of the expression is
very clear to me. Given the literal use of the expression, it is easy for me to see
how the figurative use is derived from it.
0 = The link between the literal and the figurative meaning of the expression
is not at all clear to me. Although I (now) know both the literal (source) meaning
of the expression and its idiomatic meaning, I cannot see how the literal use
could have given rise to the figurative use of the expression.

non-transparent transparent

     

The link The link The link The link The link The link
between the between the between the between the between the between the
literal and literal and literal and literal and literal and literal and
figurative figurative figurative figurative figurative figurative
meanings is meanings is meanings is meanings is meanings is meanings is
not very vague. vague. somewhat clear. very clear.
transparent at clear.
all.

The participants were encouraged to express their thoughts (in L1 if they wished)
and explain their reasoning as they carried out this rating task per idiom. At the
end of the interview, the participants were asked not to discuss individual items
encountered in the interview with other students. They were not told they would
be asked to recall the meaning of the idioms the following week.
One week later, in the second interview, the participants were presented
with the same set of idioms and asked to try and recall their figurative meaning.
They were also asked if they could recall anything else they had learned about
the idioms in the first interview. The latter question was intended to gauge if
successful recall of the idiomatic meaning often coincided with recall of the lit-
eral underpinnings.
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 231

4.1.3 Analysis

The participants’ meaning guesses in the first interview and their meaning re-
call responses in the second interview were assessed by the first author and
two L1 speakers of English (PhD students in Applied Linguistics). A three-point
scoring protocol was adopted, distinguishing between no or incorrect re-
sponses, partially correct responses, and fully correct responses. It was agreed
that responses were counted as correct if they contained all the meaning as-
pects contained in the definition from the dictionary that was used in the inter-
view. For example, pull one’s weight was defined as doing one’s share in a
common task – or, in other words, to work as hard as other people in a group.
If the response was “to do one’s share in teamwork”, it was rated as 1. If it was
just “to make an effort” or “to do one’s best”, it was rated as 0.5. Inter-rater
reliability was high (Krippendorff α=.78). In case of disagreement, a verdict was
made by majority vote.
As explained, step 1 in the interview sequence helped us to identify which idi-
oms were already known by any given participant. For example, some participants
remembered learning idioms such as let the cat out of the bag, a rule of thumb and
make ends meet from their EFL textbooks. Step 2 revealed which (if any) idioms a
participant correctly guessed the meaning of even before any hint about its origin
was given. New idioms that at least some participants managed to guess the
meaning of in step 2 included a drop in a bucket, have a lot on one’s plate and in
the driver’s seat. If a learner already knew or understood a given idiom, such an
item would likely still be known and understood one week later also in the ab-
sence of any intervention, and so these instances were discarded from further
analysis. Some initial meaning responses generated in step 2 were not fully correct
and thus given a score of 0.5. These instances were retained for analysis because
they did entail room for further learning. This pruning procedure left us with 360
initial responses indicating no knowledge or comprehension of the given idioms
and an additional 54 initial responses demonstrating partial comprehension of the
given idioms, that is, 414 out of the total 500 initial responses (25 participants x 20
idioms). The idioms, with exclusion of those which all members of the group al-
ready knew or interpreted correctly at the start, are listed in the appendix.

4.2 Results and Discussion

Informing the participants about the origin of the expressions was found to con-
siderably increase the likelihood of correct inferences in the first interview. In
49.7% of the 360 instances where participants had initially failed entirely to
232 Xinqing Wang, Frank Boers and Paul Warren

propose a correct interpretation, they now produced correct interpretations, and

in an additional 19.5% they produced partially correct ones. For the 54 instances
where guessing had initially already been partially successful, the information
about the origin of the idiom also occasionally (11.1%) helped learners to arrive
at a fuller understanding.
One week later, no fewer than 65% of the items which participants had
demonstrated no comprehension of at the start of the first interview were re-
called correctly, and an additional 10.5% of the recalls were partially correct. In
addition, 44.5% of the items that had already elicited partially correct guesses
at the very beginning were now accurately recalled. In sum, learning gains
were thus attested for close to 71.5% of the instances where learners’ prior com-
prehension was either nil or incomplete. When the participants recalled the
precise meaning of an idiom in the second interview, this typically (91.25%) co-
incided with their recollection of the idiom’s suggested origin, as displayed in
Table 1. If they failed to recall an idiom’s meaning altogether, they would often
(65.59%) also fail to recall its origin.

Table 1: Meaning recall and recall of literal underpinnings (total N = 414).

Meaning recall Recall of origin No recall of origin

n % n %

Correct  .  .

Partially correct  . Versus  .
Failed  .  .

The next question is whether the proposed literal underpinnings of the idioms
were perceived by the learners as relatively transparent motivations of the idi-
oms’ meanings. The participants’ ratings of the degree of transparency (at the
end of the first interview) were very unevenly spread across the six-point scale,
with the lowest points, 0 to 2, ticked very seldom (together only about 8%).
Idioms and their motivations that attracted such occasional low ratings in-
cluded on the back burner, let the cat out of the bag, take a back seat, hit the
roof, a loose cannon, a wet blanket, hit the roof, jump the gun, have cold feet, be
on the same page, in the wake of, not be up to scratch, hold your horses, bark up
the wrong tree and hand over fist. The two highest points on the scale, 4 and 5,
were selected the most often (together about 73.5%), which suggests that, by
and large, the participants thought the origin of the idioms that they had been
presented with offered a relatively clear motivation for the idioms’ meanings. It
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 233

is worth mentioning, however, that there was considerable disparity among

participants’ judgements. It was not uncommon for idioms to receive a rating of 1
or 2 from one learner but ratings of 4 or 5 from others. Making predictions about
which etymological notes will strike individual learners as clear vs. far-fetched
thus appears problematic, even within a relatively homogeneous group of learn-
ers. This is also one of the reasons why we felt it was important to analyse the
data by individual responses instead of using averaged ratings (see below).
It is somewhat ironic that the ratings for some of the idioms which we men-
tioned earlier as examples of likely transparent versus non-transparent cases
did not confirm our expectations. For instance, take a back seat elicited an av-
erage rating of only 3.50, whereas red herring elicited an average rating of 4.33.
We need to be cautious about these comparisons, though, because they cross
over from one idiom set to another and do thus not reflect the same learners’
judgements. Still, the data do suggest that it may be very hard for teachers, lex-
icographers and researchers to make reliable predictions about which motiva-
tions for idioms’ meanings will be experienced as “making good sense” by an
individual learner or group of learners.
The literal underpinning that elicited the lowest average rating (the only
average below 3.00) concerned a wet blanket. One participant explained his
low rating (0) as follows: “If it [the idiom] comes from putting out fire, I see it
as something useful in a difficult or dangerous situation. This is in contrast
with the negative meaning of the expression as stopping other people’s enthu-
siasm” (translated from Mandarin by the first author). It seems this learner
found the proposed source-target domain mapping improbable, because of his
association of fire with danger. His low transparency rating did not, however,
prevent him from accurately recalling both the idiom’s meaning and its pro-
posed origin one week later.
This brings us to the key question we set out to address here, notably
whether the degree of perceived transparency of the literal-figurative link influ-
ences the mnemonic effect of etymological elaboration. Table 2 gives a first im-
pression of a trend that appears to be in favour of idioms receiving relatively
high transparency ratings. Because the participants’ ratings were so unevenly
distributed, we divide them here into just two categories for preliminary, de-
scriptive purposes: ratings 0 to 3 (“low”) versus ratings 4 and 5 (“high”).
What Table 2 fails to capture, of course, is variation in recall successes due
to many other potentially influential characteristics of individual idioms as well
as characteristics of the individual participants. Regarding the latter (and as
discussed previously), the students’ comparative success as EFL learners may
be particularly relevant.
234 Xinqing Wang, Frank Boers and Paul Warren

Table 2: Meaning recall of idioms: Transparent vs. non-transparent underpinnings

(N = 414).

Meaning recall

Correct partially correct failed

Transparency n % n % n %

–  .  .  .

–  .  .  .

We therefore turned to mixed-effects regression models for ordinal data, using

clmm (cumulative link mixed modelling) in the ordinal package in R (Christensen,
2018). The dependent variable was the Recall score, with the scores of 0, 0.5 and 1,
representing ordered categories of wrong/null, partially correct and fully correct.
The fixed effect predictors were Transparency (i.e., the learner’s rating of the trans-
parency of the proposed connection between the literal underpinning and the
meaning of the idiom) and the student’s TEM-4 grade (i.e., the measure of the
learner’s EFL achievement or English proficiency). Since we hypothesized (see
above) that the effect of perceived transparency may vary with the level of EFL
achievement or proficiency, we included the interaction between Transparency
and TEM-4 grade as another predictor. The random effects were participants and
items (i.e., the different idioms). Including these random effects allowed the mod-
els to take into account variation both between participants and between idioms.
Comparison of models with and without the interaction of Transparency and EFL
achievement showed that the interaction made a significant contribution to ex-
plaining the variance in recall scores (χ2 (1) = 4.72, p < .05). The full model, with
the two simple fixed effects and their interaction, was therefore retained. A Type-
III ANOVA test showed that in addition to the interaction, the simple fixed effects
of Transparency (χ2 (1) = 4.01, p < .05) and TEM-4 grade (χ2 (1) = 7.13, p < .01) also
made significant contributions to the full model. The interaction effect is illus-
trated in Figure 1.1
These effects reflect the following patterns. Overall, fully correct recall was
more likely for idioms perceived to have transparent literal-figurative connections,
while unsuccessful recall was more likely for those whose proposed underpin-
nings were perceived to be rather obscure. This supports an affirmative answer to

1 Because the predict function has not been developed for clmm, this plot shows values derived
using the predict function in clmm2. However, because clmm2 only allows one random effect, we
chose to keep items as the random effect since items explain more variance than participants.
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 235

Figure 1: Probability of correct recall based on TEM-4 grade and transparency.

our principal research question: Transparency matters. Fully correct recall was
also more likely for the learners with comparatively high TEM-4 scores. This is in
line with our prediction that the high-achieving students in the EFL programme
would also be the high achievers in our experiment.
The interaction effect revealed by the regression model is as follows. While
the students with low TEM-4 grades performed generally more poorly than their
high-achieving peers on the recall test, this was especially acute for idioms
whose transparency they had rated as low. Idioms whose literal-figurative con-
nection they deemed highly transparent were recalled the best at this lower end
of the TEM-4 grades. However, as TEM-4 grades increased, the impact of trans-
parency gradually diminished and disappeared by grade 80 (see Figure 1). So,
if the transparency factor was found to be a predictor of recall, this is attribut-
able to the low achievers’ data. Returning to the example of the student who
had given a wet blanket the lowest transparency rating but nonetheless suc-
cessfully recalled it, it is perhaps no coincidence that this was a student with a
relatively high TEM-4 grade (78). According to the interview data, this student
was clearly willing to reflect on the literal underpinning of the idiom proposed
by the researcher, perhaps precisely because its motivation for the idiomatic
meaning was puzzling to him. It is possibly this willingness to put a certain
236 Xinqing Wang, Frank Boers and Paul Warren

effort into evaluating a proposed literal-figurative connection that helped him

to entrench this association in memory (recall that the student remembered
both the idiom’s meaning and its proposed underpinning).

5 Conclusion
The findings presented here suggest that, if etymological elaboration aids learn-
ers’ retention of the meaning of idioms, its effect is not confined to idioms whose
meaning learners find to be straightforwardly derived from the proposed origins.
Rather, it appears that explanations about origins that are experienced as some-
what far-fetched can also serve this mnemonic purpose. However, the findings
also indicate that this mnemonic purpose is generally served more easily in the
case of idioms where the learner finds the proposed literal underpinning compar-
atively straightforward or plausible. Interestingly, the interaction effect that
emerged from our mixed effects regression model suggests that this influence of
the perceived transparency of the literal-figurative connection was the greatest
for students who obtained comparatively low TEM-4 exam grades, and who
could by that proxy be considered the less successful or less advanced EFL stu-
dents in the sample. For the participants with the higher TEM-4 grades, who
could be considered the high achievers in this student population, there was no
noticeable impact of the transparency variable on idiom recall.
Despite the role of said variables, the fact remains that most of the students
in this study gained knowledge of a considerable number of new idioms
through an instructional procedure which engaged them with the literal under-
pinnings of the expressions. We need to be cautious not to oversell the pro-
claimed benefits of this etymological elaboration approach, however. As
mentioned, not all evidence to date has been favourable of its implementation
(Szczepaniak and Lew 2011). We also need to bear in mind that the present
study did not include a comparison treatment. As such, it provides no evidence
that learning idioms through etymological elaboration should be given prece-
dence over other learning procedures. The jury is still out, so to speak. What we
can say with a degree of confidence is that the learning gains attested here
were quite substantial, at least according to a one-week delayed post-test. Of
course, a further delayed post-test would be required to check if the observed
learning gains are truly durable.
At the same time, it is possible that idiom knowledge was underestimated
in our experiment. That is because we only gave full credit to learners’ para-
phrases of the idioms if these paraphrases included all the meaning components
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 237

mentioned in the dictionary entry that was used as a benchmark. It cannot be

ruled out that one or the other meaning component was implied rather than
explicitly verbalized in a respondent’s paraphrase. Even in an interview proce-
dure that takes recourse to the respondent’s L1, an interviewer’s efforts to so-
licit nuanced knowledge may fail. In future (conceptual) replications, it might
be worth adding a different type of test, such as a multiple-choice meaning-
recognition test where the respondent is required to select the meaning para-
phrase that best captures the idiom’s meaning.
If degree of transparency of the literal-figurative connection was found to
play a role in learners’ recall of the idioms’ meanings in this study, this raises
the question what other variables might play a part in this regard. Further anal-
yses of the interview and test data and inclusion of alternative idiom-related
variables in new mixed effects models will be required to identify these. In a
similar vein, the potential influence of other learner traits than proficiency mer-
its further investigation in future research on L2 idiom learning and teaching.
Finally, we need to be aware that the learning aim in the present experi-
ment may in one respect be considered modest – that is, remembering the
meaning of the idioms. For one thing, the post-test presented the learners with
the English idioms again, and thus did not require them to recall the lexical
makeup of the expressions. For another, it is well documented that idioms
serve subtle pragmatic and evaluative functions in discourse (e.g., O’Keeffe,
McCarthy and Carter 2007: 80–99) which are often not captured by dictionary
definitions. Developing expert productive knowledge of idioms would take ex-
tensive exposure to L2 natural discourse (MacArthur 2010). It would nonethe-
less be interesting to explore, by means of a longitudinal investigation, if the
intervention we have described here stimulates learners’ long-term engagement
with the idiomatic dimension of their target language.

Acknowledgements: We are grateful to the 25 students who volunteered to par-

ticipate in this study, to Lisa Woods and Rolando Coto Solano for their advice
on inferential statistics, and to Mark Toomer and Brian Strong for the many
hours they invested in assessing and discussing the participants’ meaning
guesses and post-test responses.

References
Boers, Frank. 2013. Cognitive semantic ways of teaching figurative phrases: An assessment. In
Francisco Gonzálvez-García, María S. Peña Cervel & Lorena Pérez Hernández (eds.),
238 Xinqing Wang, Frank Boers and Paul Warren

Metaphor and metonymy revisited beyond the contemporary theory of metaphor,

229–263. Amsterdam: John Benjamins.
Boers, Frank, Murielle Demecheleer & June Eyckmans. 2004. Etymological elaboration as a
strategy for learning figurative idioms. In Paul Bogaards & Batia Laufer (eds.), Vocabulary
in a second language: Selection, acquisition and testing, 53–78. Amsterdam: John
Benjamins.
Boers, Frank, June Eyckmans & Hélène Stengers. 2006. Motivating multiword units: Rationale,
mnemonic benefits, and cognitive style variables. In Susan H. Foster-Cohen,
Marta M. Krajnovic & Jelena M. Djigunovic (eds.), EUROSLA Yearbook Vol. 6, 169–190.
Amsterdam: John Benjamins.
Boers, Frank, Paul Warren, Gina Grimshaw & Anna Siyanova-Chanturia. 2017. On the benefits
of multimodal annotations for vocabulary uptake from reading. Computer Assisted
Language Learning 30. 709–725.
Cermak, Laird S. & Fergus I. M. Craik. 1979. Levels of processing in human memory. Hillsdale,
NJ: Lawrence Erlbaum.
Cieślicka, Anna B. 2006. Literal salience in on-line processing of idiomatic expressions
by second language learners. Second Language Research 22. 115–144.
Christensen, Rune H. B. 2018. Ordinal - regression models for ordinal data. R package version
2018.4–19. http://www.cran.r-project.org/package=ordinal/.
Elgort, Irina & Paul Warren. 2014. L2 vocabulary learning from reading: Explicit and tacit
lexical knowledge and the role of learner and item variables. Language Learning 64.
365–414.
Gibbs, Raymond W., Nandini P. Nayak & Cooper Cutting. 1989. How to kick the bucket and not
decompose: Analyzability and idiom processing. Journal of Memory and Language 28.
576–593.
Gu, Yongqi. 2003. Fine brush and freehand: The vocabulary learning art of two successful
Chinese EFL learners. TESOL Quarterly 37. 73–104.
Gu, Yongqi. 2013. Vocabulary learning strategies. In Carole A. Chapelle (ed.), The encyclopedia
of applied linguistics. Oxford: Wiley-Blackwell.
Hu, Ying-hsueh & Yu-Ying Fong. 2010. Obstacles to conceptual-metaphor guided L2 idiom
interpretation. In Sabine De Knop, Frank Boers & Antoon De Rycker (eds.), Fostering
language teaching efficiency through cognitive linguistics, 293–317. Berlin: Mouton de
Gruyter.
Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago, IL: University of Chicago
Press.
Liontas, John I. 2017. Why teach idioms? A challenge to the profession. Iranian Journal of
Language Teaching Research 5: 5–25.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where are we and where should we
be going? A view from the language classroom. AILA Review 23. 155–173.
O’Keeffe, Anne, Michael McCarthy & Ron Carter. 2007. From corpus to classroom: Language
use and language teaching. Cambridge, UK: Cambridge University Press.
Paivio, Allan. 1986. Mental representations: A dual coding approach. New York, NY: Oxford
University Press.
Piquer-Píriz, Ana M. 2008. Reasoning figuratively in early EFL: some implications for the
development of vocabulary. In Frank Boers & Seth Lindstromberg (eds.), Cognitive
Linguistic Approaches to Teaching Vocabulary and Phraseology, 233–257. Berlin: Mouton
de Gruyter.
 Using Literal Underpinnings to Help Learners Remember Figurative Idioms 239

Robinson, Peter. 2013. Aptitude in second language acquisition. In Carole A. Chapelle (ed.),
The encyclopedia of applied linguistics. Oxford: Wiley-Blackwell.
Siyanova-Chanturia, Anna, Kathy Conklin & Norbert Schmitt. 2011. Adding more fuel to the
fire: An eye-tracking study of idiom processing by native and non-native
speakers. Second Language Research, 27. 251–272.
Skehan, Peter. 2015. Foreign language aptitude and its relationship with grammar: A critical
overview. Applied Linguistics 36. 367–384.
Steinel, Margarita. P., Jan H. Hulstijn & Wolfgang Steinel. 2007. Second language idiom
learning in a paired associate paradigm: Effects of direction of learning, direction of
testing, idiom imageability, and idiom transparency. Studies in Second Language
Acquisition 29. 449–484.
Szczepaniak, Renata & Robert Lew. 2011. The role of imagery in dictionaries of idioms. Applied
Linguistics 32. 323–347.

Appendix
Target idioms (excluding ones known/understood by all respondents in the
group already before the instructional procedure)
Set A: go belly-up; a drop in the bucket; on the same wavelength; get into
gear; red tape; follow suit; pull one’s weight; bread and butter; throw in the
towel; a sitting duck; give the green light; spill the beans; have/get cold feet;
jump ship; play into someone’s hand; par for the course; on the back burner;
take the bull by the horns; take it on the chin.
Set B: the ball is in your court; win hands down; on the ropes; weigh some-
one down; pass the baton; give someone the cold shoulder; leave someone
high and dry; in the doldrums; beat around the bush; waiting in the wings; a
shot in the arm; throw your hand in; a loose cannon; on the same page; let the
cat out of the bag; (not) up to scratch; take a back seat; hit the roof/ceiling; a
feeding frenzy.
Set C: in the driving/driver’s seat; play your cards close to your chest;
down and out; rub someone the wrong way; turn over a new leaf; (hit) below
the belt; make ends meet; ring a bell; a red herring; bury the hatchet; have a lot
on your plate; sit on the fence; pass the buck; jump the gun; stick your neck
out; take the plunge; teething problems; hold your horses; in the wake of some-
thing; a rule of thumb; show someone the ropes.
Set D: a hot potato; have a green thumb; go with the flow; (start) from
scratch; a can of worms; flex your muscles; give someone a leg up; pull a rabbit
out of the hat; bite the bullet; get your second wind; cut corners; on automatic
pilot come out of your shell; get something off your chest; hand over fist; throw
your hat/cap into the ring; play it by ear; a wet blanket.
Rafael Alejo-González and Verónica García-Bermejo
“The Manage of Two Kingdoms Must”:
An Analysis of Metaphor in Two CLIL
Textbooks
1 Introduction
In the last two decades, Content and Language Integrated Learning (CLIL) has be-
come a common educational approach widely implemented in European Primary
and Secondary schools. Following on the footsteps of Canadian bilingual experien-
ces and promoted by the European Union, CLIL refers to those bilingual school ex-
periences in which a foreign language, typically English, is used to teach, in part
or in whole, Science, History, Mathematics, Physical Education, Music or other
subjects of the school curriculum (Coyle, Hood and Marsh 2010; Marsh 2008).
As a result of its considerable expansion, CLIL has attracted the attention of
applied linguists, who have turned their attention to the analysis of its impact in
terms of language acquisition, cognitive development or content education, the
general consensus being that CLIL programmes are helpful in developing both stu-
dents’ foreign language abilities and content learning (cf. Pérez-Cañado 2012 for
an overview, although see Fernández-Sanjurjo, Fernández-Costales and Arias
Blanco 2017 for a less optimistic view). There has also been much attention to the
role of language and pragmatics in CLIL and to the different discourse functions
and speech acts (e.g., questions, requests, directives) that are typically used in
CLIL contexts (see for example Dalton-Puffer 2007; Llinares, Morton and Whittaker
2012), an analysis which has mostly been carried out within the framework of
Systemic Functional Linguistics. The type of interaction in CLIL classrooms, follow-
ing an interactionist approach, has also been the focus of some other research
showing in what way the interaction happening in content subjects is supportive
of L2 learning (see García Mayo and Basterrechea 2017). However, to the best of
our knowledge, the research on CLIL has not incorporated a cognitive-linguistic
perspective to the examination of the language being used in these experiences
and, although another kind of metaphor, i.e.“grammatical metaphor”, has been

Note: King John 1.1, 35.

Rafael Alejo-González, English Philology (Faculty of Education), Universtiy of Extremadura,

Badajoz, Spain, e-mail: [email protected]
Verónica García-Bermejo, Badajoz, Spain, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-012
242 Rafael Alejo-González and Verónica García-Bermejo

studied with some detail (cf. Llinares et al. 2012; Järvinen 2010), metaphor, as
understood in Cognitive Linguistics, has not been considered when providing a
characterisation of this educational approach.
Given our focus on CLIL as a communicative context and a methodology
based on the analysis of real CLIL textbooks, the present paper adheres to what
some scholars call ‘real world metaphor research’ (cf. Deignan, Littlemore and
Semino 2013), a research strand which has revealed the role of this figure of
speech in a diversity of communicative contexts (e.g., Koller 2006 in business,
Charteris-Black 2004 in political communication, Pérez-Sobrino 2016 in advertis-
ing and Caballero 2012 in tennis commentaries), among which university contexts
feature prominently (Littlemore 2001; Littlemore, Chen, Barnden and Koester
2011; MacArthur 2016). This same research perspective was the one adopted to
analyse metaphor use in the two educational contexts that more closely relate to
CLIL, i.e. L1 school contexts (cf. Cameron 2003) and foreign language teaching
(see MacArthur 2017 for an overview), in which metaphor, although not particu-
larly frequent, is mainly used to mitigate potentially face threatening situations
to children in the classroom or to provide support for vocabulary learning to L2
learners respectively.
The peculiarities of CLIL, described above, impose a research programme
that analyses to what extent the basic metaphor patterns and goals revealed for
L1 school settings (cf. Cameron 2003) will be affected by the use of a language
in which the main participants (teachers and learners) are not fully proficient.
In other words, we need to investigate whether the use of an L2 will make meta-
phors less frequent, whether this will change some of its functions, or whether
there will be variation according to the topic or the specific moment of the ped-
agogic process. For example, it would be important to know if metaphor still
plays a role in explanations in facilitating access to the abstract content of the
academic curriculum. It must be borne in mind that in order to do this one has
to assume the previous knowledge of the more concrete vocabulary used as
metaphor vehicles, a knowledge that many L2 learners may not have.
It is the purpose of the present study to provide a first approach to the impor-
tance of metaphor in CLIL by analysing textbooks as one of the main sources of
input. As an exploratory study, this paper describes some of the metaphors used
by two textbooks used by a group of CLIL students in year 5 of Primary Education
in a bilingual school in Extremadura, Spain. More precisely, we report the results
of the identification of open-class metaphors (i.e. adverbs, adjectives, verbs and
nouns) in selected lessons from the Social and Natural Science books. The impor-
tance of this exploration is highlighted by two factors. First, together with teach-
ers’ explanations, written language input has a fundamental role in students’
educational success and development, and it is to be expected that in CLIL
 “The Manage of Two Kingdoms Must” 243

contexts, where teachers will feel less confident in their ability to provide explana-
tions in an L2, this role becomes reinforced. Second, the ability to use the text-
books will depend, on the one hand, on the students’ reading abilities, which as
shown by the literature, are fundamentally determined by their vocabulary knowl-
edge, and, on the other hand, the characteristics of the vocabulary used in the text-
books (see de Zeeuw, Grootjen, Jan Kootstra and Tellings 2019). These vocabulary
features are related to word properties such as word frequency, morphological
family size or polysemy and, although metaphor is not included among them, it
seems only natural to hypothesize that this figure of speech will also have an im-
pact on students’ understanding of texts.
The chapter is organised as follows. After introducing the relevant litera-
ture on the subject and reporting the methodology used, we present the results
of our metaphor identification for tokens and lemmas together with the distri-
bution of this figure of speech by Part of Speech (PoS). We also analyse the in-
ternal distribution of metaphors in the text taking into account the different
stages or moves that can be defined within the lessons. We finish by discussing
the results obtained and by giving a preliminary characterisation of the use of
metaphor in CLIL textbooks.

2 Metaphor Variation
2.1 Register

The importance of the relationship between metaphor and register has been
highlighted by Berber Sardinha (2015), who has explored how metaphor –
together with tense, aspect, modality, etc.- is another of the linguistic features that
can help us identify the different functional dimensions of register, understood
here as “a language variety associated with both a particular situation of use and
with pervasive linguistic features” (Biber and Conrad 2009: 31). Replicating the
methodology used by Biber and associates (Biber 1988, 2006; Biber and Conrad
2009) Berber Sardinha’s research shows how metaphor contributes to the different
dimensions of language and as a result metaphor can explain up to 41% of the
variation between registers (cf. Berber Sardinha 2015).
By using a multi-dimensional (MD) statistical analysis, Berber Sardinha’s work
elaborates on the analysis started by the VUAMC (Vrije Universiteit Amsterdam
Metaphor Corpus) team, who had already undertaken the study of metaphor
in four major registers in the English language (Steen et al. 2010). Using a
sample of the BNC Baby corpus, this group of researchers analysed the presence
244 Rafael Alejo-González and Verónica García-Bermejo

of metaphor in academic discourse (Herrmann 2013), news (Krennmayr 2011),

fiction (Dorst 2011) and conversation (Kaal 2012). Contrary to expectations, aca-
demic discourse, whose subject matter would seem to require the use of literal
meanings and avoidance of meaning displacements, was found to be the most
metaphorically dense register (18.5%). Another informational register, the news,
was the second most metaphorically dense register (16.4%), ahead of fiction
(11.9%), which in principle could be assumed to be prone to making more use of
expressive resources like metaphor. Finally, at a great distance from the rest,
conversation showed the least metaphorically dense language (7.7%).
From a functional perspective, these results are less of a surprise. Using
Biber’s functional dimensions, we learn that texts with a high informational
value, rich on explicit reference and a non-narrative dimension would provide
‘a natural environment’ for metaphor to thrive. The information packaging and
abstraction that is typical of academic discourse and news could explain their
higher densities, while the involved and situational dependence of conversa-
tion could account for its low density.

2.2 Part of Speech

However, metaphor density is not the only element of variation among the differ-
ent registers. The VUAMC team also analysed the distribution of metaphors accord-
ing to PoS. It was found (see Krennmayr 2017 for a summary of the finding by the
VUAMC team) that register and PoS together interacted significantly with meta-
phor in such a way that certain registers had a higher proportion of metaphorical
word classes. Thus, in the academic register, prepositions, verbs and adjectives
have a greater metaphorical weight than other PoS, including nouns, which, al-
though particularly frequent in that register, are not used as much with metaphori-
cal senses. This distribution of metaphorically-used PoS is almost reproduced in
the case of the news, the only difference being that, in this case, metaphorical
nouns are also common. For their part, fiction and especially conversation show a
different profile, the former having adjectives and prepositions at the top of their
list and verbs in third position, and the latter incorporating prepositions, determin-
ers and adjectives as the PoS with greatest metaphorical weight.
The analysis of PoS is interesting because it forms the basis of many of
Biber’s analyses and because, as Krennmayr (2017) acknowledges, “associations
between metaphor and register cannot be interpreted directly but need to be in-
terpreted by looking at the distribution of metaphor across word classes per reg-
ister” (2017: 173). In other words, the number of metaphors per word class needs
 “The Manage of Two Kingdoms Must” 245

to be put into perspective as, for example, a high number of noun metaphors
might be expected in academic discourse given that nominal groups are typical
of this register (Biber 2006; Conrad 2001).

2.3 Disciplinary Variation and Genre

An additional dimension of variation is disciplinary variation, a topic dealt

with by Herrmann (2015). In her study, this author analyses the different distri-
butional patterns of metaphor use in the four general ‘subregisters’ or broad
disciplines that she defines within academic discourse: 1) Natural Science; 2)
Humanities and Arts; 3) Politics, Law and Education; and 4) Social Science. The
general conclusion is that there seem to be no significant differences in meta-
phor density between the ‘subregisters’ with the one exception of Natural
Science, which shows a lower percentage of Metaphor Related Words (MRW).
However, the differences are not marked (17% in Natural Science vs 18.8% in
the Humanities, 18.1% in Politics and 19% in Social Science).
Finally, genre, whose defining feature is not a broad situation, as in the
case of register, but the specific purpose and the discourse community with
which a text type is created (cf. Deignan et al. 2013: 40), provides a more pre-
cise source of variation (see Caballero 2017 for a review), although it has not
been explored as much as register when dealing with metaphor in general and
with metaphor density in particular. The reason could be that for a linguistic
feature to be typical of a specific genre it need not be very frequent. For exam-
ple, the expressions and linguistic elements used in the opening salutation of a
letter (cf. Biber and Conrad 2009) are bound to be infrequent, although are ac-
tually part the conventions characterising that genre. Thus, if a linguistic ele-
ment, in our case metaphor, is described as frequent it certainly is a feature to
consider when describing that genre.
However, the number of studies giving an account of metaphor frequency
in specific genres is relatively low. Exceptions include Low, Littlemore, and
Koester’s (2008) article on lectures, where we learn that the sample studied
ranges from 11 to 13%, and MacArthur’s (2016) study on office hours consulta-
tions, another oral academic genre, which is reported to have a metaphor
density of 12.2%. Other authors approach the phenomenon from a different
angle and, instead of concentrating on metaphor density overall, they focus
on the presence of metaphor bursts, i.e. on the specific stretches of text
where metaphor is highly frequent (cf. Corts and Pollio 1999). In this way,
specific rhetorical or communicative goals of the participants of the discourse
246 Rafael Alejo-González and Verónica García-Bermejo

can be identified. Metaphor is thus examined as one of the features of genres,

i.e. their staged nature (Caballero, 2017). The extent to which metaphor fre-
quency fits a previously defined move structure of a specific genre has yet to
be explored.

3 Metaphor in School Contexts

As already stated, the focus of the present article is not only linguistically defined
by register or genre, it is also established by the situational context in which the
textbooks under analysis are used, i.e. the Primary School context. Unfortunately,
in spite of the amount of attention that metaphor has attracted in educational
contexts, the school context has been neglected. According to Littlemore, with
the exception of Cameron’s (2003) seminal work, “there have been no other de-
tailed studies of metaphor use in either primary or secondary education” (2017:
285). But even Cameron’s work cannot strictly provide a model for the present re-
search as it was carried out in a primary school in the UK, where the L1 of the
students was English and the educational context was British education, and the
main source for her corpus was the oral interaction in the classroom. We cannot
take for granted that the use of metaphor by teachers with the function of agenda
management, summarizing, providing feedback to the students or helping in giv-
ing explanations is an adequate model for the role of metaphor in a context or for
a genre like the ones studied here.
Another strand of research related to teaching or educational contexts is the
one dealing with second/foreign language learners. According to Hoang and
Boers (2018), this exploration has mostly focussed on four main areas: metaphors
conforming teachers’ and learners’ mind-set (Wan and Low 2015), awareness rais-
ing or use of linguistic motivation (MacArthur 2017), metaphor related comprehen-
sion problems (Littlemore 2001) and learners’ production (Nacey 2013; Littlemore,
Krennmayr, Turner, and Turner 2014). However, as in L1 contexts, young learners
have not explicitly been the focus of attention in the literature on metaphor use in
L2 learning (for an exception see Piquer-Píriz 2008). In all these cases, the studies
are performed in the context of EFL classes, materials or activities addressed at
improving the learners’ competence in a foreign or second language. As we stated
in the introduction, there is, to the best of our knowledge, no similar research on
CLIL contexts, where the attention is not uniquely placed on how to learn a lan-
guage but also on how to acquire the subject content, which makes it particularly
interesting to investigate metaphor in this context.
 “The Manage of Two Kingdoms Must” 247

It seems, therefore, appropriate to explore more in depth the CLIL context

and to analyse the textbook as a particularly prominent genre by addressing
the following research questions:
– What are the patterns of metaphor use in CLIL textbooks and how do these
patterns expose, or not, connections to the registers described by the re-
search? That is:
– Is metaphor a frequent phenomenon in CLIL textbooks? Are these texts
metaphorically dense?
– Are certain parts of speech overused or underused as metaphors?
– What are the main sources of internal variation in metaphor use?
– Is there disciplinary variation in metaphor use?
– Is there variation in the different moves or stages of a CLIL lesson?

4 Methodology
To carry out the proposed analysis, we selected, from a corpus of textbooks pre-
viously compiled (García-Bermejo 2015), six lessons belonging to two different
textbooks, which gave a total of 18,023 tokens, a figure that was considered
more manageable given that the time-consuming process of metaphor tagging
was carried out manually.
The two textbooks selected were Natural Science ByMe: Primary 5 (2014) and
Social Science ByMe: Primary 5 (2014), which were used to teach two compulsory
subjects, Natural Science and Social Science respectively, to students in 5th grade
of Primary Education in Spain (10–11 years old). These textbooks are widely used
in bilingual programmes in Spain and, as a consequence, they provided a good
example of some of the language students are exposed to in these programmes.
The units to be analysed were chosen so that the first lesson of each aca-
demic term (i.e., units 1, 4 and 7 of each textbook) was included. In this way,
we made sure that different thematic subjects were covered and a that the pro-
gression from beginning to end was also taken into account.
Once the corpus had been compiled, the text was scanned to tag open class
parts of speech, i.e., nouns, verbs, adverbs and adjectives, a process that was car-
ried out using the PoS tagger provided by the Wmatrix online software (Rayson
2008). The identification of metaphor related words (MRWs) was carried out manu-
ally by the two authors. Each researcher read the selected chapters from the books
cited above and tagged potential MRWs individually. The researchers then met to
reach a final agreement on the words we considered metaphor-related. In this pro-
cess of MRW identification we followed MIPVU (Steen et al. 2010).
248 Rafael Alejo-González and Verónica García-Bermejo

The only case in which we did not follow MIPVU was phrasal verbs, consid-
ered within MIPVU’s methodology as polywords. As explained by Nacey (2013:86),
MIPVU analyses phrasal verbs as a single unit because there is evidence that
speakers ‘lump’ them as a whole. However, if we take into account the rich cogni-
tive-linguistic literature on prepositions and particles (Lakoff 1989; Tyler and
Evans 2003), it would be difficult to argue that these constituents of phrasal verbs,
i.e. verbs and particles, do not contribute a meaning of their own, a stance also
taken by other metaphor researchers (MacArthur 2016; MacArthur, Krennmayr,
and Littlemore 2015).
The dictionary used to establish the basic meaning of the words was the
MacMillan Dictionary for Advanced Learners, which is the usual reference work
used in metaphor identification procedures (cf. MacArthur 2015; Steen et al.
2010). However, when there were instances where the dictionary was not help-
ful (see MacArthur 2015 for a complete explanation), we consulted the VUAMC
corpus (available online at http://www.vismet.org/) to see whether instances
similar to ours had been identified. If after all these steps, we still felt that the
information provided in the dictionary was not helpful, we consulted other dic-
tionaries, especially the Longman Dictionary of Contemporary English, to check
for the basic meaning of lexical units.
It was also important to establish how to proceed with the specific termi-
nology related to Biology or to History. As pointed out by Herrmann (2013), it is
sometimes difficult to establish how incongruent the meaning of some terms
can be in relation to the context in which they are used. This is the case, for
example, of certain verbs like breathe, whose meaning is usually applied to
mammals but it is not exactly clear to what extent it is used metaphorically
when applied to plants. Do plants literally breathe? Here we follow Cameron
(2003: 67), who suggests that in educational contexts technical language may
be more prone to be interpreted metaphorically. As a result, we consider these
terms as metaphorical.
Given the pedagogic goal of the texts, direct metaphors were also tagged. It
is undertandable that some of the comparisons should be made directly to facili-
tate students’ comprehension. We read the text and looked for the typical
markers, or flags, of direct metaphors, which include ‘like’, ‘as’, ‘seem’, ‘appear’,
‘sort of’, ‘kind of’, etc. We also discovered that there was a section –‘In other
words’-, which typically featured a simile to help students to understand specific
concepts explained in the text.
Once the text had been tagged for metaphor, a database using Excel was
created and processed (i.e. tags converted to numbers) to be able to carry out
the different statistical analyses by using the SPSS statistical package (v.17.0).
 “The Manage of Two Kingdoms Must” 249

5 Results
5.1 Metaphor Density

The first research question concerns the frequency of metaphor in CLIL text-
books. We were surprised by the constant presence of this figure of speech in
the discourse. See for example the opening paragraph of the Science book:

Science helps us understand the world we live in. It shows us the importance of asking
questions about natural phenomena that occur around us. Science helps us answer these
questions by observing, predicting, researching, doing experiments and drawing conclu-
sions. It also shows us the importance of technological and scientific advances in our
lives.

The use of 7 metaphors in a paragraph with 53 words at the beginning of lesson 1

clearly shows that this figure of speech is going to play an important role in this
textbook. A good initial explanation could be that personifying Science brings it
closer to the students, who can thus see it not so much as an abstraction but as a
person, a ‘helping teacher’, who is going to ‘accompany’ them in the learning
process.
The importance of metaphor in the complete corpus is not just a mere first
impression but is supported by a global quantitative analysis (see Table 1).
Thus, the total number of open-class metaphors identified in the 6 lessons used
in this study is 1,204, which given the number of tokens amounts to a density
of 10.5%. This means that 1 in every 10 open class words is used metaphorically
in the texts.

Table 1: Number of metaphor related words (tokens and lemmas).

Open class Tokens % Lemmas % Lemma/

words token ratio

MRW  .%  .% .

Non- MRW  .%  .% .
Total  .%  .% .

It must be pointed out that the metaphor density of CLIL textbooks is not the
result of the repetition of the same metaphors again and again. Thus, if we con-
sider the number of lemmas used metaphorically, we can see that the figure of
metaphor density is even higher (13.1%) than that of tokens or that the stan-
dardized type/token ratio is also quite high: 31.9. In other words, 1 in every 3
lemmas used metaphorically is not repeated. This means that, when reading
250 Rafael Alejo-González and Verónica García-Bermejo

these textbooks, students will be frequently faced with the task of understand-
ing words used with a meaning that is different from their basic one.

5.2 Metaphor Density and Part of Speech

Nouns and verbs are the most important word classes found in the textbooks
analysed. Together, nouns (with 41.03%) and verbs (with 40.37%) comprise
more than three quarters of the open-class metaphors used in these texts, while
metaphorically used adverbs and adjectives make up a smaller fraction of the
total number of metaphorically used words. However, nouns and verbs are the
most frequent PoS in the textbooks and so it is to be expected that many instan-
ces of metaphor will involve these word classes.
We therefore performed a Chi square analysis of the percentages of MRWs
per word class (see Table 2) as a way to avoid the uneven representation of cer-
tain PoS in the corpus. The results are statistically significant (χ2 = 173.664;
df=3; p<.0001; Cramer’s V 0.123) and clearly show that adjectives and nouns
are less likely to be used metaphorically, while verbs and adverbs occur com-
paratively often with a metaphorical meaning. Unlike verbs, the predominance
of metaphor-related nouns in absolute terms does not reflect a particular ten-
dency to use them metaphorically but is a side-effect of the higher frequency of
this PoS in the textbooks analysed.

Table 2: Metaphor tokens and parts of speech.

MRW % Non-MRW % total tokens

Adverbs  .%  .% 

Verbs  .%  .% 
Nouns  .%  .% 
Adjectives  .%  .% 
Total  .%  .% 

As far as lemmas are concerned (see Table 3), the results are also statistically
significant (χ2 = 164.46; df=3; p<.0001; Cramer’s V 0.26) and the analysis of
residuals shows that, as with tokens, the proportion of metaphorically used
adjectives and nouns is comparatively small, particularly so in the case of
nouns (Std. Res.=−5.92), whereas the proportion of metaphorically used verbs
and adverbs is more substantial, most notably so in the case of verbs (Std.
Res.=+10.44).
 “The Manage of Two Kingdoms Must” 251

Table 3: Metaphor lemmas and parts of speech.

MRW % MRW Non-MRW % NON.MRW total lemmas

Adjectives  .%  .% 

Nouns  .%  .% 
Verbs  .%  .% 
Adverbs  .%  .% 

5.3 Disciplinary Variation

Our analysis on whether metaphor density was affected by the topic focussed
on two of the most important subjects in the Spanish curriculum: Natural
Science and Social Science.
By comparing the metaphors used in Natural Science with those used in
Social Science (see Table 4), we find that there is a statistically significant differ-
ence in metaphor use (χ2 = 60.640; df=1; p<.0001; Cramer’s V 0.073) and that the
number of metaphors used in the Natural Science textbook is greater than in
Social Science. Thus, while Natural Science lessons have a density of nearly
13.0%, Social Science texts only reach 8.4%, which means a difference of 4.5 per-
centage points, enough to indicate that there is a substantial difference between
the two textbooks analysed in terms of metaphor use.

Table 4: Percentage of MRWs by subject.

MRW % Non-MRW % Total tokens

Natural Sc.  .%  .% 

Social Sc.  .%  .% 
Total  .%  .% 

5.4 Move Variation

The second source of internal variation that we set out to examine is the struc-
ture of the textbooks and the lessons in our corpus. The structure is described
in Table 5, where we include labels for each part together with a short explana-
tion, the corresponding move according to Parodi’s (2010) taxonomy of text-
books moves and the kind of pedagogic function they perform. This structure is
easily identifiable as its limits are marked by page breaks, except in the case of
252 Rafael Alejo-González and Verónica García-Bermejo

Table 5: Structure of the textbooks’ lessons.

Textbook Description Move (cf. Pedagogic

label Parodi ) function

Getting Opening introduction to the textbooks. Only Preamble Introductory

started before unit  of each textbook (Macro-move)

Read and Short overview of unit’s topic followed by Practice Warm up

discover questions

Analyse and Exercises to activate previous knowledge

organize

Content Explanations about the topic of the unit Concept Explanation

It is the only section consisting of more than definitions
one page

Project Practical work on the topic of the subject Practice Follow up

Fragile World Focus on a specific issue related to the topic of Concept

the unit definitions

Revise Exercises to monitor comprehension and Recapitulation

learning

the Content section (see Table 5), which consists of more than one page and
uses titles identifying the topic. The structure is fixed to such an extent that,
except for the opening macro-move in each textbook (Getting started), it is re-
peated in both the Natural Science and the Social Science lessons in the order
appearing in the table.
To check whether metaphor densities varied across these moves, we carried
out a Chi square analysis, whose results, shown in Table 6, are statistically sig-
nificant (χ2 = 20.837; df=6; p<.002; Cramer’s V 0.043). The association is not
very strong, as shown by the low Cramer’s V value, but certain sections do tend
to show greater metaphor use than others. These are Getting started, Read &
Discover and Fragile World. This would indicate that there is greater reliance on
metaphors in these sections, which in principle do not seem to share a common
pedagogic function. At the same time, there is much less use of metaphor in the
case of sections more related to practical work like Analyse & Organise, Project
or Revise. The central section of Content, considered by Parodi (2010) the nu-
cleus of a chapter and usually making up the bulk of the unit, does not stand
out as having either particularly high or low metaphor density.
 “The Manage of Two Kingdoms Must” 253

Table 6: MRW frequency by move.

MRW % Non-MRW % total

Getting started  .%  .% 

Read & discover  .%  .% 
Analyse & organise  .%  .% 
Content  .%  .% 
Project  .%  .% 
Fragile World  .%  .% 
Revise  .%  .% 
Total  .%  .% 

6 Discussion
After having presented the results, we are now in a position to discuss the re-
search questions posed in our introduction. The first question sought to estab-
lish whether CLIL textbooks had a high, or low, degree of metaphor density, a
question that can only be answered by referring to benchmarks against which
our results can be measured. Two main points of reference are available.
First, we can compare our results to the percentages provided by the VUAMC
team (Steen et al. 2010; Krennmayr 2011; Dorst 2011; Kaal 2012; Herrmann 2013).
This comparison should be taken with a certain degree of caution as there is a
methodological difference – relating to phrasal verbs – between our identifica-
tion procedure and MIPVU’s (see methodology). However, it is not very likely
that this should essentially alter the percentages, as only a small number of
words are affected by this methodological decision.
Given the nature of CLIL textbooks, one would expect to find a metaphor
density typical of non-involved informational texts. In other words, the number
of metaphors to be found in the textbooks should be closer to that of academic
register (Herrmann 2013). In fact, this is not what we find if we look at Figure 1,
where we can see that the density in our corpus is nearer the Conversation reg-
ister (Kaal 2012). This would mean that the density could be considered as low
and that school textbooks do not follow the patterns of metaphor use that are
common in academic texts in general.
Secondly, we can consider the most important study to date on metaphor
use in educational contexts (Cameron 2003). In this case, we can see that the
figure obtained in our study for just the open class words (10.5%) is clearly
higher than the 2.7% which Cameron identified in her corpus. We must ac-
knowledge that Cameron’s results are less comparable to ours than the VUAMC
254 Rafael Alejo-González and Verónica García-Bermejo

19.8%
18.3%

14.0%

10.5%
8.9%

Academic News Fiction CLIL Conversation

textbooks

Figure 1: Comparison of metaphor density in CLIL textbooks with other registers.

team’s cited above (Steen et al 2010), as she used a different method of meta-
phor identification and her corpus is made up of mostly spoken language.
However, the number of metaphors that would be added by transforming
Cameron’s unit of analysis into words, a calculation that can be easily made as
she provides the number of single metaphors (i.e. words) and the number of
metaphorical phrases (noun, verb, etc., cf. Cameron 2003:88), would not signif-
icantly alter the tentative conclusion that oral school language in L1 educa-
tional contexts would have a lower metaphor density than the textbooks we are
analysing. Other studies on metaphor density in texts directed at children, also
cited by Cameron, would provide a similar conclusion as “around 10 metaphors
per 1000 words might be expected in classroom texts” (Cameron 2003: 57).
In short, although the number of metaphors found in our corpus could be
considered small if compared with academic texts in general, it seems that high
metaphor densities are not to be expected in school texts directed at children,
much less in those addressed to L2 learners. As a consequence, a figure of
10.5% may very well describe a not unimportant role for metaphor in the texts
analysed, especially if we take into account that the type/token ratio indicates
a high degree of variation in the metaphors used.
This conclusion is reinforced by our results on the breakdown of meta-
phors by PoS, which relates to our second research question. The relatively
high proportion of metaphorically used verbs and a comparatively smaller pro-
portion of metaphorically used nouns was also characteristic of the academic
register (Herrmann 2013), which in this case matches what happens in school
language (Cameron 2003). These PoS results therefore corroborate that
 “The Manage of Two Kingdoms Must” 255

patterns of metaphor use also conform to the academic register, although an

explanation for each word class is needed.
The explanation for the high number of verbs is provided by Krennmayr
(2017). According to her, the use of verbs is related to personification, which
renders the text more comprehensible and accessible for students by providing
them with a ‘human-like’ approach to abstract concepts. This personification
function of metaphor related verbs is at its peak in textbooks when definitions
are provided. General abstractions are thus rendered more manageable:

Volume describes how much space matter occupies.” (Nat. Science, less.7)
Density explains why some objects float in water while others sink (Nat. Science, less.7)
A force is a push or pull that acts on an object. (Nat. Science, less.7)
Friction acts in the opposite direction to the direction of movement (Nat. Science, less.7)

Unlike verbs, the results for adverbs were unexpected, as adverbs have not
been shown to be particularly prone to being used metaphorically in the aca-
demic register (Herrmann 2013). Here the explanation is simple. On the one
hand, it is related to our methodological decision to consider the particle of
phrasal verbs not as parts of a single multi-word verb but as an adverb. On the
other, as typical of academic discourse, adverbs are used as metadiscursive de-
vices which indicate that texts are conceived as locations (Herrmann 2013: 199).
Here are some examples of this use in CLIL textbooks:

What are the functions of the elements in a city listed above? (Nat. Science, lesson 1)
What does the image above illustrate? (Soc. Science lesson 4)
Explain the importance of the advances listed below. (Soc. Science lesson 7)
Use the words from the words in the box below. (Nat. Science, lesson 1)
In your notebook, complete the graphic organiser below by choosing the words from the
box. (Nat. Science, lesson 1)

As for metaphorically-used nouns, contrary to what has been shown in aca-

demic discourse, where their frequency is the same as in the overall discourse
(Herrmann 2013: 197), the relatively small proportion of metaphor-related
nouns in our corpus was not expected and seems to be a characteristic of the
textbooks we have analysed. If we have a look at the list of 15 most frequent
nouns in our corpus, we find that there are only 3 nouns that are frequently
used metaphorically: kingdom, cell and thing. The rest, as can be seen from
Table 7 are concrete nouns referring to the basic entities or components of the
disciplines being studied (plant, water, matter, city), to specific singular items
(Earth, Moon, Spain) or to the materials used in class (notebook). Reference
rather than explanation makes it appropriate to avoid metaphor.
256 Rafael Alejo-González and Verónica García-Bermejo

Table 7: Most frequent nouns in the corpus studied.

Lemma Freq % of Nouns Lemma Freq % of

nouns

 plant  .%  Moon  .%

 water  .%  City  .%
 Earth  .%  matter  .%
 kingdom  .%  Force  .%
 cell  .%  Space  .%
 Spain  .%  Word  .%
 notebook  .%  animal  .%
 planet  .%

We now move on to our third research question, which asked whether

there was disciplinary variation (Social vs. Natural Science) in our corpus.
The results indicate a positive answer, although they contradict the ones ob-
tained by Herrmann (2015), where the Natural Science ‘subregister’ shows a
lower metaphor density. It would be fair to say that the jury is still out since
neither Herrmann’s nor the present study offer conclusive statistical results.
From a qualitative perspective, though, this different behaviour of the two
‘subregisters’, or thematic areas, can be best exemplified by looking at a spe-
cific noun like kingdom (see Table 8), which is used 33 times in the Natural
Science textbook and 26 in the Social Science textbook.
As the Shakespearean quote in our title indicates, it seems that Primary
school children have to learn how to interpret the word kingdom in two clearly
differentiated contexts. In the context of the Natural Science textbook, the word is
exclusively used metaphorically and appears in the beginning chapters of the
book. In contrast, in the Social Science textbook the meaning is literal and the
word is used in the middle and final chapters. In terms of language processing,
this separation of contexts is probably helpful for students. What remains to be
explored is whether the use of the metaphorical meaning before the literal sense,
as the former is used in the beginning chapters, may pose a problem to them.
Finally, as regards the distribution of metaphors in the different sections of
the lessons, i.e. the fourth research question, the results obtained are also sta-
tistically significant but in this case there are no similar studies with which
they can be compared. The explanation does not pose a problem in the case of
sections where metaphor density is relatively low. In these sections, the main
concern is with practical work, which means that in terms of language there is
greater use of a regulative register (Christie 2000). Students are given
 “The Manage of Two Kingdoms Must” 257

Table 8: Concordance of ‘kingdom’ in the two subcorpora.

Natural  tula belongs to the Animal Kingdom.  It 's usual

Science  d into five groups called kingdoms. Organisms of
 Organisms of the same kingdom share similarities
 t from organisms in other kingdoms. Plant Kingdom
 kingdoms. Plant Kingdom. Plants are multicellu
 MRW oxygen. Protist Kingdom. Protists are usually
 tosynthesis. Monera Kingdom. All organisms in this
 _All organisms in this kingdom are unicellular.

Social  r false . Visigothic Kingdom. The Middle Ages began in

Science  ished the capital of their kingdom in Toledo and adopted
 redo I, converted his kingdom to Catholicism during the
 iscuss any problems in the kingdom. The rest of the p
 tal city of the Visigothic Kingdom. c . The Vis
 ans invaded the Visigothic Kingdom in AD . San Isi
 ical system Christian Kingdoms The Reconquista began

instructions that need to be precise and factual. As a consequence, there is less

room for meaning displacements and for metaphor.
The reason behind the greater metaphor density of the introduction to the
books (Getting started), the warm-up section giving an overview of the unit’s topic
(Read & Discover) and the follow-up focus section on a specific issue related to the
subject matter (Fragile World) is less obvious. We would argue that it has to do
with the wider scope and the more abstract nature of these sections. It seems as if
their preparatory and exploratory nature calls for summarizing and generalizing
statements for which metaphor is particularly apt. See for example this Read &
Discover extract from unit 7, which is not particularly dense in terms of metaphor:

During the Middle Ages, different civilisations lived on the Iberian Peninsula. Visigoths
settled after the Roman Empire fell. Then Muslims invaded, calling their new territories
Al-Andalus. Here, three religious groups lived together: Muslims, Jews and Christians.
Over the next centuries, the Christians in the north fought against the Muslims. By the
end of 1492, the whole Iberian Peninsula was under Christian rule again. What was the
period before the Middle Ages called?

In this extract, the whole series of defeats and the collapse of the Roman army
and institutions, which developed over a long period of time, and the emer-
gence of a new social and political system after the Reconquista, which took
several centuries to enforce, are condensed by using a verb of motion (fell) and
a specific noun (rule). For their part, religious groups are personified (live) to
258 Rafael Alejo-González and Verónica García-Bermejo

indicate that they can have a particular kind of life whose limits go beyond in-
dividual human beings.
Finally, the content section is focussed on providing the basic definitions
and explanations of the topic and, as a consequence, tends to make use of spe-
cific vocabulary and terminology, which is sometimes metaphorical, together
with more general language used to perform the basic functions of scientific and
academic language (e.g., defining, classifying, describing, giving examples, com-
paring and contrasting).
It is interesting to note that, in spite of metaphor not being manifested espe-
cially frequently in this central section, it is here where we can find most of the
examples of direct metaphor. In fact, incorporated to all the content sections of
the Natural Science book we can find a subsection called In other words. This
subsection is nothing but an analogy expressed as a simile and intended to help
students understandthe concepts explained in the lesson. Here we include the
subsection in unit 4:

Chloroplasts are like solar panels. They capture solar energy and transform it. Instead of
electricity, chloroplasts produce glucose.

Typographically separated from the rest of the section and accompanied by a

photograph of solar panels, this subsection is clearly designed to explain some
of the concepts being used. Sometimes, it is the learners who have to provide
the explanation, as in this example from unit 1 from the Natural Science book:

A cell is like a city:

nucleus=town hall
vacuole=water tower
cytoplasm=atmosphere
cell membrane=guards
chloroplasts=parks
cell wall= city wall
What are the functions of the elements in a city listed above?

These are clearly examples of direct metaphors being used with a pedagogic or il-
lustrating function, but they are not used frequently (once in a whole unit) and are
not given particular prominence as they are placed in a corner of the content
pages. We can only find two other examples of metaphor being signposted. In a
less obvious and more condensed simile, a part of the cell, the cytoplasm, is de-
scribed as jelly-like and the leaves of a tree are referred to as needle-like. MacArthur
(2016) already pointed out the relatively low frequency of similes and direct meta-
phors in a corpus of office hour consultations (EuroCoAT). Skorczynska and
 “The Manage of Two Kingdoms Must” 259

Deignan (2006) obtained similar findings in other genres like research and periodi-
cal articles related to economics. They predicted, however, a higher proportion in
textbooks given their pedagogical goals. However, it seems that this is not the case
of CLIL textbooks.
All in all, indirect metaphor seems to be the order of the day, especially in
those sections less directly focussed on either the specific content (definitions
and explanations) or on the practical work students have to carry out. The
more prominent and more marked use of direct metaphor, however, should not
conceal its low frequency.

7 Conclusion
The present chapter has attempted to make a preliminary analysis of metaphor in
CLIL. This educational approach to bilingualism, mostly adopted in Europe, has
attracted a lot of attention from applied linguists, but up to now has been ne-
glected in the extensive literature on metaphor, which is increasingly covering a
wider range of communication contexts but has left this one unattended. The
study of metaphor in two CLIL textbooks has provided us with a preliminary pic-
ture of the workings of metaphor in this context and has shown that, although it
is not as frequently used as in academic prose (Herrmann 2013), the density is
quite high, especially if we compare it with the levels found by Cameron (2003)
for L1 educational contexts, which in a way points to the difficult task students
reading these textbooks are faced with. However, the textbooks show similarities
with some of the patterns found by Cameron (2003), as we found that they make
use of a great number metaphorical verbs, in contrast to the relatively low now
number of nouns, for example, which indicates a need to make the actions they
express more accessible to the students by personifying the abstract nominal
phrases used as their subjects. The picture of the use of metaphor in CLIL text-
books is completed by the study of the two main sources of variation we identi-
fied. On the one hand, the use of metaphor in the Natural Science textbook is
clearly higher than in the Social Science one, which indicates that even at these
initial stages of education the discourse strategies used are not homogenous and
that the different subjects are already beginning to impose their stamp on the lan-
guage used. On the other hand, we found that there are certain parts of the units
when metaphor is more frequent and the internal variation between the different
structural elements of the lesson evidences the need to pay more attention to
those areas where the use of metaphor may go unnoticed, like for example the
introduction of lessons. These conclusions should, however, be taken with
260 Rafael Alejo-González and Verónica García-Bermejo

caution as the study is only preliminary and another study with a corpus includ-
ing a greater number of books and lessons would be needed.

References
Berber Sardinha, Tony. 2015. Register variation and metaphor use: A multi-dimensional
perspective. In Berenike Herrmann & Tony Berber Sardinha (eds.), Metaphor in specialist
discourse, 17–52. Amsterdam: John Benjamins.
Biber, Douglas. 1988. Variation across speech and writing. Cambridge: Cambridge University
Press.
Biber, Douglas. 2006. University language: A corpus-based study of spoken and written
registers. Amsterdam: John Benjamins.
Biber, Douglas & Susan Conrad. 2009. Register, genre and style. Cambridge: Cambridge
University Press.
Caballero, Rosario. 2003. Metaphor and genre: The presence and role of metaphor in the
building review. Applied Linguistics 24(2). 145–67.
Caballero, Rosario. 2017. Genre and metaphor: Use and variation across usage events.
In Elena Semino & Zoltán Demjén (eds.), The Routledge handbook of metaphor and
language, 193–218. London: Routledge.
Caballero, Rosario. 2012. The role of metaphor in tennis reports and forums. Text & Talk
32 (6). 703–726.
Cameron, Lynne. 2003. Metaphor in educational discourse. London: Continuum
Cameron, Lynne & Juurd H. Stelma. 2004. Metaphor clusters in discourse. Journal of Applied
Linguistics 1(2). 107–36.
Charteris-Black, Jonathan. 2004. Why “an angel rides in the whirlwind and directs the storm”:
A corpus-based comparative study of metaphor in British and American political
discourse. Language and Computers 49. 133–50.
Christie, Frances. 2000. The pedagogic device and the teaching of English. In Frances Christie
(ed.), Pedagogy and the shaping of consciousness, 156–184. London: Continuum.
Conrad, Susan. 2001. Variation among disciplinary texts. A comparison of textbooks and
journal articles in Biology and History. In Susan Conrad & Douglas Biber (eds), Variation
in English: Multidimensional studies, 94–107. London: Routledge.
Corts, Daniel P. & Howard R. Pollio. 1999. Spontaneous production of figurative language and
gesture in college lectures. Metaphor and Symbol 14(2). 81–100.
Coyle, Do, Philip Hood & David Marsh. 2010. CLIL: Content and language integrated learning.
Cambridge: Cambridge University Press.
Dalton-Puffer, Christiane. 2007. Discourse in content and language integrated learning (CLIL)
classrooms. Amsterdam/Philadelpia: John Benjamins.
Deignan, Alice, Jeannette Littlemore & Elena Semino. 2013. Figurative language, genre and
register. Cambridge: Cambridge University Press.
Dorst, Aletta G. 2011. Metaphor in Fiction: Language, thought and communication. Oisterwijk:
Box Press Uitgeverij.
Dorst, Aletta G. 2017. Textual patterning of metaphor. In Elena Semino and Zoltán Demjén
(eds.), The Routledge handbook of metaphor and language, 178–92. London: Routledge.
 “The Manage of Two Kingdoms Must” 261

Fernández-Sanjurjo, Javier, Alberto Fernández-Costales & José Miguel Arias Blanco. 2017.
Analysing students’ content-learning in science in CLIL vs. non-CLIL programmes: Empirical
evidence from Spain, International Journal of Bilingual Education and Bilingualism. 1–14.
García-Bermejo, Verónica. 2015. Análisis del vocabulario en libros de texto para la enseñanza
AICLE en 5° de educación primaria. Campo Abierto. Revista de Educación 34 (1). 29–47.
García-Mayo, María del Pilar & María Basterrechea. 2017. CLIL and SLA. In Ana Llinares & Tom
Morton (eds.), Applied Linguistics Perspectives on CLIL, 33–50. Amsterdam: John Benjamins.
Herrmann, Berenike. 2013. Metaphor in academic discourse: Linguistic forms, conceptual
structures, communicative functions and cognitive representations. Utrecht: LOT
dissertation series.
Herrmann, Berenike. 2015. High on metaphor, low on simile? An examination of metaphor type
in sub-registers of academic prose. In Berenike Herrmann & Tony Berber Sardinha (eds.),
Metaphor in specialist discourse, 163–190. Amsterdam: John Benjamins.
Hoang, Ha & Frank Boers. 2018. Gauging the association of EFL learners’ writing proficiency
and their use of metaphorical language. System 74. 1–8.
Järvinen, Heini-Marja. 2010. Language as a meaning making resource in learning and teaching
content. in Christiane Dalton-Puffer, Tarja Nikula & Ute Smit (eds.), Language use and
language learning in CLIL classrooms,145–168. Amsterdam: John Benjamins.
Kaal, Anna A. 2012. Metaphor in conversation. Oisterwijk: Uitgeverij BOX Press.
Koller, Veronika. 2006. Of critical importance: Using corpora to study metaphor in business
media discourse. In Anatol Stefanowitsch & Stefan T. Gries (eds.), Corpus-based
approaches to metaphor and metonymy, 229–257. Berlin: Mouton de Gruyter.
Krennmayr, Tina. 2011. Metaphor in newspapers. Utrecht: LOT dissertation series.
Krennmayr, Tina. 2017. Metaphor and parts-of-speech. In Elena Semino and Zoltán Demjén
(eds.), The routledge handbook of metaphor and language, 165–177. London: Routledge.
Lakoff, George. 1989. Women, fire and dangerous things: What categories reveal about the
mind. Chicago: University of Chicago Press.
Littlemore, Jeannette. 2001. The use of metaphor in university lectures and the problems that
it causes for overseas students. Teaching in Higher Education 6(3). 333–49.
Littlemore, Jeannete, Phillys Chen, John Barnden & Almut Koester. 2011. Difficulties in
metaphor comprehension faced by international students whose first language is not
English. Applied Linguistics 32(3). 408–29.
Littlemore, Jeannette, Tina Krennmayr, James Turner & Sara Turner. 2014. An investigation into
metaphor use at different levels of second language writing. Applied Linguistics 35(2).
117–144.
Llinares, Ana, Tom Morton & Rachel Whittaker. 2012. The roles of languages in CLIL.
Cambridge: Cambridge University Press.
Low, Graham D., Jeannette Littlemore & Almut. Koester. 2008. Metaphor use in three UK
university lectures. Applied Linguistics 29(3). 428–55.
MacArthur, Fiona. 2015. On using a dictionary to identify the basic sense of words. Metaphor
and the Social World. 5(1). 124–136.
MacArthur, Fiona. 2016. Overt and covert uses of metaphor in the academic mentoring in
English of Spanish undergraduate students at five European universities. Review of
Cognitive Linguistics 14(1). 23–50.
MacArthur, Fiona. 2017. Using metaphor in the teaching of second/foreign language. In Elena
Semino & Zoltán Demjén (eds.), The Routledge handbook of metaphor and language,
193–218. London: Routledge.
262 Rafael Alejo-González and Verónica García-Bermejo

MacArthur, Fiona, Tina Krennmayr & Jeannette Littlemore. 2015. How basic Is UNDERSTANDING
IS SEEING when reasoning about knowledge? Asymmetric uses of SIGHT metaphors in
office hours’ consultations in English as academic lingua franca. Metaphor and Symbol
30(3). 184–217.
Marsh, David. 2008. Language awareness and CLIL. In Jasone Cenoz & Nancy Hornberger
(eds): Encyclopedia of Language and Education, vol 6. Knowledge about language.
233–46. Boston, MA: Springer.
Nacey, Susan. 2013. Metaphors in learner English. Amsterdam: John Benjamins.
Parodi, Giovanni. 2010. The rhetorical organization of the textbook genre across disciplines:
A ‘colony-in-loops’? Discourse Studies 12(2). 195–222.
Pérez-Cañado, María Luisa. 2012. CLIL research in Europe: Past, present and future.
International Journal of Bilingual Education and Multilingualism 15(3). 315–41.
Pérez-Sobrino, Paula. 2016. Multimodal metaphor and metonymy in advertising: A corpus-
based account. Metaphor and Symbol 31(2). 73–90.
Piquer-Píriz, Ana M. 2008. Young learners’ understanding of figurative language. In Maria
S. Zanotto, Lynne Cameron & Marilda C. Cavalcanti (eds.), Confronting metaphor in use:
An applied linguistic approach, 183–198. Amsterdam: John Benjamins.
Rayson, Paul. 2008. Wmatrix: A web-based corpus processing environment (version 2009).
Lancaster University: Computing department.
Skorczynska, Hanna & Alice Deignan. 2006. Readership and purpose in the choice of
economics metaphors. Metaphor and Symbol 21(2). 87–104.
Steen, Gerard J., Aletta G. Dorst, Berenike Herrmann, Anna A. Kaal, Tina Krennmayr & Trijntje
Pasma. 2010. A method for linguistic metaphor identification: From MIP to MIPVU.
Amsterdam: John Benjamins.
Tyler, Andrea & Vyvyan Evans. 2003. The semantics of English prepositions explained.
Cambridge: Cambridge University Press.
Wan, Wan & Graham D. Low. 2015. Elicited metaphor analysis in educational discourse.
Amsterdam: John Benjamins.
Zeeuw Marlies de, Franc Grootjen, Gerrit Jan Kootstra & Agnes Tellings. 2019. Lexical
characteristics of written language input across primary grades: An analysis of a Dutch
corpus based lexicon. Linguistics and Education 49, 11–21.
Rawan A. Saaty
An Enactment-Based Approach to the
Teaching of Metaphoric Expressions:
A Case of Saudi EFL Learners
1 Introduction
Metaphor awareness-raising activities which are based on enactment involve
promoting awareness of the embodied motivations behind conceptual meta-
phors. This approach to the teaching of EFL metaphors draws on insights from
embodied cognition research (Barsalou 2009; Gibbs 2006), which views embod-
ied metaphors as grounded in sensorimotor experience. Gibbs (2014) described
embodied metaphors as being more fundamental than conceptual metaphors
because they unfold through human actions and interactions with the world.
This study is based on the hypothesis that promoting awareness of embodied
metaphor in the EFL/ESL classroom through actions can improve the learning
of related metaphoric expressions. To investigate this, the study examines the
impact of enactment-based metaphor awareness on learners’ comprehension
and retention of the meaning of 11 metaphoric expressions instantiating the
conceptual metaphor LIFE IS A JOURNEY. It also explores whether the interven-
tion leads to productive use of the target vocabulary in writing.
As outlined by Lindstromberg and Boers (2005), enactment-based metaphor
awareness involves encouraging language learners to enact the literal meanings
of metaphorically used lexical items. Lindstromberg and Boers found inspiration
in Asher’s ([1977] 2012) Total Physical Response (TPR) for the teaching of meta-
phoric expressions as it allows for the natural acquisition of vocabulary in a phys-
ically active environment. Through the use of repetitive commands, learners
experience the physical meaning of vocabulary from hearing and performing it
multiple times, thereby taking an active role in vocabulary understanding (Asher
([1977] 2012). Applied to enactment-based metaphor awareness, learners can act
out ‘sailing through life’ by waving their hands forwards to help them internalise
the image of sailing in terms of life. Ray and Sealy’s ([1998] 2004) advanced ver-
sion of TPR, Teaching Proficiency through Reading and Storytelling (TPR
Storytelling), relies on students enacting short stories rather than single words
and allows them to learn vocabulary items in context. Both variations of TPR are

Rawan A. Saaty, English Language Institute, King Abdulaziz University, Jeddah,

Saudi Arabia, e-mail: [email protected]

https://doi.org/10.1515/9783110630367-013
264 Rawan A. Saaty

employed in this study, so metaphoric expressions are learned both individually

and in context.

2 Awareness of Conceptual Metaphor and

Semantic Clustering
Figurative language teaching research has raised awareness of conceptual meta-
phor in L2 and guided this awareness through guessing strategies (Skoufaki
2008), etymological elaboration (Boers, Demecheleer, and Eyckmans 2004;
Boers, Eyckmans, and Stengers 2007) and pictorial elucidation (Boers et al. 2008;
Boers et al. 2009) among others. In an experimental study with 30 EFL Greek
learners, Skoufaki (2008) employed learners’ problem-solving skills to promote
awareness of conceptual metaphors. The study involved three metaphor groups.
Group 2 (the metaphor-guessing group) obtained meaning clues that helped
them guess the relationship between the target idioms and conceptual meta-
phors. Groups 1 and 3, on the other hand, received conceptual metaphor aware-
ness training minus the guessing, and differed only in that group 1 focused on
meaning while group 3 focused on form. The metaphor-guessing group received
higher scores than the metaphor-only groups in a cloze test (4.30 versus 1.60 and
1.90) as well as a comprehension test (62.50% versus 50% and 42.50%). Skoufaki
concluded that the cognitive effort used in mental reasoning benefited the learn-
ers’ retention of the metaphoric expressions. In another study, Boers, Demecheleer
and Eyckmans (2004) employed etymological elaboration to promote awareness of
the literal, historical and cultural origins of idioms. Results from the 2002 experi-
ment indicated that the etymology group outperformed the control group by 11%
in post-tests which required students to recall the key content words of the idioms
in a blank-fill test format.
While figurative language teaching research calls for promoting awareness of
metaphor in language teaching, most current EFL/ESL textbooks present meta-
phoric vocabulary in general semantic clusters with no reference to the metaphori-
cal motivations behind them. Wilcox and Medina (2013: 1057) define semantic
clustering as the presentation of vocabulary items as “groups of words whose
meaning would fall under one superordinate concept” without further indication
of the items’ contextual use or cognitive motivation. Tinkham (1993) argued that
textbooks often present new vocabulary items under semantic clusters, such as
words for animals. He clarified that although learners would be able to memorize
such lists, they would struggle to differentiate between the meanings of individual
words, to employ those words communicatively, and to retain the vocabulary.
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 265

So far, figurative language teaching research has been confined to aware-

ness-raising activities involving conceptual metaphor, which according to
Boers (2004), do not guarantee that learners will be capable of producing the
metaphoric expressions in natural contexts. Boers (2004) also noted that due to
a lack of longitudinal investigation, it is not clear whether conceptual metaphor
awareness would provide long-term mnemonic benefits. Figurative language
teaching research is also limited in that it can view the language classroom as a
static environment. In her dynamic system view of the language classroom,
MacArthur (2010: 158) argued that language learning should be viewed as a com-
plex phenomenon in which the replication of results of controlled classroom ex-
perimental research would be “variable, and no one method or technique will
work for all learners; initial states, cognitive styles, mood swings, the L1 of the
learners, group dynamics, and a host of other variables will be working together
to affect the developing system”. Thus, even explicit teaching of metaphoric ex-
pressions may not offer satisfactory results without considerable attention to the
learners’ needs and abilities.

3 Awareness of the Embodied Nature

of Metaphors through Actions
The sensorimotor experiences that we use to interact with the world provide
grounding for our language and thought. Grady (1997) and Gibbs (2006) viewed
conceptual metaphors as being rooted in bodily and sensorimotor experiences
through primary metaphors. For example, Grady (1997) classified the concep-
tual metaphor LIFE IS A JOURNEY as a complex metaphor that inherits its map-
pings from primary metaphors such as PURPOSES ARE DESTINATIONS and
CIRCUMSTANCES ARE SURROUNDINGS which are grounded in physical reality.
Gibbs (2006) noted that since many metaphoric expressions are also rooted in
bodily experiences, they can be understood through awareness of their sensori-
motor links. To learn metaphoric expressions such as to take a step forward and
to climb a career ladder, learners may be asked to stand up and take steps with
their feet and gesture with their hands as if they were climbing a ladder.
While figurative language teaching research is yet to explore the extent of
enactment-based metaphor awareness, research carried out in the fields of psy-
chology, neuroscience, and cognitive linguistics has enhanced our understand-
ing of the potential of embodied metaphors. In behavioural psychology, research
has employed embodied metaphors to modulate suspicion (Lee and Schwarz
2012) and influence social proximity (IJzerman and Semin 2009). For example,
266 Rawan A. Saaty

Lee and Schwarz’ (2012) experiments on the olfactory experience of fishy smells
and social suspicion indicated that exposing participants to fishy smells pro-
vokes suspicion and influences financial distrust. At the same time, the findings
of this study indicated that encouraging participants to be suspicious heightens
their ability to identify fishy smells. This bi-directionality of metaphor can be
beneficial in the teaching of L2 metaphors, where the embodied sensorimotor
motivations of primary metaphors can be manipulated to promote awareness of
abstract target domains.
In neuro-cognition research, Desai et al. (2011) and De Grauwe et al. (2014)
suggested dual linguistic and motor processing of action metaphors in L1 and L2.
As to L1, Desai et al. (2011) compared the mental activation of literal, abstract, and
metaphoric sentences through the use of functional magnetic resonance imaging
(fMRI). Participants were instructed to think of the meanings of sentences as they
read them. Metaphoric sentences with action metaphors (e.g. the jury grasped the
idea) were processed similarly to physical action words (e.g. the daughter grasped
the flowers). fMRI results indicated that both literal and metaphoric sentences acti-
vated the left anterior inferior partial lobule, which is an area of the brain generally
activated when planning concrete actions. In addition, metaphoric sentences also
activated the left superior temporal regions, which are activated with abstract sen-
tences. Regarding metaphor processing in L2, Xue et al. (2014) compared Chinese-
English bilinguals’ understanding of spatial-time metaphors. Results indicated
that the mental processing of L1 and L2 time metaphors involved different sensori-
motor simulations for each language, suggesting that even in their L2, bilinguals
processed metaphors through embodied grounding.
In the cognitive linguistic paradigm, studies that have investigated enactment
of the embodied nature of metaphors were conducted in L1 (e.g. Gibbs and
Perlman 2006; Wilson and Gibbs 2007; Gibbs 2013). Studies that investigate enact-
ment-based metaphor awareness in L2 are scarce. Concerning L1, Gibbs (2013) ex-
amined the understanding of two groups of participants with regards to the
metaphor LOVE IS A JOURNEY through actions. Gibbs asked participants to walk
across a field blindfolded after they had listened to successful and failing relation-
ship stories. Those who listened to the successful love story walked faster and fur-
ther than those who had heard the failing love narrative. Gibbs’ findings suggest
that employing the embodied nature of metaphor influences related actions.
Regarding enactment-based metaphor awareness in L2, Lindstromberg and Boers
(2005) employed enactment and mime to teach English action verbs to Dutch uni-
versity learners who were divided into enactment groups and comparison groups.
Participants in the enactment groups played charades by miming metaphoric ac-
tion verbs, such as leap and pounce, and the other students guessed the verbs.
Participants in the comparison groups prompted guesses using only verbal
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 267

paraphrases of the verbs. The learners in the enactment groups performed better
in immediate post-tests which required students to supply the missing verbs in
gapped sentences than the learners who had learned the vocabulary items through
verbal representations. One week later, the students were given a new test in
which the same action verbs were presented in contexts where they were used
metaphorically. Accompanying these metaphorical uses were L1 translations,
many of which did not fully capture the precise meaning of the verbs. The stu-
dents’ task was to evaluate these translations. Also, on this test the students who
had learned the literal meanings of the verbs through enactment performed better
than their peers who had learned the verbs only through verbal representations.
Lindstromberg and Boers (2005) concluded that actions and motoric imagery
benefited not only the students’ retention of the literal meanings of the action
verbs, but also helped them to more fully appreciate their meaning when used
metaphorically. While Lindstromberg and Boers’ study highlighted the benefits of
enactment-based metaphor awareness, the study did not investigate the impact of
this teaching technique beyond a one-week delayed test.
Employing enactment-based metaphor awareness to teach metaphoric expres-
sions in L2 thus offers possible mnemonic benefits. Cognition research (Kelly et al.
2009; Macedonia et al. 2011) in support of Engelkamp and Krumnacker’s (1980)
enactment effect hypothesis indicated that enacting abstract words promotes re-
tention for longer periods than verbal or auditory modalities in L2. For example,
Macedonia and Klimesch (2014) performed a longitudinal 14-month study testing
participants’ ability to commit artificial abstract vocabulary to their long-term
memory after a lesson that involved gesture and enactment. Memory performance
was assessed at five time points. At each time point, enacted vocabulary items
were remembered significantly more often than vocabulary items explained
through audio-visual modalities. Since metaphoric vocabulary describes abstract
concepts, cognition research serves as further support of the usefulness of enact-
ment in figurative language teaching. Together, the findings of cognitive psychol-
ogy, neuro-cognition, cognitive linguistics and cognition research are valuable
starting points for an investigation of enactment-based metaphor awareness in lan-
guage learning.

4 Production of Metaphoric Expressions

Only a handful of studies have investigated interventional methods in promot-
ing production of metaphoric expressions. Boers (2004) noted that this lack of
focus is partly due to a consensus amongst cognitive linguists that metaphor
268 Rawan A. Saaty

awareness-raising activities, while useful as tools for helping learners compre-

hend and retain expressions they are presented with, are limited as generative
methods. MacArthur (2010) suggested promoting the production of metaphor
through the use of written and spoken texts, dictionaries, and corpora, which
could be employed both inside and outside the classroom. However, studies in-
vestigating such techniques are scarce.
Boers (2000) performed one of the few studies that have investigated learn-
ers’ production of metaphoric expressions in writing. In his study, he taught 73
French business students vocabulary items on ‘upward’ and ‘downward’ eco-
nomic trends. The experimental group was made aware of the nature of concep-
tual metaphors, whereas the control group was not. The learners were then
encouraged to use the vocabulary items they had been taught in a short essay
to describe graphs on economic growth. The experimental group produced
more of the targeted vocabulary items than the control group (7.1 versus 4.9 in
average). However, Boers (2013) advised that his early (2000) study should be
viewed with caution because no pre-test was administered.
Viewing the topic of vocabulary production in Second Language Acquisition
(SLA) research helps in clarifying difficulties faced in promoting metaphor produc-
tion. Nation ([2001] 2013) viewed vocabulary acquisition in terms of the receptive-
productive lexical knowledge continuum, where understanding of vocabulary
items is placed at one end of the continuum and production of the items is at the
other end. In most models of vocabulary acquisition (Laufer 1998; Nation 2013),
lexical knowledge begins with receptive knowledge, or the ability to understand a
word when it is heard or read (passive vocabulary), and ends with productive
knowledge, or the ability to use the word in writing or speaking (active vocabu-
lary). Passive vocabulary knowledge develops at a faster and more systematic rate
than the much more limited active knowledge. Placing metaphoric vocabulary on
the active-passive vocabulary continuum can explain why promoting the produc-
tion of taught metaphoric expressions is so difficult.
In summary, although a gap exists in empirical classroom testing, the research
conducted suggests that exploiting the embodied nature of metaphors through ac-
tions in the teaching of L2 metaphor can lead to longer-term retention. Promoting
awareness of the bodily motivations behind a conceptual metaphor may not only
clarify the metaphor’s perceptual elements but also help learners retain associated
metaphoric expressions. Thus, by shifting the focus from metaphor understanding
to retention and from conceptual metaphor awareness to enactment-based meta-
phor awareness, we can broaden the scope of figurative language teaching
research. In addition, MacArthur’s (2010) dynamic system view is considered in
the experimental design of this study so that it is closer to authentic classroom
teaching. Instead of taking a standalone teaching approach to metaphor, the
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 269

teaching of metaphor is embedded in teaching methodologies adopted as frames

for the interventional teaching sessions. The teaching methodologies are Asher’s
(2012) TPR, Ray and Sealy’s (2004) TPR storytelling and Presentation – Practice –
Production (PPP) teaching approach. Also, the metaphoric expressions, teaching
materials and tests were chosen and developed from authentic sources and online
corpora. Moreover, since the difficulties of promoting the production of taught
metaphoric expressions have not been thoroughly investigated in cognitive lin-
guistics, this study also explores this difficulty in a classroom-based environment
to further support the notion that metaphor awareness-raising activities may not
BE sufficient to promote metaphor production.

5 Research Question and Aim

The experimental study seeks to answer the research question: How do 60 female
Saudi EFL learners respond to awareness-raising activities in the teaching of meta-
phoric expressions when these activities are based on enactment-based metaphor
awareness, conceptual metaphor awareness, and semantic clustering? This study
investigates the use of bodily actions to make EFL learners aware of the embodied
nature of the conceptual metaphor LIFE IS A JOURNEY and to help these learners
understand the sensorimotor motivations of metaphoric expressions.

6 Method
The 5-week experimental study involved 60 female Saudi EFL students. The stu-
dents were divided into an enactment metaphor group, a conceptual metaphor
group, and a control group. Data collection involved metaphor comprehension
tests and metaphor production tests.

6.1 Participants

The participants of this study were female Saudi university students aged 18 to
20 years old whose first language is Arabic. All participants were in their
foundation year at a Saudi university studying at the university’s English
Language Institute. The courses at the English Language Institute teach stu-
dents who are placed at A1, A2, B1, B2 levels as described by the Common
European Framework of Reference (CEFR) for Languages. These courses are
270 Rawan A. Saaty

intensive 7-week courses that require students to take EFL classes for 18 hours
per week. Two weeks prior to the EFL course, students undertook the Oxford
Online Placement Test and were placed at the B2 upper-intermediate level.
Three B2 level classes, which ranged from 31 to 32 students each, were chosen
at random. Participants signed consent forms that ensured privacy and ano-
nymity. Participants were then assigned as the enactment metaphor group (31
students), the conceptual metaphor group (32 students), and the control group
(31 students). As some students were absent in some of the sessions, their data
were excluded from the study. In the 5-week period of the study, 25 participants
remained in the enactment metaphor group (mean age= 18.36), 18 participants
remained in the conceptual metaphor group (mean age= 18.47), and 17 partici-
pants (mean age= 19.52) remained in the control group.

6.2 Selection of Metaphoric Expressions

The target metaphoric expressions for this study are 11 metaphoric expressions
that have metaphoric and literal relations to the conceptual metaphor LIFE IS A
JOURNEY. These metaphoric expressions include verbs, nouns, multi-word
units, and collocations. I narrowed the number of vocabulary items to 11 meta-
phoric expressions because the teaching methodology TPR (Asher [1977] 2012)
is a teaching methodology that focuses on the quality of vocabulary taught in-
stead of the quantity. Asher ([1977] 2012) recommends that the target vocabu-
lary in a TPR lesson should be kept fairly small in order to make sure that
students participate in the actions and gestures, so that every vocabulary item
is thoroughly practised by every student during class time.
I chose the metaphoric expressions according to three criteria. First, I adopted
some of Boers and Lindstromberg’s (2008) criteria for choosing metaphoric ex-
pressions for teaching. Namely, I selected the metaphoric expressions according
to their relevance to students in their foundation year and how useful the meta-
phoric expressions can be to them. I also selected metaphoric expressions based
on coverage and chose items whose meanings are more general (e.g. a path). I
selected items based on how wide the range of the expression is in the discourse
of life experiences (e.g. a step in the right direction). In addition, I selected colloca-
tions of different levels of difficulty and frequency to expose the participants to
different frequencies of vocabulary. I asked participants in the piloting stage to
rate the difficulty of the piloted metaphoric expressions and I chose items with
varied difficulty ratings.
Second, I consulted Macmillan and Collins Online dictionaries to ensure au-
thentic use of the chosen metaphoric expressions and that both the metaphoric
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 271

and literal meanings of metaphoric expressions are documented uses. Third, I

collected the list of metaphoric expressions from Lazar’s (2003) Meanings and
Metaphors textbook and from websites that discussed life experiences. I cross-
checked the initial set of metaphoric expressions with the British National
Corpus (the BNC) for frequency and kept only those expressions with a statisti-
cally significant occurrence rate (i.e. above 2.00 frequency band T-score). This
technique for the selection of metaphoric expressions is based on the method-
ology outlined by Walker (2008), who explains that when the collocate has a
T-score of 2.00 or above, the combination of words is statistically significant
and did not occur by chance. With multi-word expressions such as to take a
step forward, the T-score is extracted for the entire expression; and if the ex-
pression contained a verb, different verb forms are taken into consideration.
The initial list of metaphoric expressions included 15 metaphoric expres-
sions. Piloting was performed with 18 language learners and 7 British speakers,
and based on its results, the most frequently answered 11 vocabulary items
were chosen. As stated earlier, different levels of difficulty were taken into con-
sideration as well. Table 1 provides the final set of metaphoric expressions
taught per teaching week and their frequency collocations in the BNC:

Table 1: Metaphoric expressions and their frequencies in the BNC.

Week  BNC Week  BNC

metaphoric expressions T-score metaphoric expressions T-score

to take a step forward . to climb .

to overcome an obstacle . in the fast lane .
to follow in someone’s footsteps . a step in the right direction .
a path . to stumble into .
at a crossroads . a dead-end job .
up the career ladder .

6.3 Experiment Time Frame

The study was designed as a 5-week experimental study to run alongside a 7-week
teaching programme at the Saudi university. I coordinated with the teachers of
the participant groups to spend 6 teaching-hours with the students during week 2
and week 3 (three 1-hour sessions per week). Table 2 presents the timeframe for
each group, which is detailed in the following section.
272 Rawan A. Saaty

Table 2: Experiment time frame for participant groups.

Control group Conceptual metaphor Enactment metaphor

group group

Week  Consent form

Metaphor comprehension pre-test
Metaphor production pre-test

Week  Semantic clustering Conceptual metaphor Enactment-based

Reading a metaphoric awareness metaphor awareness
short story Reading metaphoric and TPR Storytelling of
literal short stories metaphoric and literal
short stories

Week  Semantic clustering Conceptual metaphor Enactment-based

Reading a metaphoric awareness metaphor awareness
short story Reading metaphoric and TPR Storytelling of
literal short stories metaphoric and literal
short stories

Metaphor comprehension post-test

Week  Metaphor production post-test

Week  Metaphor comprehension -week delayed test

6.4 Interventional Teaching

I designed teaching materials describing ‘the journey of life’ to teach the 11 meta-
phoric expressions surrounding LIFE IS A JOURNEY. Week 2 focused on teaching
five metaphoric expressions, and week 3 focused on teaching six metaphoric ex-
pressions. The following subsections detail the interventional teaching for partic-
ipant groups.

6.4.1 Enactment Metaphor Group

The enactment metaphor group learned the literal and figurative senses of the
expressions through enactment and mime. Day one of each week was spent in-
troducing the literal and metaphoric meanings of the metaphoric expressions
which were obtained from Macmillan and Collins Online dictionaries. Students
also elaborated on the relationships between the metaphoric expressions and
the LIFE IS A JOURNEY metaphor. For example, I explained ‘to climb’ in both
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 273

its metaphoric and literal senses and then asked the students to guess the simi-
larities between climbing social ranks and physical journeys. Students brain-
stormed ideas about hiking and mountain climbing and related them to
overcoming difficulties in one’s career or education. Students were then asked
to enact each vocabulary item in its literal sense and follow simple TPR com-
mands. For example, to enact the phrase to overcome an obstacle, a group of
learners were instructed to stand together and form an obstacle to prevent
others from passing through. Another student would then attempt to overcome
the obstacle. Then, day two followed Ray and Seely’s ([1998] 2004) TPR
Storytelling. The students read and acted out a literal journey story and a figu-
rative journey story. Snapshots of the stories are presented in Figures 1 and 2
below. Finally, students spent Day three of every week repeating the commands
and TPR stories. They also created and acted out their own TPR stories employ-
ing the taught metaphoric expressions.

Directions to the Grocery Store

When we moved to our new house, my mom wanted me to go to the grocery store to get us
some food and supplies. I did not know the area yet, so she gave me some simple directions.
She said to take my first step forward out of the house then to avoid the busy street by
taking the green footpath to the left instead. At the crossroads, I should turn left again. She
warned me to watch out for a small stream along the path.

Figure 1: A sample of the literal story.

Sarah’s Career Path

I’d like to tell you the story of how I became a painter. After I finished high school, I started
to go down the path of earning my medical degree. This wasn’t what I wanted, but I felt like
I had to follow in my parents’ footsteps. The first step in medical school was to take a science
placement exam. I failed it miserably! My instructors told me that medicine may not be the
right path for me, but they still offered to allow me to retake the test in a week.

Figure 2: A sample of the figurative story.

6.4.2 Conceptual Metaphor Group

Students in the conceptual metaphor group learned the same 11 metaphoric ex-
pressions in the framework of conceptual metaphor awareness through the PPP
teaching methodology. The three one-hour teaching sessions followed the
274 Rawan A. Saaty

phases of PPP across the two teaching weeks. On day one (Presentation Phase),
the literal and figurative meanings of the first five metaphoric items were dem-
onstrated through conceptual metaphor awareness. Similar to the enactment
metaphor group, learners in the conceptual metaphor group learned the dictio-
nary definitions of the metaphoric expressions. They were also encouraged to
guess and create links between the source and target domains beyond the lit-
eral meanings of the metaphoric expressions. However, they were not asked to
enact the metaphoric expressions as the intervention for the conceptual meta-
phor group targets only verbal explanations of the conceptual metaphor. For
instance, ‘a path’ was explained as a way from one place to another that people
can walk along’ as well as ‘the way that someone’s life develops’. Students
were also encouraged to come up with ideas on how literal and figurative paths
relate to each other and how a path can be a part of one’s life journey. On day
two (Practice Phase), the students read the two short stories and they were
given worksheets to practice the learned materials. Students then extracted the
metaphoric expressions from the two stories and compared the use of their lit-
eral and metaphoric meanings. On day three (Production Phase), the students
practiced integrating the taught metaphoric expressions in their own speech by
creating their own stories.

6.4.3 Control Group

Participants in the control group dealt with the metaphoric expressions only as
figurative expressions according to the theme ‘travelling through life’. The aim of
the control group was not to remove the metaphor entirely, but to focus the stu-
dents’ attention on the metaphoric expressions without their conceptual motiva-
tions. As a control measure, the teaching of the control group did not involve
referencing conceptual metaphor awareness, the source domain JOURNEY, or en-
actment; it only involved semantic clustering of the 11 metaphoric expressions.
The PPP teaching methodology was employed as parallel to the intervention
given to the conceptual metaphor group. On day one (Presentation Phase), the
figurative meanings of the metaphoric expressions, which were extracted from
Macmillan and Collins Online dictionaries, were introduced within the domain of
life. On day two (Practice Phase), students read the figurative story, did the prac-
tice worksheets, extracted the metaphoric expressions from the text, and dis-
cussed use of the expressions within the context of life experiences. The literal
stories did not constitute a part of the control group’s treatment as the compari-
son between the two stories might have led the students to notice the source do-
main JOURNEY. Day three (Production Phase) was a revision day of the
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 275

metaphoric expressions and figurative stories in which the students practiced

using the metaphoric expressions in their own stories.
In relation to treatment design, the quality of teaching received by the two
experimental groups and the control group was kept as similar as possible ex-
cept for the variables tested. The three groups received the same amount of
training time (six 1-hour sessions) and the same teaching materials (11 vocabu-
lary items and two figurative short stories). The only differences between the
groups were that they underwent different types of treatment according to the
aims of the study. Only the experimental groups were given literal short stories,
as the experimental groups’ training required awareness of the source domain
JOURNEY, and only the enactment metaphor group enacted the metaphoric ex-
pressions. This difference of treatment between the three groups could raise
some questions with regards to the different variables introduced to each
group. There was a concern about the higher cognitive effort exerted by the en-
actment metaphor group, who used actions and gestures to internalize the met-
aphoric senses of the vocabulary, as opposed to the effort required of the
control group and the conceptual metaphor group whose teaching was mainly
teacher-centered. To this we can say that requiring this cognitive effort on the
part of the enactment metaphor group was intentional and gives more support
to the hypothesis that employing enactment-based metaphor awareness leads
to better understanding and retention. In addition, since two teaching method-
ologies (TPR and PPP) were employed across the participant groups, these
methods are considered as variables for participant groups along with the met-
aphor awareness-raising activities.
Following the experiment completion and to ensure that all groups were
treated in an ethical manner, I went back to the control group and the conceptual
metaphor group and gave them a taste of the teaching of metaphoric expressions
through enactment-based metaphor awareness. As to the control group, I intro-
duced the conceptual metaphor LIFE IS A JOURNEY and reconnected the 11 vo-
cabulary items to their source and target domains as well as their literal senses.
Then, I had the learners enact the vocabulary and stories. As to the conceptual
metaphor group, I had them enact the 11 vocabulary items and the four stories
since they already learned them through conceptual metaphor awareness.

6.5 Metaphor Comprehension Tests

I designed the metaphor comprehension tests as multiple-choice questions that

measured the students’ understanding of the different meanings of the taught
metaphoric expressions. Each question in the pre-test, the post-test, and the 2-
276 Rawan A. Saaty

week delayed test consisted of a statement that used the vocabulary in either
its metaphoric or literal sense. Each test included 11 questions; half of the tested
items were presented in their metaphoric sense while the other half were pre-
sented in their literal sense totalling in five literal items and six metaphoric
items or vice versa. For example, the following two sentences from the pre-test
and the post-test use the term crosswords both literally and metaphorically:
– Metaphoric: After she had finished her master’s degree, she was at a cross-
roads. She could pursue a PhD degree or find a job to establish her career.
– Literal: To get to the Children’s Hospital, drive down Second Street until
you are at the crossroads, then turn left.

I selected the metaphoric and literal statements in the tests using the BNC and
simplified them to suit learners’ language levels. I also selected the possible lit-
eral and metaphoric answers using Macmillan and Collins online dictionaries.
Figure 3 shows that the multiple-choice answers comprised an incorrect an-
swer, a literal sense, a metaphoric sense, an option for other possible mean-
ings, and an option for a lack of knowledge.

10. Hoping to take over the family business one day, Mary followed in her father’s footseps
and became a lawyer. The problem is that she does not like being a lawyer at all.

A. To carry a heavy object for a long period of time

B. To walk or drive behind someone or to go in the same direction as them
C. To do the same work or achieve the same success as someone else before you
D. It means something else:
E . I do not know

Figure 3: A sample of the metaphor comprehension post-test.

During data coding, multiple-choice options A, B, C, and E were easily coded. As

to option D, students filled it in 21 instances in total. Because the students’ an-
swers were open to interpretation, an inter-rater reliability procedure was used by
which two raters coded option D answers for correctness of meaning indepen-
dently, and the rating scores were analysed through Cohen’s Kappa. The test re-
vealed that agreement between the scores of the two raters in the pre-test was
96%, with a Kappa score of 0.89 at (p= .000), while in the post-test, this agree-
ment level was 97%, with a Kappa score of .95 at (p= .000). This result indicates
that the agreement between the two raters on the pre-test and the post-test is al-
most perfect (Landis and Koch 1977). The raters then discussed the different
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 277

answers until they agreed on the correctness of 11 cases in the pre-test and 6 cases
in the post-test.
A statistical analysis of the metaphor comprehension tests was performed
using SPSS Version 22. Since the number of participants in each of the three
participant groups was below 30, the Kruskal-Wallis Test, which is the non-
parametric equivalent of One-Way ANOVA, was used to compare their perform-
ances in the pre-test, the post-test and the 2-week delayed test. As only similar-
ity of distribution of data was expected in the Kruskal-Wallis Test, the test
results appeared to follow similar distribution shapes in histogram plots.

6.6 Metaphor Production Tests

The metaphor production pre-test and post-test consisted of free-writing assess-

ments that were designed to measure free production of the 11 taught metaphoric
expressions. Due to conflicts with university exams, the production post-test was
postponed to Week 4 8 days after the metaphor comprehension post-test. The
amount of time between the two posttests may not be enough to avoid test-
taking effects, but it was the best option available at the time. The writing in-
structions are presented in Figures 4 and 5:

Have you ever been in a situation where you had to make a decision between two alternative
choices? Write about this situation. Explain how you were undecided between the two choices
and what made you make this life-changing decision. What factors influenced you to reach this
decision? What would have happened if you had taken the alternative choice?

Figure 4: Instructions for the metaphor production pre-test.

Think of two alternative careers that you would like to pursue in the future. Write about these
two jobs. Which career would you enjoy more and why? Which career do you think you will
finally choose? Are there any external or internal factors that might influence your decision to
pursue one rather than the other? How will you reach your final decision?

Figure 5: Instructions for the metaphor production post-test.

The writing topics were chosen because they revolved around life experiences,
but there was no guarantee that the topics would promote productive use of the
metaphoric expressions. Participants were given 60 minutes to answer each
278 Rawan A. Saaty

test and were instructed to write a 200-word paragraph without thinking about
spelling, as this concern may hinder the writing process.
As this study aims to provide awareness-raising activities for the teaching of
metaphors rather than measuring language learners’ metaphor use, an overall
metaphor identification procedure (e.g. MIP or MIPVU) capable of scanning the
texts for each metaphoric instance was not needed. Instead, scanning the partic-
ipants’ writing samples was limited to the 11 taught metaphoric expressions. I
read each writing sample twice to familiarise myself with the styles of writing.
Following this process, I counted the number of words in each sample because
the students’ written samples varied in length considerably. Shortened uses (e.g.
I’ll) were counted as separate individual words (i.e. I will) to maintain conformity
in the number of words per student. In the case where a metaphoric expression
was used, I treated errors in spelling, verb tense, and singular versus plural
forms as irrelevant errors and therefore counted the uses as correctly used meta-
phoric expressions. However, if the metaphoric expression consisted of a multi-
word unit, I verified that the entire unit had been used correctly and considered
errors in the use of prepositions that constituted a part of the multiword-unit
and/or phrasal verb as incorrect uses (i.e. in the fast-lane). Moreover, if a partici-
pant used a taught metaphoric expression on two different occasions, I counted
this as two single uses of metaphoric expressions and two points were awarded
for each use. The use of taught metaphoric expressions in writing by the three
participant groups was then compared using the Kruskal-Wallis Test in SPSS.

7 Results
This study investigates how teaching metaphoric expressions through enact-
ment-based metaphor awareness of the metaphor LIFE IS A JOURNEY could in-
fluence Saudi learners’ comprehension, retention and production of taught
metaphoric expressions in comparison with conceptual metaphor awareness
and semantic clustering. Results are presented in terms of metaphor compre-
hension tests and metaphor production tests.

7.1 The Results of the Metaphor Comprehension Tests

To measure the differences in improvement between the control group, the con-
ceptual metaphor group, and the enactment metaphor group, the Kruskal-
Wallis Test was used twice. It was used first to study the differences between
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 279

the metaphor understanding in the post-test and the pre-test and then to ex-
plore the differences between the 2-week delayed test and the pre-test. Table 3
presents the average scores of the metaphor comprehension pretest, posttest,
and 2-week delayed test received by the control group, the conceptual meta-
phor group, and the enactment metaphor group. Table 4 presents a summary of
the means, medians, standard deviations, and the 2-tailed significance of both:

Table 3: Average results of the metaphor comprehension tests.

n. Pretest avg. Posttest avg. -week delayed

test avg.

Control group  . . .

Conceptual metaphor group  . . .
Enactment metaphor group  . . .

Table 4: The differences of improvement between the post-test and the

pre-test, and between the 2-week delayed test and the pre-test.

n Post-test/pre-test Delayed test/pre-test

difference difference

Control  M= . M= -.

group Md= . Md= -.
SD= . SD= .

Conceptual  M= . M= .

metaphor Md= . Md= .
group SD= . SD= .

Enactment  M= . M= .

metaphor Md= . Md= .
group SD= . SD= .

-tailed significance p= . p= .

The Kruskal-Wallis Test reveals that the difference in improvement between

the post-test and the pre-test between the medians of the control group (M= 1.65,
Md= 2.00, SD= 2.57), the conceptual metaphor group (M= 5.06, Md= 5.00,
SD= 2.94), and the enactment metaphor group (M= 4.84, Md= 5.00, SD= 2.42)
is highly significant (p= .001). A highly significant difference (p= .000) is also
found regarding the differences in improvement between the 2-week delayed test
and the pre-test between the medians of the control group (M= -0.53, Md= -1.00,
SD= 2.15), the conceptual metaphor group (M= 2.72, Md= 3.00, SD= 2.96), and the
280 Rawan A. Saaty

enactment metaphor group (M= 5.16, Md= 5.00, SD= 2.37). Post-hoc tests in the
form of a series of Mann-Whitney U Tests reveal the specific differences between
the control group, the conceptual metaphor group, and the enactment metaphor
group. Applying the Bonferroni adjustment to the alpha level, significance in the
post-hoc analyses is judged at (p<.017). Table 5 presents the significant compari-
sons between the individual groups and their effect sizes:

Table 5: Post-hoc test results for differences in improvement with the effect sizes.

Difference between the Difference between the

post-test and pre-test delayed test and pre-test
significance significance

Control group and p = . – r = . p = . – r = .

conceptual metaphor group

Control group and p = . – r = . p = . – r = .

enactment metaphor group

Conceptual metaphor group NA p = . – r = .

and enactment metaphor
group

The individual Mann-Whitney U-Tests in Table 5 indicate that the only signifi-
cant differences in improvement between the post-test and the pre-test are be-
tween the control group and the conceptual metaphor group (p= .001) with a
medium effect size (r= .4), and between the control group and the enactment
metaphor group (p= .000) with a medium effect size (r = .45) (Cohen 1988). This
finding suggests that by being made aware of the metaphoric motivations of
the metaphoric expressions, the experimental groups have understood the
taught metaphoric expressions more accurately than the control group.
In addition, the post-hoc tests reveal that there are significant differences in
improvement between the 2-week delayed test and the pre-test between the con-
trol group and the conceptual metaphor group (p= .001) with a medium effect
size (r= .41), the control group and the enactment metaphor group (p= .000) with
a large effect size (r= .64), and between the conceptual metaphor group and the
enactment metaphor group (p= .008) with a medium effect size (r= .34) (Cohen
1988). These findings suggest that learning metaphoric expressions through ac-
tions has aided the enactment metaphor group’s retention of metaphoric expres-
sions up to a period of two weeks in comparison to the control and conceptual
metaphor groups. In addition, the results show that longer-term retention seems
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 281

to have also happened in the conceptual metaphor group, but not as pronounced
as in the enactment metaphor group. This indicates that enactment-based meta-
phor awareness is more effective in terms of retention than the teacher-centred
conceptual metaphor awareness. In addition, employing conceptual metaphor
awareness appears to aid longer-term retention of the senses of the metaphoric
expressions, just not as much as learning the metaphoric expressions through
enactment and mime.

7.2 The Results of the Metaphor Production Tests

As none of the participant groups used the target metaphoric expressions in the
metaphor production pre-test, only the results of the metaphor production post-
test are statistically analysed using the Kruskal-Wallis test. The Kruskal-Wallis test
indicates that the difference in numbers of taught metaphoric expressions in the
control group (M= .31, Md= .00, SD= .1), the conceptual metaphor group (M= .64,
Md= .00, SD= .19) and the enactment metaphor group (M= .85, Md= .00, SD= .25)
is not significant. The 2-week interventional teaching sessions did not yield a sig-
nificant difference in the productive use of the taught metaphoric expressions be-
tween the three groups.
These results suggest that enactment-based metaphor awareness, concep-
tual metaphor awareness, and semantic clustering may be unsuitable teaching
methods for the production of metaphoric expressions. There is also the possi-
bility that the testing measure that consisted of a free-writing assessment
played a role in the less than encouraging results. Comparing pre- and post-
intervention production in the form of free-writing assessments which may con-
tain opportunities for metaphor production, may not be the most appropriate
method to test for metaphor production.

8 Discussion
The results of this study support the notion that adopting enactment for the teach-
ing of the conceptual metaphor LIFE IS A JOURNEY aids the understanding and
retention of metaphoric expressions than that of conceptual metaphor awareness
alone. This finding is important, as Boers (2004) has pointed out the shortcom-
ings of conceptual metaphor awareness as a technique limited to the understand-
ing and retention of taught metaphoric expressions. By associating metaphoric
expressions with their embodied nature through actions, longer-term retention,
282 Rawan A. Saaty

in this case in terms of two weeks, is achieved. However, a number of issues

should be acknowledged when discussing the results of this study.
First, in terms of the results of the metaphor comprehension tests, there ap-
pears to be a dual advantage that can be gained from enactment-based metaphor
awareness: one for the awareness of embodied meanings of the metaphoric ex-
pressions, and one for the enactment effect on memory. First, enactment-based
metaphor awareness familiarised the learners with the sensorimotor motivations
behind LIFE IS A JOURNEY and its metaphoric expressions, which was also ob-
served in Lindstromberg and Boers’ (2005) study on action verbs. The important
addition provided by the current study is that the learners acted out not only ac-
tion verbs but also complex constructions like nouns, collocations, lexical phrases,
etc. Second, while Lindstromberg and Boers (2005) did not explore beyond 1-week
retention of vocabulary, the current study has found that enactment-based aware-
ness of metaphor has had stronger mnemonic benefits than conceptual metaphor
awareness and semantic clustering two weeks after teaching. The results of this
study also align with those of Macedonia and Klimesch’s (2014) longitudinal
study, which show highly significant results in 2-week delayed tests for enacted
vocabulary compared to vocabulary items explained using audio-visual modali-
ties. It is important to note that the difference between the three groups is based
on a combination of variables including their respective teaching methodologies
(PPP for the control group and the conceptual metaphor group and TPR for the
enactment metaphor group), the amount of input for each group, the time spent
by each group on the items along with the different metaphor awareness-raising
activities. While it is not clear if the results of the enactment metaphor group are
due to one of these variables or all, it appears that employing enactment-based
awareness of metaphor through TPR aids the comprehension and retention of
taught metaphoric expressions for learners.
In terms of the results of the metaphor production tests, the 60 learners
provided writing samples ranging from 28 words to 199 words in the pre-test
(M= 98.13, SD= 40.83) and from 20 words to 184 words in the posttest (M= 73.1,
SD= 34.33) and appeared to be able to express themselves well without using
taught metaphoric expressions. The results were also rather surprising espe-
cially that the students were reminded yet again of the metaphoric expressions
in the metaphor comprehension test which was administered 8 days prior to
the metaphor production test. As discussed in Section 4, Nation ([2001] 2013)
details how productive vocabulary progresses at a slower rate than passive vo-
cabulary, which could explain the delayed use of the taught metaphoric expres-
sions. Further cognitive linguistic research should investigate the fostering of
productive knowledge of metaphoric expressions in EFL learners focusing on
the development of testing measures for metaphor production in L2.
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 283

In conclusion, since the current study was unable to test for the impact of
enactment-based metaphor awareness for more than two weeks after the inter-
ventional teaching sessions, its results can be interpreted as indicative rather
than conclusive. More longitudinal classroom-based experimental research is
crucial to gain an accurate picture of the impact of enactment-based metaphor
awareness-raising activities in L2.

References
Asher, James. 2012 [1977]. Learning another language through actions: The complete
teacher's guidebook, 7th edn. Los Gatos: Sky Oaks.
Barsalou, Lawrence. 2009. Simulation, situated conceptualization, and prediction.
Philosophical Transactions of the Royal Society 364. 1281–1289.
Boers, Frank. 2000. Metaphor awareness and vocabulary retention. Applied Linguistics 21.
553–571.
Boers, Frank. 2004. Expanding learners’ vocabulary through metaphor awareness: What
expansion, what learners, what vocabulary? In Susanne Niemeier & Michel Achard (eds.),
Cognitive linguistics, second language acquisition, and foreign language teaching,
211–234. Berlin: Mouton de Gruyter.
Boers, Frank. 2013. Cognitive linguistic approaches to teaching vocabulary: Assessment and
integration. Language Teaching 46(2). 208–224.
Boers, Frank & Seth Lindstromberg. 2008. How cognitive linguistics can foster effective
vocabulary teaching. In Frank Boers & Seth Lindstromberg (eds.), Cognitive linguistic
approaches to teaching vocabulary and phraseology, 1–65. Berlin: Mouton de Gruyter.
Boers, Frank, Seth Lindstromberg, Jeannette Littlemore, Hélène Stengers & June Eyckmans.
2008. Variables in the mnemonic effectiveness of pictorial elucidation. In Frank Boers &
Seth Lindstromberg (eds.), Cognitive linguistic approaches to teaching vocabulary and
phraseology, 189–116. Berlin: Mouton de Gruyter.
Boers, Frank, Murielle Demecheleer & June Eyckmans. 2004. Cross-cultural variation as a
variable in comprehending and remembering figurative idioms. European Journal of
English Studies 8(3). 375–388.
Boers, Frank, June Eyckmans & Hélène Stengers. 2007. Presenting figurative idioms with a touch
of etymology: More than mere mnemonics? Language Teaching Research 11(1). 43–62.
Boers, Frank, Ana Piquer Píriz, Hélène Stengers & June Eyckmans. 2009. Does pictorial
elucidation foster recollection of figurative idioms? Language Teaching Research, 13(4).
367–388.
Cohen, Jacob. 1988. Statistical power analysis for the behavioral sciences. Hillsdale: Lawrence
Erlbaum.
De Grauwe, Sophie, Roel Willems, Shirley-Ann Rueschemeyer, Kristin Lemhöfer & Herbert
Schriefers. 2014. Embodied language in first- and second-language speakers: Neural
correlates of processing motor verbs. Neuropsychologia 56. 334–349.
Desai, Rutvik H., Jeffrey R. Binder, Lisa L. Conant, Quintino R. Mano & Mark S. Seidenberg.
2011. The neural career of sensorimotor metaphors. Journal of Cognitive Neuroscience
23(9). 2376–2386.
284 Rawan A. Saaty

Engelkamp, Johannes. & Horst Krumnacker. 1980. Imaginale und motorische Prozesse beim
Behalten verbalen Materials. Zeitschrift für experimentelle und angewandte Psychologie,
27, 511–533.
Gibbs, Raymond. 2006. Embodiment and cognitive science. New York: Cambridge University
Press.
Gibbs, Raymond. 2013. Walking the walk while thinking about the talk: Embodied interpretation
of metaphorical narratives. Journal of Psycholinguistic Research. 42(4). 363–378.
Gibbs, Raymond. 2014. Embodied metaphor. In Jeannette Littlemore & John Taylor (eds.),
Bloomsbury companion to cognitive linguistics, 167–184. London: Bloomsbury.
Gibbs, Raymond & Marcus Perlman. 2006. The contested impact of cognitive linguistic
research on the psycholinguistics of metaphor understanding. In Gitte Kristiansen,
Michel Achard, René Dirven & Francisco Ruiz de Mendoza Ibáñez (eds.), Cognitive
linguistics: Current applications and future perspectives. 211–228. Berlin & New York:
Mouton de Gruyter.
Grady, Joseph. 1997. Foundations of meaning: Primary metaphors and primary scenes.
Berkeley: University of California Berkeley PhD Dissertation
IJzerman, Hans & Gün R. Semin. 2009. The thermometer of social relations: Mapping social
proximity on temperature. Psychological Science 20(10). 1214–1220.
Kelly, Spencer D., Tara Mcdevitt & Megan Esch. 2009. Brief training with co-speech gesture
lends a hand to word learning in a foreign language. Language Cognitive Processes 24.
313–334.
Landis, J. Richard & Gary Koch. 1977. The measurement of observer agreement for categorical
data. Biometrics 33. 159–174.
Laufer, Batia. 1998. The development of passive and active vocabulary in a second language:
Same or different? Applied Linguistics 19. 255–271.
Lazar, Gillian. 2003. Meanings and metaphors: Activities to practice figurative language.
Cambridge: Cambridge University Press.
Lee, Spike W. S. & Norbert Schwarz. 2012. Bidirectionality, mediation, and moderation of
metaphorical effects: The embodiment of social suspicion and fishy smells. Journal of
Personality and Social Psychology. 103(5). 737–749.
Lindstromberg, Seth & Frank Boers. 2005. From movement to metaphor with manner-of-
movement verbs. Applied Linguistics 26(2). 241–261.
MacArthur, Fiona. 2010. Metaphorical competence in EFL: Where are we and where should we
be going? A view from the language classroom. AILA Review 23. 155–173.
Macedonia, Manuela & Wolfgang Klimesch. 2014. Long-term effects of gestures on memory for
foreign language words trained in the classroom. Mind, Brain and Education 8(2),74–88.
Macedonia, Manuela, Karsten Müller & Angela D. Friederici. 2011. The impact of iconic
gestures on foreign language word learning and its neural substrate. Human Brain
Mapping. 32. 982–988.
Nation, Paul. 2013 [2001]. Learning vocabulary in another language, 2nd edn. Cambridge:
Cambridge University Press.
Ray, Blaine & Contee Seely. 2004 [1998]. Fluency through TPR storytelling: Achieving real
language acquisition in school, 4th edn. Berkley: Command Performance Language
Institute.
Skoufaki, Sophia. 2008. Conceptual metaphoric meaning clues in two idiom presentation
methods. In Seth Lindstromberg & Frank Boers (eds.), Cognitive linguistic approaches to
teaching vocabulary and phraseology, 101–132. Berlin: Mouton de Gruyter.
 An Enactment-Based Approach to the Teaching of Metaphoric Expressions 285

Tinkham, Thomas. 1993. The effect of semantic clustering on the learning of second language
vocabulary. System 21(3). 371–380.
Walker, Crayton. 2008. Factors which influence the process of collocation. In Frank Boers &
Seth Lindstromberg (eds.), Cognitive linguistic approaches to teaching vocabulary and
phraseology. Berlin & New York: Mouton de Gruyter.
Wilcox, Amanda & Almitra Medina. 2013. Effects of semantic and phonological clustering on
L2 vocabulary acquisition among novice learners. System 41. 1056–1069.
Wilson, Nicole L. & Raymond Gibbs. 2007. Real and imagined body movement primes
metaphor comprehension. Cognitive Science 31. 721–731.
Xue, Jin, Jie Yang & Qian Zhao. 2014. Chinese–English bilinguals processing temporal–spatial
metaphor. Cognitive Processing 15(3). 269–281.
Subject Index
Applied Cognitive Linguistics (ACL) 3, 5, Vrije Universiteit Amsterdam Metaphor
6, 71 Corpus (VUAMC) 243, 248
academic discourse 155, 244, 245 creative figurative language 65
creative metaphors 90
blending theory 39, 52 cross-sensory metaphor 107
bodily-basis 103, 104, 110, 119, 123, 125, culture 27, 34, 39, 103, 104, 114, 116, 125,
126, 128 221, 223

classroom discourse 5 deliberate metaphor 39, 135, 195

closed-class metaphors 173, 184 direct metaphors 179, 189, 258
Cognitive Linguistic (CL)-oriented dual coding theory 222
approach 57, 59, 72
collocations 128, 193, 194 ecology of metaphor 18, 34
colour 50, 51, 66–69, 103, 107–112, 116, 119, embodiment, embodied 18, 19, 21, 27, 59,
123, 126, 129, 166 68, 71, 103, 109, 120, 123, 125, 126, 128,
Common European Framework of Reference 263, 266, 268, 282
for Languages (CEFRL) 4, 86, 92, 175, emotions 24, 45, 103, 111, 116
202, 269 enactment effect hypothesis 267
complexity theory 39, 52 English as a Lingua Franca (ELF) 28, 150, 151,
conceptual metaphors 155, 169
ARGUMENT IS WAR 135, 136 English as a Foreign language (EFL) 37, 71,
LIFE IS A JOURNEY 270, 278 72, 81–84
UNDERSTANDING IS SEEING METAPHOR 18–21, etymological elaboration 222, 224, 225, 226,
151 227, 236, 264
KNOWING IS SEEING 19, 166
conceptual metaphor theory (CMT) 37, 70, figurative awareness 71, 72, 204
154, 174, 186 figurative competence 57–59, 65–67, 204
conceptual fluency 59, 201, 203 figurative language 57, 60, 62, 64, 66, 199,
constructional grounding theory 64 200
Content and Language Integrated Learning figurative language production 60, 132
(CLIL) 8, 240–243, 249, 253, 259 figurative language understanding 60, 62,
corpora 200, 205, 208
Tracking Written Learner Language figurative reasoning 59, 60, 65, 68, 69
(TraWL) 173, 176, 177, 179, 181, figurative thinking 57, 60, 139, 140, 201, 203
182, 183, 184, 186, 188, 192, 194, 196 foreign language instruction/teaching 3–6,
European Corpus of Academic Talk 9, 37, 48, 71, 80, 85, 201, 204, 214, 242
(EuroCoAT) 19, 155, 258
British National Corpus (BNC) 151, 154, grammatical metaphor 47, 241
174, 243, 271, 276 genre 245, 246
Cambrige Learner Corpus 175, 202
Norwegian International Corpus of homonymy 222
Learner English (NICLE) 201
Louvain Corpus of Native English Essays
idioms 50, 51, 119, 129, 221–228, 232–236
(LOCNESS) 201

https://doi.org/10.1515/9783110630367-014
288 Subject Index

irony 58, 60 metaphor translation 131, 135

implicit metaphors 179 metaphor translation strategies 132, 137
indirect metaphors 179, 259 creative translations 145
metaphorical (metaphoric) competence
keyness analysis 154, 156, 157 5–9, 23, 48, 53, 58–60, 62, 80–81,
85, 86, 89, 90, 174, 175, 199–203,
languages 205
Cantonese 104, 105, 107–111, 113–116, metaphorical discourse 18
120, 125, 129 metaphorical reasoning 23, 80, 131
Chinese 80, 103, 143, 223 metaphorical thinking 22, 23, 72
English 19, 24–29, 37, 67, 86, 103–105, metonymy 41, 51, 58, 60, 63, 71, 144, 145,
107, 112, 113, 116, 119, 120, 125, 194
127 motivation 57, 58, 69, 70, 71, 104, 111, 112,
Italian 133, 139 204, 225, 232, 233, 246, 263, 266, 268,
Mandarin 228 282
Spanish 24, 25, 27, 79, 80, 204, 213
learner errors 180, 181 native language 127
levels of processing theory 222 native speaker 28, 84, 151
lexical competence 202, 203, 205, 214 non-native speakers 29, 30, 84, 151
linguistic metaphors 174 novel metaphor 28, 30, 83, 84, 140

metaphor awareness 205, 263–269, 274, open-class metaphors 173, 184, 242,
281–283 249, 250
metaphor bursts 45, 134, 245
metaphor clusters 181, 182, 186, 190, 191, Part of Speech (PoS) 153, 243, 244, 250
196 personification 46, 57, 58, 194, 255
metaphor density 8, 84, 85, 173, 181, 183–186, polysemy 58, 59, 70, 71, 202, 223
189, 244, 245, 249, 251, 253, 257 primary metaphors 5, 18, 19, 21, 22, 39, 41,
moving metaphor density 181 64, 150, 265, 266
metaphor elicitation 44 Productive Vocabulary Level Tests
metaphor flags 40, 44, 45, 190, 248 (PVLT) 200, 206
metaphor functions 37, 42–47, 187, 188, proverbs 41, 42
191, 196, 246, 258
metaphor identification 33, 154, 167, 243, scenario 42, 43
247, 248, 254 second language (L2) 3, 8, 42, 48, 53,
MIP 20, 30, 32, 135, 153, 154, 157, 57, 59, 66, 72, 80, 82, 84, 104, 119,
208, 278 127, 128, 132, 174, 175, 199–202, 266
MIPVU 30, 32, 135, 153, 154, 173, second language (L2) competence 132, 145
178, 180, 208, 247, 248 semantic networks 71
metaphor in educational contexts 149, 242, sight metaphors 20, 21, 150, 155,
246, 253, 254 158–166
metaphor processing 139 simile 40, 190, 248, 258
metaphor production 28, 58, 81, 83–85, 133, Spanish as a Foreign Language (SFL) 81, 84,
173–175, 177, 178, 185, 190, 192, 199, 85, 87–89, 90
201, 267, 268, 277, 281 subregisters 245, 256
metaphor understanding 22, 82, 85, 133,
138, 199, 202, 213, 266, 268, 279, 281
 Subject Index 289

vocabulary breadth 202 word-colour associations 104, 110, 112, 114,

vocabulary depth 202 119, 123, 127, 128
vocabulary learning strategies 205, 227
Vocabulary Levels Tests (VLT) 200, 206 young L2 learners 22, 57, 60, 66–67, 90, 91,
204, 226, 246