![Acknowledgements
The Seventh International Conference on Researching and Applying Metaphor
(RaAM7) was the first of the RaAM conferences to be organized under the auspices of
the recently created Association for Researching and Applying Metaphor (http://www.
raam.org.uk). It was held in May 2008 at the Faculty of Arts of the University of
Extremadura (Cáceres, Spain) during the European Year of Intercultural Dialogue. In
line with the European drive to foster increased awareness of cultural diversity, the
hosts of this international conference – the editors of this volume – chose as its over-
arching theme ‘Metaphor in Cross-Cultural Communication’. The Year of Intercultural
Dialogue, like the conference itself, aimed to encourage all those living in Europe and
elsewhere to explore the benefits of our rich cultural heritages and to take advantage of
opportunities to learn from different cultural traditions.
Like previous RaAM meetings, RaAM7 gathered metaphor researchers from
many disciplines from all over the world, providing a forum for high-quality research
into metaphor in ‘real world’ contexts. Many of the chapters included in this volume
were originally presented as papers at this conference and were subsequently enriched
by the supportive and sometimes lively debate and discussion that characterizes RaAM
meetings. We gratefully acknowledge the expert advice and support given to the local
organizers by the RaAM Association, and most particularly that of Lynne Cameron,
Graham Low, and Jeannette Littlemore, respectively Chair, Secretary, and Treasurer of
the Association at that time.
We are also grateful for the support given to us by the University of Extremadura
– and especially that of the Dean of the Faculty of Arts, Luis Merino Jerez – and for the
funding given to us by the Spanish Ministerio de Ciencia e Innovación (Dirección
General de Programas y Transferencia de Conocimientos-Acciones Complementarias
[HUM2007–30872-E] and by the Junta de Extremadura (CON08020). Their help
contributed to making this conference possible and also enabled us to offer a number
of bursaries so that a number of young metaphor researchers from different parts of
the world could attend this conference.
We extend our thanks to all those at John Benjamins who have contributed to
making this volume possible, especially to Hanneke Bruintjes for her help in the early
stages and Els van Dongen later on. The Series Editors have provided crucial support
and advice at different stages of preparing the manuscript, and the anonymous review-
ers who carefully read the entire manuscript made a number of helpful suggestions for
its improvement.](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-14-320.jpg)
![ Fiona MacArthur and José Luis Oncins-Martínez
Even though the primary site for human communication is conversation, speakers
of English do not appear to use linguistic metaphors as frequently when they are chat-
ting to each other face to face as they do in the written medium (Steen et al. 2010), so
another factor that contributes to metaphor variation is the discourse contexts in
which it is used. Furthermore, certain written registers display a much greater density
of metaphor use than others. Steen et al. (2010) have found that metaphor is used
much more frequently in academic discourse than in fiction, a perhaps somewhat sur-
prising finding given the traditional emphasis on metaphor as a trope peculiar to
poetry and fictional prose. But even within academic discourse, for example, meta-
phor use varies: different academic discourse communities use metaphor in different
ways. The metaphors used by economists, for example, when writing and talking about
their discipline are not the same as those used by architects when dealing with theirs
(Alejo 2010, Caballero 2006), for the metaphor systems or models that constitute par-
ticular theories or frame the problems that disciplines seek to explore and resolve
(Kuhn 1993) vary across different areas of enquiry. Indeed, major paradigm shifts may
be marked by changes in the metaphors conventionally used in a field of scientific
enquiry (see, for example, Aitchison’s [2003] discussion on competing metaphors for
understanding linguistic change), which recalls the importance of the diachronic di-
mension as one more factor that contributes to metaphor change and variation.
When studied in a historical time scale, metaphor has been revealed to play an
important role in motivating semantic change in English (e.g. Allan 2008, Kay 2000,
Sweetser 1990), and research adopting a diachronic perspective on metaphor use has
not only provided details about the processes involved in how word meanings change
in the course of time, but has also shed light on the status of particular utterances as
“metaphors” for speakers of earlier and later generations (Alm-Arvius, this volume;
Geeraerts and Grondelaers 1995, Oncins-Martínez 2006), for consideration of meta-
phor in various time scales reveals that what might count as a metaphor at one time
and in one context might be regarded somewhat differently in another. For example,
one of the time scales in which metaphor has been widely researched – the ontoge-
netic – has further revealed the complexity of this phenomenon and how difficult it
may be to decide on whether the unconventional ‘metaphor-like’ utterances of
children should be considered metaphors at all (Cameron 1996). Piaget (1962) re-
ported his daughter between the ages of 3 : 6 and 4 : 7 saying that a winding river was
like a snake and comparing a bent twig with a machine for putting in petrol. While
Piaget himself regarded these as ‘child metaphors’ as opposed to ‘real metaphors’
(describing them as nothing more than products of the symbolic, imagistic type of
thinking that characterizes the pre-operational stage), other researchers have used
different criteria to distinguish metaphors and pseudo-metaphors in children’s speech
(e.g. Billow 1981, Nerlich et al. 1999, Vosniadou and Ortony 1983, or Winner 1988),
reaching different conclusions about what distinguishes a child’s use of metaphor
from an adult’s, and how the changes in children’s use and understanding of meta-
phor at different ages can be accounted for.](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-17-320.jpg)
![Introduction
The complexity of the task of researching metaphor is perhaps most apparent
when we move away from a consideration of metaphor solely in relation to English
speakers or even speakers of other standard European languages. As Leezenberg
(2001: 15) has pointed out, there are certain “cultural prerequisites for a notion of
metaphor”. A similar point is made by Goddard when he notes that the term ‘meta-
phor’ lacks precise equivalents in many of the world’s languages, and warns of the
dangers of uncritically adopting the category as a starting point for cross-cultural com-
parison (2004: 1212). Both authors discuss the issue in relation to A is B (active or
expository) metaphors, and Leezenberg (2001: 15) cites the disagreement over inter-
pretations of the much debated utterance of the Bororo Indians of Brazil pa e-do nabure
(‘we are parrots’). Early accounts (e.g. Durkheim and Mauss 1963: 6–7) suggested that
the Bororo did not distinguish between the categories of people and animals, and this
expression could not therefore be classed as a metaphor. However, close attention to
the linguistic form of the utterance (Turner 1991: 135–136) has provided grounds for
thinking that it should not be regarded as a ‘literal’ statement or a conflation of the
categories people and birds/animals, because it can only be used to refer to men and
the verb is marked for ‘customary form’ rather than ‘permanent state’ (Leezenberg
2001: 16). In the light of close linguistic analysis, then, the utterance can be regarded
as instantiating the metaphorical mapping people are animals. In fact, as numerous
studies over the years have shown, there appears to exist a very widespread tendency
to ‘see’ people as animals, although the instantiation of the mapping varies consider-
ably across different language-speaking communities. The use of the same animal
names to refer to people may be similar or quite different in different languages (e.g.
Hines 1999, Hsieh 2006, López Rodríguez 2009, Talebinejad and Dastjerdi 2005), as
are the preferred ways of instantiating the metaphor in everyday speech (Deignan
1999). Similarly, while it seems true that “the existence of the semantic prime body
invites people to theorise about the other parts of a person” (Goddard 2003: 122), the
way that speakers of different languages establish these relations varies considerably.
The head, the heart, the liver, the ear, and the stomach are some of the body parts and
organs associated with ‘thinking’ or ‘feeling’ in different languages (Goddard 2003,
Wierzbicka 1992, Yu 2007, 2009) but although body part for thought/feeling
might be a common pattern, the type and value of the thoughts or emotions associated
with each body part is often different across languages. Goddard (2003: 124) describes,
for example, the hati (liver) concept in Malay as:
very ‘feeling-oriented’ but focused primarily on interpersonal feelings. [...] the
hati is viewed as an inner domain of experience, but there is a heightened empha-
sis on its motivational consequences, along with a certain moral ambivalence. On
account of the hati, a person may have an urge to do bad things as much as good
things (hence one ought not unthinkingly or impulsively follow one’s hati; as the
saying goes, ikuthati mati ‘follow the hati, die’).](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-18-320.jpg)
![Introduction
first language subtly alter the way that these mappings are conceived. These findings
have implications not only for foreign language teaching, but also for cross-cultural
metaphor research, because they show the importance of language in shaping culture-
specific conceptualizations.
Like Johansson Falck’s, the study reported by Anne Golden in Chapter 6 –
“Metaphorical expressions in L2 production: The importance of text topic in corpus
research” – focuses on one specific area of language use: in this case, the high fre-
quency Norwegian verb ta (roughly equivalent to English ‘take’) as used by NNSs with
three different mother tongues (L1s): German, Spanish, and Russian. Golden com-
pares these learners’ uses of this verb with that of Norwegian students’ in order to ex-
plore the differences between the way these groups of speakers employ the verb in its
basic or metaphorical sense, but distinguishing also between the use of ta in fixed col-
locations or as ‘bridge terms’ (Kittay 1990). Among the findings that emerge from this
study is that, although differences in metaphorical uses of ta can be observed among
the three different NNS groups, related both to their L1 and to their command of the
second language (L2), the topic of the written discourse proves the most important
variable: in the control corpus employed, the NSs of Norwegian were found to use ta
with metaphorical senses less frequently than the NNSs. The conclusions drawn echo
Cameron’s observation (2008: 203) that the absence of metaphor is as significant as its
presence in discourse, and the density of metaphor use is often related to what is being
talked about. In this regard, Golden’s chapter sheds light on some of the problems that
are involved in attempting to relate competence in a foreign language with metaphoric
competence (Danesi 1993, Littlemore and Low 2006). Context or the topic a NNS
needs to talk about also influences L2 learners’ use of metaphor.
Unlike the preceding two chapters, in “Researching linguistic metaphor in native,
non-native, and expert writing”, Claudia Chapetón-Castro and Isabel Verdaguer-
Clavera do not start from consideration of the metaphorical use of any particular
lexical items when comparing NSs and NNSs writing, or how the first language may
influence metaphor production in the second language, but rather seek to discover
more general patterns in the different corpora they examine. In order to do so, their
study involved identifying all potentially metaphorical uses of language. In their chap-
ter, they describe in detail how the combination of two different methods of identify-
ing metaphors in discourse (through the identification of vehicle terms, as developed
by Cameron [2003] and MIP [Pragglejaz 2007]) enabled them not only to reliably
identify the metaphors in the texts they examined, but also to point out how this pains-
taking approach to metaphor identification obliges the researcher to engage closely
with the linguistic form of the metaphors used. Using this combined procedure,
Chapetón-Castro and Verdaguer-Clavera carry out a three-way comparison between
the use of metaphor by undergraduate Spanish learners of English with that of NS
undergraduate students and that of NS expert writers. The findings provide interesting
detail about the similarities and differences between the three groups of writers, not
only as far as the linguistic forms of the metaphors used are concerned, but also as](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-24-320.jpg)
![Chapter 1.╇ An assessment of metaphor retrieval methods
techniques with respect to how accurate they are in terms of offering good samples
(i.e. with minimal error) for researchers.
Unlike sampling techniques, with census techniques a question that arises is how
reliable the identification is, that is, whether the steps were correctly followed by all
coders involved. In census projects, usually more than one person is responsible for
doing the coding, and in order for the coding to be reliable, there must be a high
degree of agreement among coders (see Dorst and Kaal, this volume; Chapetón -Castro
and Verdaguer-Clavera, this volume).
As can be seen, the methodological issues surrounding sampling and census tech-
niques are markedly different. Because census techniques involve questions related to
coder agreement and not sample adequacy, they are not examined here.
Pre-defining a search string or pool of string is perhaps the most frequently used
sampling technique in corpus-based metaphor research. It consists in determining
ahead of time one or more search strings based on particular research goals. A number
of different researchers have applied this technique. For instance, Deignan (2005)
chose words such as ‘hunt’ (8) and ‘warm’ (68), and expressions like ‘hot under the
collar’ (21) and ‘in the running’ (28), and then searched for them in a large corpus
(Bank of English). Her choice of each of these terms was motivated by theoretical con-
siderations, including the link between linguistic and conceptual metaphor, the rela-
tionship between metaphor and metonymy, patterning of linguistic metaphor, and
frequency of metaphor compared to literal senses.
Stefanowitsch (2006) also defined a set of words to focus on, independent of the
corpus to be analysed, with the primary purpose of detailing the procedure known as
Metaphor Pattern Analysis (MPA), which is aimed at finding metaphorical expres-
sions in corpora. A metaphorical pattern was defined as “a multi-word expression
from a given source domain (SD) into which one of more specific lexical items from a
given target domain (TD) have been inserted” (66), and lexical items included content
words (nouns, verbs, adjectives, adverbs). In Stefanowitsch (2006), MPA was used to
identify metaphorical conceptualizations of emotions, such as anger, happiness and
sadness. It involved the following steps. First, a target domain, for example, anger, was
selected based on the previous literature on metaphors and emotions. Second, a lexical
item was chosen to represent that domain, for instance ‘angry’. Third, a corpus
(the British National Corpus [BNC]) was searched for that lexical item and a sample of
up to 1,000 concordance lines was retrieved. Fourth, these lines were analysed by hand
to determine whether each occurrence was a metaphor or not, and if so, what concep-
tual metaphor motivated it; a metaphor was counted when the chosen lexical item and
other lexical items nearby expressed a source-target domain mapping, for example,
‘angry’ and ‘boiling’ were considered to express the conceptual metaphor anger is
hot fluid in a container. MPA and introspection were compared as to their ability
to find metaphorical mappings. Results showed that MPA found well over 90% of the
mappings identified by introspection, and that it also spotted mappings that were not
arrived at introspectively.](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-38-320.jpg)
![ Tony Berber Sardinha
In terms of the techniques involved, MPA is a concordance-based procedure that
relies on the choice of appropriate candidates: (“choose the lexical items wisely”
[Stefanowitsch 2006: 66]), which in turn depends on the selection of particular do-
mains that are of interest to a researcher. Hence, it is not suited for the analysis of a
whole corpus, because it would mean having to analyse each word in the corpus, and
therefore was not included for testing here. It must be said, though, that the idea of
metaphorical patterns can be used to automate metaphor retrieval in corpora, accord-
ing to Stefanowitsch (2006: 102–103):
[W]e might even envision a lexical database containing a large number of lexical
items and the metaphorical patterns they occur with (analogous to the FrameNet
project at the UC Berkeley), which would allow easy retrieval of all metaphors as-
sociated with a particular lexical item (or semantic field) and vice versa.
In fact, such databases were used by the Metaphor Candidate Identifier (MCI), the
metaphor detecting tool reviewed in Section 7.
In both of these techniques, the set of strings to be searched for was defined in
top-down mode, that is, the selection arose from theoretical or methodological con-
cerns. Some researchers, though, have used bottom-up approaches as means for deter-
mining which words to investigate.
One such procedure is that developed by Cameron and Deignan (2003). These
researchers point out problems with both small and large corpora. With small corpora,
the main issue is a lack of generalizability: “the frequency and metaphorical use of a
particular word form is inevitably influenced by the collection of data from a limited
number of discourse events” (Cameron and Deignan 2003: 151). And with large
corpora, there are two main problems, the first being a lack of information about the
context, which may make at least some of the data difficult to interpret, and the sec-
ond, the very issue being discussed in this chapter:
[A] problem in searching large corpora is that patterns may be missed, because
the researcher usually begins by searching for particular linguistic forms. If he or
she has not identified a particular form as worthy of study, it may not emerge from
the data during the analysis, and an important metaphorical use may be missed.
This reflects a fundamental difficulty in researching linguistic metaphors through
a corpus: We are trying to trace patterns of meaning but can only begin our analy-
sis by looking at forms. (Cameron and Deignan 2003: 151)
Their goal was to find “tuning devices” (‘just’, ‘sort of’, ‘actually’, etc.), or words and
expressions normally referred to as hedges or vague language, and to look at how they
were used in conjunction with metaphor. In order to find a good sample of tuning
devices to search for in the large corpus, without predicting them by either introspec-
tion or examples from the previous literature, they decided to read a small corpus
(28,285 words) and note down any relevant search terms. These were then searched for
in the large corpus (9 million words of the Bank of English), which contributed further](https://image.slidesharecdn.com/2040616-250528074538-53ad3bbd/85/Metaphor-In-Use-Context-Culture-And-Communication-Fiona-Macarthur-39-320.jpg)
Metaphor In Use Context Culture And Communication Fiona Macarthur
- 1. Metaphor In Use Context Culture And Communication Fiona Macarthur download https://ebookbell.com/product/metaphor-in-use-context-culture- and-communication-fiona-macarthur-4081232 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Metaphor In Use Context Culture And Communication Fiona Macarthur Et Al https://ebookbell.com/product/metaphor-in-use-context-culture-and- communication-fiona-macarthur-et-al-4688976 Confronting Metaphor In Use An Applied Linguistic Approach 1st Mara Sophia Zanotto Ed https://ebookbell.com/product/confronting-metaphor-in-use-an-applied- linguistic-approach-1st-mara-sophia-zanotto-ed-4442756 Metaphor Its Therapeutic Use And Construction A Professional Guide To Using Metaphor In Psychotherapy And Counseling Annotated Martin Cohen https://ebookbell.com/product/metaphor-its-therapeutic-use-and- construction-a-professional-guide-to-using-metaphor-in-psychotherapy- and-counseling-annotated-martin-cohen-57976596 Representational Change And The Use Of Metaphors In Problem Solving Benjamin Angerer https://ebookbell.com/product/representational-change-and-the-use-of- metaphors-in-problem-solving-benjamin-angerer-57319544
- 3. Metaphor In Specialist Discourse J Berenike Herrmann Ed Tony Berber Sardinha Ed https://ebookbell.com/product/metaphor-in-specialist-discourse-j- berenike-herrmann-ed-tony-berber-sardinha-ed-44870886 Metaphor In European Philosophy After Nietzsche 1st Edition Andrew Hines https://ebookbell.com/product/metaphor-in-european-philosophy-after- nietzsche-1st-edition-andrew-hines-47401896 Metaphor In Homer Time Speech And Thought Andreas T Zanker https://ebookbell.com/product/metaphor-in-homer-time-speech-and- thought-andreas-t-zanker-47475252 Metaphor In Foreign Language Instruction Ana Mara Piquerpriz https://ebookbell.com/product/metaphor-in-foreign-language- instruction-ana-mara-piquerpriz-48723924 Metaphor In Language And Culture Across World Englishes Marcus Callies Marta Degani Editors https://ebookbell.com/product/metaphor-in-language-and-culture-across- world-englishes-marcus-callies-marta-degani-editors-50219296
- 7. Volume 38 Metaphor in Use. Context, culture, and communication Edited by Fiona MacArthur, José Luis Oncins-Martínez, Manuel Sánchez-García, and Ana María Piquer-Píriz Human Cognitive Processing (HCP) Cognitive Foundations of Language Structure and Use This book series is a forum for interdisciplinary research on the grammatical structure, semantic organization, and communicative function of language(s), and their anchoring in human cognitive faculties. For an overview of all books published in this series, please see http://benjamins.com/catalog/hcp Editorial Board Bogusław Bierwiaczonek University of Economics and Humanities, Poland Mario Brdar Josip Juraj Strossmayer University, Croatia Barbara Dancygier University of British Columbia N.J. Enfield Max Planck Institute for Psycholinguistics, Nijmegen & Radboud University Nijmegen Elisabeth Engberg-Pedersen University of Copenhagen Ad Foolen Radboud University Nijmegen Raymond W. Gibbs, Jr. University of California at Santa Cruz Rachel Giora Tel Aviv University Elżbieta Górska University of Warsaw Martin Hilpert Freiburg Institute for Advanced Studies Zoltán Kövecses Eötvös Loránd University, Hungary Teenie Matlock University of California at Merced Carita Paradis Lund University Günter Radden University of Hamburg Francisco José Ruiz de Mendoza Ibáñez University of La Rioja Doris Schönefeld University of Leipzig Debra Ziegeler University of Paris III Editors Klaus-Uwe Panther Nanjing Normal University & University of Hamburg Linda L. Thornburg Nanjing Normal University
- 8. Metaphor in Use Context, culture, and communication Edited by Fiona MacArthur José Luis Oncins-Martínez Manuel Sánchez-García Ana María Piquer-Píriz University of Extremadura John Benjamins Publishing Company Amsterdamâ•›/â•›Philadelphia
- 9. Library of Congress Cataloging-in-Publication Data Metaphor in use : context, culture, and communication / edited by Fiona MacArthur, José Luis Oncins-Martínez, Manuel Sánchez-García, and Ana María Piquer-Píriz. p. cm. (Human Cognitive Processing, issn 1387-6724 ; v. 38) Includes bibliographical references and index. 1. Metaphor. 2. Communication. I. MacArthur, Fiona. P301.5.M48M469â•…â•… 2012 808.032--dc23 2012021736 isbn 978 90 272 2392 0 (Hb ; alk. paper) isbn 978 90 272 7346 8 (Eb) © 2012 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa 8 TM The paper used in this publication meets the minimum requirements of the╯American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.
- 10. Table of contents List of contributors vii Acknowledgements ix Introduction: Metaphor in use 1 Fiona MacArthur and José Luis Oncins-Martínez part 1.╇ Contexts of research 1. An assessment of metaphor retrieval methods 21 Tony Berber Sardinha 2. Metaphor in discourse: Beyond the boundaries of MIP 51 Aletta G. Dorst and Anna Kaal 3. Metaphor identification in Dutch discourse 69 Trijntje Pasma 4. Locating metaphor candidates in specialized corpora using raw frequency and keyword lists 85 Gill Philip part 2.╇ Contexts of production 5. Metaphor variation across L1 and L2 speakers of English: Do differences at the level of linguistic metaphor matter? 109 Marlene Johansson Falck 6. Metaphorical expressions in L2 production: The importance of the text topic in corpus research 135 Anne Golden 7. Researching linguistic metaphor in native, non-native, and expert writing 149 Claudia Marcela Chapetón-Castro and Isabel Verdaguer-Clavera
- 11. Metaphor in Use part 3.╇ Contexts of interpretation 8. Appreciation and interpretation of visual metaphors in advertising across three European countries 177 Margot van Mulken and Rob Le Pair 9. English native speakers’ interpretations of culture-bound Japanese figurative expressions 195 Masumi Azuma 10. The limits of comprehension in cross-cultural metaphor: Networking in drugs terminology 217 Richard Trim part 4.╇ Metaphor, topic, and discourse 11. Conceptual types of terminological metaphors in marine biology: An English-Spanish contrastive analysis from an experientialist perspective 239 José Manuel Ureña 12. Gestures, language, and what they reveal about thought: A music teacher’s use of metaphor in Taiwan 261 Ya-Chin Chuang part 5.╇ Metaphor and culture 13. Armed with patience, suffering an emotion: The conceptualization of life, morality, and emotion in Turkish 285 Yeşim Aksan and Mustafa Aksan 14. Trolls 309 Christina Alm-Arvius 15. A computational exploration of creative similes 329 Tony Veale part 6.╇ Afterword and prospects for future research 16. Metaphors, snowflakes, and termite nests: How nature creates such beautiful things 347 Raymond W. Gibbs, Jr. Name index 373 Terms index 375
- 12. List of contributors Mustafa Aksan Mersin University Turkey Yeşim Aksan Mersin University Turkey Christina Alm-Arvius Stockholm University Sweden Masumi Azuma Kobe Geijutsukoka/Design University Japan Tony Berber Sardinha São Paulo Catholic University Brazil Claudia Marcela Chapetón-Castro Universidad Pedagógica Nacional Colombia Ya-Chin Chuang University of York United Kingdom National Cheng Kung University Taiwan Aletta G. Dorst VU University Amsterdam The Netherlands Raymond W. Gibbs, Jr. University of California Santa Cruz United States of America Anne Golden University of Oslo Norway Marlene Johansson Falck Umeå University Sweden Anna Kaal VU University Amsterdam The Netherlands Rob Le Pair Radboud University Nijmegen The Netherlands Trijntje Pasma VU University, Amsterdam The Netherlands Gill Philip University of Macerata Italy Richard Trim Université de Provence France José Manuel Ureña University of Castilla la Mancha Spain Margot van Mulken Radboud University Nijmegen The Netherlands Tony Veale University College Dublin Ireland Isabel Verdaguer-Clavera University of Barcelona Spain
- 14. Acknowledgements The Seventh International Conference on Researching and Applying Metaphor (RaAM7) was the first of the RaAM conferences to be organized under the auspices of the recently created Association for Researching and Applying Metaphor (http://www. raam.org.uk). It was held in May 2008 at the Faculty of Arts of the University of Extremadura (Cáceres, Spain) during the European Year of Intercultural Dialogue. In line with the European drive to foster increased awareness of cultural diversity, the hosts of this international conference – the editors of this volume – chose as its over- arching theme ‘Metaphor in Cross-Cultural Communication’. The Year of Intercultural Dialogue, like the conference itself, aimed to encourage all those living in Europe and elsewhere to explore the benefits of our rich cultural heritages and to take advantage of opportunities to learn from different cultural traditions. Like previous RaAM meetings, RaAM7 gathered metaphor researchers from many disciplines from all over the world, providing a forum for high-quality research into metaphor in ‘real world’ contexts. Many of the chapters included in this volume were originally presented as papers at this conference and were subsequently enriched by the supportive and sometimes lively debate and discussion that characterizes RaAM meetings. We gratefully acknowledge the expert advice and support given to the local organizers by the RaAM Association, and most particularly that of Lynne Cameron, Graham Low, and Jeannette Littlemore, respectively Chair, Secretary, and Treasurer of the Association at that time. We are also grateful for the support given to us by the University of Extremadura – and especially that of the Dean of the Faculty of Arts, Luis Merino Jerez – and for the funding given to us by the Spanish Ministerio de Ciencia e Innovación (Dirección General de Programas y Transferencia de Conocimientos-Acciones Complementarias [HUM2007–30872-E] and by the Junta de Extremadura (CON08020). Their help contributed to making this conference possible and also enabled us to offer a number of bursaries so that a number of young metaphor researchers from different parts of the world could attend this conference. We extend our thanks to all those at John Benjamins who have contributed to making this volume possible, especially to Hanneke Bruintjes for her help in the early stages and Els van Dongen later on. The Series Editors have provided crucial support and advice at different stages of preparing the manuscript, and the anonymous review- ers who carefully read the entire manuscript made a number of helpful suggestions for its improvement.
- 15. Metaphor in Use Most of all, we would like to thank those students and colleagues from our Department who kindly lent their help with the organization of the RaAM7 confer- ence and hence made this volume possible: Carolina Amador, Elisabeth Amaya, Naomi Chaillou, Gemma Delicado, Montaña Durán, Denise Elekes, Montaña González, Sara Hoyas, Kerr Marín, Ignacio Portero, and Rosa Sánchez.
- 16. Introduction Metaphor in use Fiona MacArthur and José Luis Oncins-Martínez University of Extremadura, Spain 1. Background Although metaphor, or the human drive to ‘see’ or understand one thing in terms of another, is probably a universal, even perennial phenomenon, its manifestations most certainly are not. Even if we were only to consider the way that metaphor is used in communication among speakers of English, one of the most striking facts to emerge from research in recent years is how variable metaphor use is and how its production and interpretation in context depends on the interplay of many different factors. Among these is the means people use to convey a metaphorical idea, for it must be borne in mind that metaphors are not realized solely in language: gesture, visuals (whether static or moving), and other modes of expression are also vehicles that pub- licly display the way that people conceive of one thing in terms of another. In turn, these different modes of metaphorical communication may also interact with each other and with language in various different ways (Chuang, this volume; Cienki 1998, Cienki and Müller 2008, Forceville 2007, Forceville and Urios-Aparisi 2009), which adds further complexity to the use of metaphor in context. Apart from the different modes employed (speech, writing, gesture, or visuals, for example), another factor that has been shown to influence metaphor production and comprehension is the time scale in which it is used. Since metaphor use occurs in real time, attention to its presence and absence as discourse unfolds reveals the variability and unevenness of this phenomenon both within and across discourse events. Several researchers have noted that metaphors are not evenly distributed in discourse events such as conversation or lectures, but tend to occur in bursts, or cluster in response to different factors, such as management of the ongoing discourse, the topic, or even in- terpersonal relations (Cameron 2008, Cameron and Stelma 2004, Corts and Pollio 1999). Cameron (2008: 200), for example, has observed that “when one speaker uses metaphor, other speakers seem more likely to adapt their own talk and become meta- phorical in response”.
- 17. Fiona MacArthur and José Luis Oncins-Martínez Even though the primary site for human communication is conversation, speakers of English do not appear to use linguistic metaphors as frequently when they are chat- ting to each other face to face as they do in the written medium (Steen et al. 2010), so another factor that contributes to metaphor variation is the discourse contexts in which it is used. Furthermore, certain written registers display a much greater density of metaphor use than others. Steen et al. (2010) have found that metaphor is used much more frequently in academic discourse than in fiction, a perhaps somewhat sur- prising finding given the traditional emphasis on metaphor as a trope peculiar to poetry and fictional prose. But even within academic discourse, for example, meta- phor use varies: different academic discourse communities use metaphor in different ways. The metaphors used by economists, for example, when writing and talking about their discipline are not the same as those used by architects when dealing with theirs (Alejo 2010, Caballero 2006), for the metaphor systems or models that constitute par- ticular theories or frame the problems that disciplines seek to explore and resolve (Kuhn 1993) vary across different areas of enquiry. Indeed, major paradigm shifts may be marked by changes in the metaphors conventionally used in a field of scientific enquiry (see, for example, Aitchison’s [2003] discussion on competing metaphors for understanding linguistic change), which recalls the importance of the diachronic di- mension as one more factor that contributes to metaphor change and variation. When studied in a historical time scale, metaphor has been revealed to play an important role in motivating semantic change in English (e.g. Allan 2008, Kay 2000, Sweetser 1990), and research adopting a diachronic perspective on metaphor use has not only provided details about the processes involved in how word meanings change in the course of time, but has also shed light on the status of particular utterances as “metaphors” for speakers of earlier and later generations (Alm-Arvius, this volume; Geeraerts and Grondelaers 1995, Oncins-Martínez 2006), for consideration of meta- phor in various time scales reveals that what might count as a metaphor at one time and in one context might be regarded somewhat differently in another. For example, one of the time scales in which metaphor has been widely researched – the ontoge- netic – has further revealed the complexity of this phenomenon and how difficult it may be to decide on whether the unconventional ‘metaphor-like’ utterances of children should be considered metaphors at all (Cameron 1996). Piaget (1962) re- ported his daughter between the ages of 3 : 6 and 4 : 7 saying that a winding river was like a snake and comparing a bent twig with a machine for putting in petrol. While Piaget himself regarded these as ‘child metaphors’ as opposed to ‘real metaphors’ (describing them as nothing more than products of the symbolic, imagistic type of thinking that characterizes the pre-operational stage), other researchers have used different criteria to distinguish metaphors and pseudo-metaphors in children’s speech (e.g. Billow 1981, Nerlich et al. 1999, Vosniadou and Ortony 1983, or Winner 1988), reaching different conclusions about what distinguishes a child’s use of metaphor from an adult’s, and how the changes in children’s use and understanding of meta- phor at different ages can be accounted for.
- 18. Introduction The complexity of the task of researching metaphor is perhaps most apparent when we move away from a consideration of metaphor solely in relation to English speakers or even speakers of other standard European languages. As Leezenberg (2001: 15) has pointed out, there are certain “cultural prerequisites for a notion of metaphor”. A similar point is made by Goddard when he notes that the term ‘meta- phor’ lacks precise equivalents in many of the world’s languages, and warns of the dangers of uncritically adopting the category as a starting point for cross-cultural com- parison (2004: 1212). Both authors discuss the issue in relation to A is B (active or expository) metaphors, and Leezenberg (2001: 15) cites the disagreement over inter- pretations of the much debated utterance of the Bororo Indians of Brazil pa e-do nabure (‘we are parrots’). Early accounts (e.g. Durkheim and Mauss 1963: 6–7) suggested that the Bororo did not distinguish between the categories of people and animals, and this expression could not therefore be classed as a metaphor. However, close attention to the linguistic form of the utterance (Turner 1991: 135–136) has provided grounds for thinking that it should not be regarded as a ‘literal’ statement or a conflation of the categories people and birds/animals, because it can only be used to refer to men and the verb is marked for ‘customary form’ rather than ‘permanent state’ (Leezenberg 2001: 16). In the light of close linguistic analysis, then, the utterance can be regarded as instantiating the metaphorical mapping people are animals. In fact, as numerous studies over the years have shown, there appears to exist a very widespread tendency to ‘see’ people as animals, although the instantiation of the mapping varies consider- ably across different language-speaking communities. The use of the same animal names to refer to people may be similar or quite different in different languages (e.g. Hines 1999, Hsieh 2006, López Rodríguez 2009, Talebinejad and Dastjerdi 2005), as are the preferred ways of instantiating the metaphor in everyday speech (Deignan 1999). Similarly, while it seems true that “the existence of the semantic prime body invites people to theorise about the other parts of a person” (Goddard 2003: 122), the way that speakers of different languages establish these relations varies considerably. The head, the heart, the liver, the ear, and the stomach are some of the body parts and organs associated with ‘thinking’ or ‘feeling’ in different languages (Goddard 2003, Wierzbicka 1992, Yu 2007, 2009) but although body part for thought/feeling might be a common pattern, the type and value of the thoughts or emotions associated with each body part is often different across languages. Goddard (2003: 124) describes, for example, the hati (liver) concept in Malay as: very ‘feeling-oriented’ but focused primarily on interpersonal feelings. [...] the hati is viewed as an inner domain of experience, but there is a heightened empha- sis on its motivational consequences, along with a certain moral ambivalence. On account of the hati, a person may have an urge to do bad things as much as good things (hence one ought not unthinkingly or impulsively follow one’s hati; as the saying goes, ikuthati mati ‘follow the hati, die’).
- 19. Fiona MacArthur and José Luis Oncins-Martínez Likewise, although several languages instantiate a hand for control metonym, the way that it is realized and used by different language-speaking communities can also vary. Yu (2000), for example, finds that English and Chinese highlight different sub- parts of the hand in expressing this relation. More importantly, perhaps, the evaluation conveyed by the expressions that instantiate this metonym may be quite dissimilar: Charteris-Black (2001) notes that Malay expressions with tangen imply interference or meddling while English equivalents with hand evaluate this control positively. Researchers may be content to note that socio-cultural factors cause such cross-Â� linguistic and cross-cultural differences or seek to find more detailed explanation for them (e.g. MacArthur 2005). However, this should not cause us to lose sight of the possible consequences that such differences may have for cross-cultural communica- tion, where more applied metaphor research is still needed. For instance, misunder- standings or miscommunication may result when speakers whose languages differ from each other in these subtle but important ways communicate with each other, as happens when native speakers of English interpret Japanese figurative expressions us- ing body part terms when these are translated into English (see Azuma, this volume). In short, although metaphorizing may be “a natural function of the human mind” (Morgan 1993: 132) and metaphor may be used by people all over the world, the met- aphors found in different linguistic communities are subject to the contextual variation observable in a single language, and a search for universal patterns may thus detract attention from the diverse and variable ways that metaphor is employed by speakers in different cultural contexts. In an increasingly globalized world, where communication between different cul- tural groups is not only facilitated by media such as the Internet but indeed made nec- essary by large-scale transnational migration or the federation of nation states, such as the European Union, the growing interest in the relationship between metaphor, cul- ture, and context is to be welcomed. In recent years, various studies have done much to contribute to our understanding of cross-cultural and cross-linguistic differences in metaphor use worldwide and context induced variation (e.g. Kövecses 2005, 2010). For example, Kövecses (2000) describes how metaphors may be motivated by the culturally or physically salient experiences of particular language-speaking groups which may, in turn, vary quite substantially from one to another. This would account for the fact that certain source domains motivate a large number of metaphorical expressions in certain languages but not in others (e.g. Boers 1999). This would explain why a speaker of Spanish might use a metaphor such as echar un capote a alguien (lit. ‘to throw someone a cape’) in order to express the notion of helping another person, while a speaker of English would not, for bull-fighting is not an everyday, familiar area of experience for those from outside the Spanish-speaking world. However, it does not explain why an English speaker (and not a Spaniard) might use a maritime metaphor like ‘bail some- one out’ to express the same idea, because the sea is salient not only for people from the British Isles: Spain, along with other countries, also has a long sea-faring tradition. In- deed, the difficulty of establishing a direct relationship between metaphor and culture
- 20. Introduction (Deignan 2003, Deignan and Potter 2004) has led Deignan to propose that the relation- ship is indirect, and that many metaphors may survive in languages as “cultural relics” (Deignan 2003). This conclusion is not altogether surprising or unusual. After all, as Tomasello (1999) has pointed out, one of the important functions of language is to preserve the cultural lessons of the past, and to ensure their transmission – even when some may have become irrelevant or obsolete. Language can be seen as the prime means for communicating cultural ideas and beliefs (Sperber 1996). Language is both a part of a people’s culture and a vehicle for its transmission, It is tempting to see culture as a set of ideas and beliefs shared by a community that influence in relatively predictable ways the actions and behaviour of that group (e.g. Hall 1981, Hall and Hall 1990, Kövecses 2005). However, it may be more helpful to understand cultural conceptualizations as more variable and dynamic than this. For example, Sharifian (2011) considers culture as one type of complex adaptive system, which is, in turn, nested in other complex adaptive systems, including individual peo- ple, the language they speak, or the physical environments they inhabit. In this view, cultural cognition – or the shared views of a community of people – is a complex sys- tem in that an individual’s cognition does not capture the totality of his/her cultural group’s cognition (Sharifian 2011: 23). Furthermore, cultural cognitions – just like in- dividual cognitions – have their own unique history of interactions that constantly construct and reconstruct the system. And among the history of interactions of indi- viduals or groups that are of particular interest in an era of globalization are those that involve contact with other groups, a phenomenon that has always been of interest in diachronic studies of individual languages, but less so to metaphor researchers (but see Trim 2007, this volume). An example of how contact between different cultural groups may bring about changes in metaphor use is provided by Goddard (2004). He de- scribes how speakers of the Western Desert language Pitjantjatjara/Yankunytjakjara now employ a certain number of expository metaphors in non-traditional discursive domains (for example, in talk about Christianity), which Goddard attributes to con- tact between the aboriginal peoples and speakers of English, particularly through mis- sionary efforts (Goddard 2004: 1218–1219). New metaphorical language may emerge from such situations of contact and, on occasion, become entrenched in the language used by a group of speakers. Thus, a regional variety of a standard language may show traces of prolonged situations of language contact. For example, the interlanguage of Irish Gaelic speakers of English resulted in the coinage of the metaphorical idiom used in Hiberno-English: ‘to put something on the long finger’ (from Irish Gaelic chuir ar an méar fada é) (Odlin 1991). In this regard, then, studies of metaphor use in the in- terlanguage systems of learners of a foreign language, like those of Golden and Johansson Falck in this volume, are relevant not only to applied linguists interested in making pedagogical use of such studies, but also for understanding the processes in- volved in the emergence of new metaphorical uses of language and the short and long- term consequences for the varieties of languages that emerge from such contact. Sharifian (2010) rightly states that “it would be naive to expect a speaker to become a
- 21. Fiona MacArthur and José Luis Oncins-Martínez culturally and emotionally different person when speaking a second language”, so it is not surprising that culturally induced ways of ‘seeing-as’ should lead to new meta- phorical language uses, an area of study of particular relevance to the phenomenon of global Englishes. At present, non-native speakers of English far outnumber those who speak it as a first language (Kirkpatrick 2010). The spread of English is resulting in the rise of varieties that are different from native speaker norms, and these differences are also apparent in metaphor use in different varieties. For example, Polzenhagen and Wolf (2007) have described the culture-specific conceptualization of corruption in African English and how this is reflected in the linguistic metaphors speakers of this variety use when talking about this topic. 2. The contributions to this volume As these introductory remarks have aimed to show, metaphor is a complex and multi- faceted phenomenon. Indeed, it seems well-nigh impossible for any one theory of metaphor to account fully for the complexity of metaphor as used by human beings in communication with each other, as Gibbs (2006: 435) has pointed out. It is thus not surprising to find that the sixteen chapters in this volume should not adhere to one single method or approach, but range from the computational (Veale or Berber Sardinha, for example) to more traditional, philological approaches (Alm-Arvius or Trim) through research guided by the precepts of conceptual metaphor theory or CMT (Johansson Falck or Aksan and Aksan). What they all have in common, however, is their focus on the situated use of metaphor in different contexts and their use of real data to underpin the research they report, whether this comes from very large, com- mercially available corpora (for example, Johansson Falck or Dorst and Kaal), data gathered with the help of Internet search engines such as Google (Alm-Arvius or Veale), specially compiled corpora (for example, Golden, Trim, Chapetón-Castro and Verdaguer-Clavera, or Aksan and Aksan), or smaller amounts of real world data gath- ered for the specific purposes of the research being carried out (Van Mulken and Le Pair, Chuang, or Azuma). Indeed, one of the charges made against CMT is that the linguistic data used to illustrate conceptual mappings has often been the result of the analyst’s introspection and that the examples used to support their proposals often do not fully account for the way that metaphors may be realized in language (Ritchie 2003, Semino 2005, Stefanowitsch 2006). In this regard, one of the contexts of research that has revolutionized the way that metaphor may be studied in the last 30 years or so is the availability of large electronic corpora that allow researchers to have access to much larger amounts of linguistic data than was formerly possible. This new research context has contributed to providing more robust descriptions of the way that metaphors are realized in everyday discourse (for example, Deignan 2005, Gries 2006, Hanks 2006, Stefanowitsch 2006). At the same time, the task of identifying and quantifying meta- phors in large corpora poses a number of challenges to metaphor researchers and
- 22. Introduction raises a number of questions. Among these are: how can metaphors be identified and retrieved in very large corpora? How can they be quantified? Is it necessary to have identified metaphorical language uses in advance or is it possible to mine large cor- pora in a data-driven way? Are the methods that have been developed for identifying metaphors in English applicable to other languages as well? The four chapters that make up the first part of the book address these issues. 2.1 Part 1: Contexts of research In the first chapter, “An assessment of metaphor retrieval methods”, Tony Berber Sardinha evaluates a number of different techniques and tools for retrieving metaphor in large corpora, explaining in detail for researchers who are not experts in computa- tional linguistics themselves how each can be used and how reliable each procedure is in terms of the number of metaphors retrieved. As Berber Sardinha’s work in this field has shown, the methods and techniques he explores are applicable to both English and Brazilian Portuguese. The second chapter, “Metaphor in discourse: Beyond the boundaries of MIP”, by Aletta G. Dorst and Anna Kaal, two researchers in the MIPVU project at the Free University of Amsterdam, is similarly concerned with the identification and accurate quantification of metaphor in discourse, but takes a much closer look at the decisions that must be taken by researchers when identifying metaphorical uses of language. Dorst and Kaal describe some of the problems that arise in applying the Method for Identifying Metaphors (MIP) (Pragglejaz Group 2007) to direct metaphors and meta- phorical comparisons, explaining in detail how decisions can be taken in order to pro- vide robust and replicable methods of metaphor identification in discourse, which is important, above all, in quantifying such uses of language for comparative purposes. Chapter 3, “Metaphor identification in Dutch discourse”, is by another researcher in the MIPVU project, Trijntje Pasma. Unlike her colleagues, the author discusses MIP in relation to Dutch and illustrates how the method, originally conceived to deal with English discourse, can be used to identify metaphors in another European lan- guage when appropriate modifications are made for the morpho-syntactic peculiari- ties of the language involved. The last chapter in this section – “Locating metaphor candidates in specialized corpora using raw frequency and keyword lists”, by Gill Philip – is concerned with the automatic retrieval of metaphors from large corpora. However, unlike Berber Sardinha, Philip deals with corpora made up of homogeneous texts (that is, texts that all deal with the same topic), a characteristic that allows the researcher, with the help of key- words and raw frequency lists, to distinguish between metaphors and ‘terminology’ (i.e., words and expressions that appear metaphorical to people from outside the dis- course community that uses them, but that may not be regarded as such by members of the discourse community that uses them with particular fixed or stable meanings). Philip is also concerned with explicating a method for automatically retrieving
- 23. Fiona MacArthur and José Luis Oncins-Martínez metaphors from large corpora without the need for a researcher to have advanced command of corpus linguistics methodology or tools, and one that uses commercially available software. And, in line with Pasma’s chapter, she explains how this method can be applied to another language, in this case, Italian. The four chapters in this first section, then, explicate ways of identifying and re- trieving metaphorical language uses that can be applied by metaphor researchers with no background in computational linguistics or by those who do not have access to the specialized software that has been developed for these purposes. Furthermore, the various methods described extend the contexts in which metaphor identification may be reliably carried out, by considering their use with languages other than English. Although the focus here remains on standard European languages (but see Chuang, this volume, for an illustration of how MIP was applied to Mandarin Chinese), they may suggest ways of developing methods of metaphor identification and retrieval ap- plicable to other, typologically different languages, in order that future research into metaphor use in these contexts may contribute to similarly robust findings that can be compared with each other and with studies that have been carried out into English. 2.2 Part 2: Contexts of production The three chapters in this section all examine how metaphorical language is used by non-native speakers (NNS) of a language, comparing this with native-speaker (NS) norms as found in the control corpora used. In this regard, one thing that all these studies reveal is the importance of the appropriate choice of the NS corpora, depend- ing on the research questions the analyst is seeking to answer. The study reported in Chapter 5, “Metaphor variation across L1 and L2 speakers of English: Do differences at the level of linguistic metaphor matter?” by Marlene Johansson Falck, focuses on the linguistic realization of motion metaphors (actions are self-propelled movements, purposes are destinations or an activity is a journey) in ‘path’, ‘way’, and ‘road’ expressions. It offers a detailed analysis of how these are used by advanced learners of English with Swedish as their mother tongue in comparison to how these expressions are used by NSs of English in the texts contained in the British National Corpus (BNC). Johansson Falck’s study is specifically concerned with discovering to what extent the linguistic means for expressing motion metaphors in Swedish influence these learners’ use of similar metaphors in English, as Swedish has only two forms, stig and vag, to describe the different types of routes that can be taken – literally and metaphorically – from one place to another. The very detailed analysis offered of the use of ‘path’, ‘way’, and ‘road’ in English in these two contexts reveals that, while the Swedish speakers of English as a second language with advanced competence in the language did not produce any erroneous or incomprehensible ut- terances, there were interesting quantitative and qualitative differences between their uses of these expressions and that of NSs, suggesting that even when two languages share primary and complex metaphors, the precise way that these are expressed in the
- 24. Introduction first language subtly alter the way that these mappings are conceived. These findings have implications not only for foreign language teaching, but also for cross-cultural metaphor research, because they show the importance of language in shaping culture- specific conceptualizations. Like Johansson Falck’s, the study reported by Anne Golden in Chapter 6 – “Metaphorical expressions in L2 production: The importance of text topic in corpus research” – focuses on one specific area of language use: in this case, the high fre- quency Norwegian verb ta (roughly equivalent to English ‘take’) as used by NNSs with three different mother tongues (L1s): German, Spanish, and Russian. Golden com- pares these learners’ uses of this verb with that of Norwegian students’ in order to ex- plore the differences between the way these groups of speakers employ the verb in its basic or metaphorical sense, but distinguishing also between the use of ta in fixed col- locations or as ‘bridge terms’ (Kittay 1990). Among the findings that emerge from this study is that, although differences in metaphorical uses of ta can be observed among the three different NNS groups, related both to their L1 and to their command of the second language (L2), the topic of the written discourse proves the most important variable: in the control corpus employed, the NSs of Norwegian were found to use ta with metaphorical senses less frequently than the NNSs. The conclusions drawn echo Cameron’s observation (2008: 203) that the absence of metaphor is as significant as its presence in discourse, and the density of metaphor use is often related to what is being talked about. In this regard, Golden’s chapter sheds light on some of the problems that are involved in attempting to relate competence in a foreign language with metaphoric competence (Danesi 1993, Littlemore and Low 2006). Context or the topic a NNS needs to talk about also influences L2 learners’ use of metaphor. Unlike the preceding two chapters, in “Researching linguistic metaphor in native, non-native, and expert writing”, Claudia Chapetón-Castro and Isabel Verdaguer- Clavera do not start from consideration of the metaphorical use of any particular lexical items when comparing NSs and NNSs writing, or how the first language may influence metaphor production in the second language, but rather seek to discover more general patterns in the different corpora they examine. In order to do so, their study involved identifying all potentially metaphorical uses of language. In their chap- ter, they describe in detail how the combination of two different methods of identify- ing metaphors in discourse (through the identification of vehicle terms, as developed by Cameron [2003] and MIP [Pragglejaz 2007]) enabled them not only to reliably identify the metaphors in the texts they examined, but also to point out how this pains- taking approach to metaphor identification obliges the researcher to engage closely with the linguistic form of the metaphors used. Using this combined procedure, Chapetón-Castro and Verdaguer-Clavera carry out a three-way comparison between the use of metaphor by undergraduate Spanish learners of English with that of NS undergraduate students and that of NS expert writers. The findings provide interesting detail about the similarities and differences between the three groups of writers, not only as far as the linguistic forms of the metaphors used are concerned, but also as
- 25. Fiona MacArthur and José Luis Oncins-Martínez regards the density of the metaphors employed. The most significant differences were not to be found in the writers’ L1 but rather in their age or expertise: both groups of undergraduate students used metaphors less frequently than the expert writers. 2.3 Part 3: Contexts of interpretation The three chapters that make up the third section all present cross-linguistic and cross- cultural analyses of metaphor interpretation, and shed light on some of the factors that give rise to similarities and differences in metaphor interpretation and appreciation across different cultural groups. In “Appreciation and interpretation of visual meta- phors in advertising across three European countries”, Margot van Mulken and Rob Le Pair consider how advertising campaigns that target consumers in different European countries may employ visual metaphors in their advertisements, on the as- sumption, it seems, that they will be understood and appreciated in similar ways by consumers with different cultural backgrounds. These researchers investigated this as- sumption by gathering data from French, Dutch, and Spanish informants in response to different types of visual metaphors used in advertising, whose visual ‘syntax’ may encode a metaphor more or less explicitly. They found that the three cultural groups appear very similar in their preference for certain types of visual metaphors; however, subtle differences in interpretation across the three groups were detected, although these did not correspond to the division of the groups of informants as belonging to a high or low context culture. According to Hall and Hall’s (1990) classification of low and high context cultures, the interpretations of the Spanish and French informants should have borne a similarity to each other, as their communication has been claimed to rely on a specific situational context for interpretation, while the Dutch informants’ interpretations would be different, as members of this low context culture would be more dependent on clear and explicit articulation of an idea in order to interpret it successfully. These results did not obtain, however, suggesting that further cross-cul- tural research of this type would be very valuable for understanding the relationship between metaphor and culture. Researchers may confuse the effects of language knowledge with the effects of the shared cultural beliefs and values of different communities that are expressed through linguistic metaphors, and yet knowledge of one’s own language is very important when interpreting metaphors, as the study reported in Chapter 9 by Masumi Azuma shows. In her contribution, “English native speakers’ interpretations of culture-bound Japanese figurative expressions”, Azuma examines the way native speakers of English interpret culture-bound figurative expressions when they are translated literally from Japanese, pointing to some of the different factors that influence the way they may be interpreted by NNS of Japanese. A particularly interesting finding that emerges from this research is that the interpretation of familiar and unfamiliar metaphorical lan- guage uses relies heavily on knowledge of the mother tongue (a finding in line with Johansson Falck’s), for the participants in this study came from different parts of the
- 26. Introduction English-speaking world (the U.S., Britain, and Australia) and yet interpreted the Japanese metaphorical expressions in very similar ways, despite differences in their social or cultural background. The distance that separates the speakers of English and Japanese that took part in Azuma’s study is not just geographical. The two languages are typologically different, and the cultural traditions of each have developed independently of each other. This is not the case of the situation considered in Chapter 10, “The limits of comprehension in cross-cultural metaphor: Networking in drugs terminology”, by Richard Trim, where a comparison is made across different European languages. The common cul- tural heritage of Europeans is evident in many of the metaphors shared by speakers of different Western European languages, inherited, for example, from such influential texts as the Bible or Æsop’s fables. However, the various languages spoken in Europe also display culture-specific metaphorical language uses. In the last chapter in this sec- tion, Trim explores the linguistic and conceptual features of metaphors that may make them more or less transparent to non-native speakers of a language in a European context, focusing on metaphors used to talk about drugs in English, German, French, and Italian. The author finds that various factors cause the metaphors used to talk about the same topic to converge or diverge. These factors may contribute to making some metaphors reasonably transparent for speakers of other standard European lan- guages, while other metaphors will be more difficult to understand. For example, shared conceptualizations give rise to similar – and hence reasonably transparent – metaphorical language uses across Europe, although they may not be realized or used in exactly the same way in each language. In contrast, individual languages may recruit metaphorical expressions from other discourse contexts to extend the range of meta- phors used to talk about a topic like this. That is, the emergence of unfamiliar – and possibly opaque – metaphorical language uses may be influenced by the entrenched metaphorical meanings associated with certain linguistic forms used in talk about other topics in that particular language. 2.4 Part 4: Metaphor, topic, and discourse Part 4 contains two chapters that further explore the importance of topic and context in cross-cultural metaphor research. In Chapter 11, “Conceptual types of terminologi- cal metaphors in marine biology: An English-Spanish contrastive analysis from an experientialist perspective”, José Manuel Ureña examines metaphors in the field of marine biology from a cross-linguistic perspective, analysing terms in Spanish and English for designating different kinds of sea creatures. This metaphor-driven cross- linguistic analysis reveals that multiple correspondence metaphors give rise to virtually identical metaphorical names in both languages, while metaphors based on resem- blance in shape (or image metaphors) tend to be subject to greater variation and are, the author suggests, more susceptible to cultural influence.
- 27. Fiona MacArthur and José Luis Oncins-Martínez In Chapter 12, “Gestures, language, and what they reveal about thought: A music teacher’s use of metaphor in Taiwan”, Ya-Chin Chuang explores the metaphors used for explaining music in a secondary school classroom in Taiwan. The close analysis of a single class session allows Chuang to examine the interaction of metaphor realized in language and in gesture and to relate these to the different phases of the class and the different functions they fulfil, finding that, although overall the gestures used by the teacher showed a tendency to cluster at different points, this clustering was not a fea- ture of the metaphorically-used gestures. In line with earlier findings (e.g. Cienki 1998 or Cienki and Müller 2008), Chuang’s study shows that a metaphorical gesture can express the same metaphorical idea expressed in language at the same time in the on- going talk or a different one. Likewise, a gesture can express a metaphorical idea that is not accompanied by a corresponding metaphorical use of language uttered at the same time. Chuang even found instances of metaphorical mappings expressed by gesture that are never instantiated in linguistic form in Mandarin Chinese. This study thus replicates earlier work focusing on gesture that has been able to locate metaphor in thought, and – perhaps most importantly – provides evidence that this is not a conse- quence of any ethnocentric bias on the part of previous researchers, and that the phe- nomenon is not restricted to Indo-European languages, for the language used in this classroom is typologically different from those that have been the focus of attention when examining the relations between metaphor in speech and gesture. Chuang de- scribes how MIP (Pragglejaz Group 2007) was applied to Mandarin Chinese, discuss- ing the issues raised by this methodological decision, and also discusses the problems associated with accurately identifying metaphorical gestures. This chapter thus illus- trates the importance of finding robust and replicable methods for identifying meta- phors in discourse, whatever the language or the mode in which they are expressed. 2.5 Part 5: Metaphor and culture Although many of the preceding chapters have touched obliquely on the relationship between metaphor and culture, a fuller exploration of this relationship and its mani- festation in language is offered by the chapters in Part 5. In Chapter 13, “Armed with patience, suffering an emotion: The conceptualization of life, morality, and emotion in Turkish”, Yeşim Aksan and Mustafa Aksan describe in detail the cultural models expressed by the Turkish lexemes çile (very roughly, English ‘suffering’) and sabir (very roughly, English ‘patience’), tracing the root of these culturally salient concepts to centuries-old religious practice and values (çile, for example, referred to the institu- tionalized practice of Sufi ascetics of observing a 40-day period of fasting). These words, and the specific concepts they express, metaphorically extend their meaning to other realms of experience (life, morality, and emotion), and both constitute and re- flect the cultural beliefs of speakers of Turkish. The linguistic evidence provided shows that the cultural models of çile and sabir underlie contemporary speakers’ beliefs about morality, emotion, and, in general, how life should be lived, and testify to the stability
- 28. Introduction of a metaphorical idea across centuries in a particular language-speaking community. In contrast, although in Chapter 14 – “Trolls” – Christina Alm-Arvius finds that the Scandinavian cultural complex troll has likewise survived the passage of time, and in- deed changes in ideology (for speakers of Swedish no longer believe that trolls really exist), metaphorical uses of troll in contemporary language uses reveal contradictory senses and evaluations. The primarily negative evaluations of both conventional and novel metaphors with troll are employed side by side with others with positive con- notations, such as when they are used as terms of endearment or to refer to a child. Nevertheless, Alm-Arvius finds that all instances of troll metaphors are attitudinally coloured in Swedish, a feature that is lost when the term is adopted by another lan- guage such as English. As can be seen in the studies offered in these two chapters, language – and in par- ticular, metaphorical language – may preserve the enduring cultural values of a lan- guage-speaking community and prove a prime vehicle for propagating them (Sperber 1996, Tomasello 1999). However, language must also provide the means for the expression of new ideas and relations. In his chapter “A computational exploration of creative similes”, Tony Veale considers what linguistic signals are necessary for the identification and interpretation of creative as ... as similes in English. Using a large corpus of this type of simile in contemporary English, he describes how the word ‘about’ or the length of the metaphorical vehicle may function as scaffolding structures used by speakers and writers in English to alert listeners and readers to the humorous or ironic intent of a metaphorical simile. In this regard, Veale’s study proves an impor- tant complement to earlier chapters in this volume (Chapters 2 and 7, in particular) by showing how a computational approach to similes in English can supplement more qualitative approaches, and further add to our knowledge of their pragmatic function in discourse. 2.6 Part 6: Afterword and prospects for future research The final chapter, “Metaphors, snowflakes, and termite nests: How nature creates such beautiful things”, by Raymond W. Gibbs, Jr., provides an afterword to the various strands explored in the different chapters. The complexity and variety of metaphor as used and interpreted in context can best be understood, he argues, if we regard this phenomenon as one type of complex dynamic or self-organizing system. The focus of Gibbs’ chapter is on the role of multiple attractors, the hierarchy of time-scales, and the dynamics of processing, global emergence, and top-down causality in self-organiza- tional processes of metaphor use. His approach is thus very much in line with other theories of complexity emerging from the natural sciences (e.g. Holland 1995, 1998) that are having a profound effect on the social sciences and arts. For example, a major paradigm shift seems well under way in second language acquisition research in accordance with complex dynamical systems theory (e.g. de Bot et al. 2007, Larsen-Freeman 2006, The Five Graces Group 2009). Moreover, the value of this
- 29. Fiona MacArthur and José Luis Oncins-Martínez perspective has also been advocated by Cameron and her colleagues in relation to re- searching metaphor in discourse (Cameron et al. 2009). Although none of the contributors to the volume would necessarily espouse the views put forward by Gibbs, his analysis nevertheless sheds light on many of seem- ingly intractable problems in metaphor research and the on-going debates about it – particularlyonthedisagreementsamongpsycholinguistsaboutthestatusofconceptual mappings in people’s minds and how they may or may not be activated in online pro- cessing. In this regard, this important chapter sets an agenda for future research and offers a glimpse of exciting new ways of approaching many of the complex, variable, and sometimes troublesome facets of metaphor as used in communication between human beings. References Aitchison, Jean. 2003. Metaphors, models and language change. In R. Hickey, ed., Motives for Language Change, 39–53. Cambridge: Cambridge University Press. Alejo, Rafael. 2010. Where does the money go? An analysis of the container metaphor in eco- nomics: The market and the economy. Journal of Pragmatics 2 (4): 1137–1150. Allan, Kathryn. 2008. Metaphor and Metonymy: A Diachronic Approach. Chichester: Wiley- Blackwell. Billow, Richard. 1981. Observing spontaneous metaphor in children. Journal of Experimental Child Psychology 31 (3): 430–445. Boers, Frank. 1999. When a bodily source domain becomes prominent: The joy of counting metaphors in the socio-economic domain. In R. W. Gibbs, Jr. G. J. Steen, eds., Metaphor in Cognitive Linguistics, 47–56. Amsterdam Philadelphia: Benjamins. de Bot, Kees, Wander Lowie, Marjolijn Verspoor. 2007. A dynamics systems theory approach to second language acquisition. Bilingualism: Language and Cognition 10 (1): 7–21. Caballero, Rosario. 2006. Re-Viewing Space: Figurative Language in Architects’ Assessment of Built Space. Berlin: Mouton de Gruyter. 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. Cameron, Lynne. 2008. Metaphor and talk. In R. W. Gibbs, Jr., ed., The Cambridge Handbook of Metaphor and Thought, 197–211. New York: Cambridge University Press. Cameron, Lynne Juup Stelma. 2004. Metaphor clusters in discourse. Journal of Applied Linguistics 1 (2): 107–136. Cameron, Lynne, Robert Maslen, Zazie Todd, John Maule, Peter Stratton, Neil Stanley. 2009. The discourse dynamics approach to metaphor and metaphor-led discourse analysis. Metaphor and Symbol 24 (2): 63–89. Charteris-Black, Jonathan. 2001. Cultural resonance in English and Malay figurative phrases: The case of ‘hand’. In J. Cotterill I. Ife, eds., Language across Boundaries, 151–170. London: Continuum.
- 30. Introduction Cienki, Alan. 1998. Metaphoric gestures and some of their relations to verbal metaphoric ex- pressions. In J. P. Koenig, ed., Discourse and Cognition: Bridging the Gap, 189–204. Stanford, CA: Centre for the Study of Language and Information. Cienki, Alan Cornelia Müller, eds. 2008. Metaphor and Gesture. Amsterdam Philadelphia: Benjamins. Corts, Daniel P. Howard R. Pollio. 1999. Spontaneous production of figurative language and gestures in college lectures. Metaphor and Symbol 14 (2): 81–100. Danesi, Marcel. 1993. Metaphorical competence in second language acquisition and second language teaching: The neglected dimension. In J. E. Alatis, ed., Georgetown University Round Table on Languages and Linguistics 1992: Language, Communication, and Social Meaning, 489–500. Washington, D.C.: Georgetown University Press. Deignan, Alice. 1999. Corpus-based research into metaphor. In L. Cameron G. Low, eds., Researching and Applying Metaphor, 177–199. Cambridge: Cambridge University Press. Deignan, Alice. 2003. Metaphorical expressions and culture: An indirect link. Metaphor and Symbol 18 (4): 255–271. Deignan, Alice. 2005. Metaphor and Corpus Linguistics. Amsterdam Philadelphia: Benjamins. Deignan, Alice Liz Potter. 2004. A corpus study of metaphors and metonyms in English and Italian. Journal of Pragmatics 36 (7): 1231–1252. Durkheim, Émile Marcel Mauss. 1963. Primitive Classification. Chicago: The University of Chicago Press. The Five Graces Group. 2009. Language is a complex adaptive system. Language Learning 59 (1): 1–32. Forceville, Charles. 2007. Multimodal metaphor in ten Dutch TV commercials. Public Journal of Semiotics 1 (1): 15–34. Forceville, Charles Eduardo Urios-Aparisi. 2009. Multi-Modal Metaphor. Berlin: Mouton de Gruyter. Geeraerts, Dirk Stefan Grondelaers. 1995. Looking back at anger: Cultural traditions and metaphorical patterns. In J. R. Taylor R. E. MacLaury, eds., Language and the Cognitive Construal of the World, 153–179. Berlin: Mouton de Gruyter. Gibbs, Raymond W., Jr. 2006. Metaphor interpretation as embodied simulation. Mind and Language 21 (3): 434–458. Goddard, Cliff. 2003. Thinking across languages and cultures: Six dimensions of variation. Cognitive Linguistics 14 (2): 109–140. Goddard, Cliff. 2004. The ethnopragmatics and semantics of ‘active metaphors’. Journal of Prag- matics 36 (7): 1211–1230. Gries, Stefan T. 2006. Corpus-based methods and cognitive semantics: The many senses of to run. In S. T. Gries A. Stefanowitsch, eds., Corpora in Cognitive Linguistics: Corpus-Based Approaches to Syntax and Lexis, 57–99. Berlin: Mouton de Gruyter. Hall, Edward T. 1981. Beyond Culture. New York: Anchor Books. Hall, Edward T. Mildred R. Hall. 1990. Understanding Cultural Differences. Yarmouth, MA: Intercultural Press. Hanks, Patrick. 2006. Metaphoricity is gradable. In A. Stefanowitsch S. T. Gries, eds., Corpus- Based Approaches to Metaphor and Metonymy, 17–35. Berlin: Mouton de Gruyter. Hines, Caitlin. 1999. Foxy chicks and Playboy bunnies: A case study in metaphorical lexicaliza- tion. In M. K. Hiraga, C. Sinha, S. Wilcox, eds., Cultural Psychological and Typological Issues in Cognitive Linguistics, 9–23. Amsterdam Philadelphia: Benjamins
- 31. Fiona MacArthur and José Luis Oncins-Martínez Holland, John H. 1995. Hidden Order: How Adaptation Builds Complexity. New York: Basic Books. Holland, John H. 1998. Emergence: From Chaos to Order. Oxford: Oxford University Press. Hsieh, Shelley Ching-yu. 2006. A corpus-based study on animal expressions in Mandarin Chinese and German. Journal of Pragmatics 38 (12): 2206–2222. Kay, Christian J. 2000. Metaphors we lived by: Pathways between Old and Modern English. In R. Roberts J. Nelson, eds., Essays on Anglo-Saxon and Related Themes in Memory of Lynne Grundy, 273–85. London: Centre for Late Antique and Medieval Studies, King’s College London. Kirkpatrick, Andy, ed. 2010. The Routledge Handbook of World Englishes. Abingdon: Routledge. Kittay, Eva F. 1990. Metaphor: Its Cognitive Force and Linguistic Structure. Oxford: Clarendon. Kövecses, Zoltán. 2000. The scope of metaphor. In A. Barcelona, ed., Metaphor and Metonymy at the Crossroads: A Cognitive Perspective, 79–92. Berlin: Mouton de Gruyter. Kövecses, Zoltán. 2005. Metaphor in Culture: Universality and Variation. Cambridge: Cambridge University Press. Kövecses, Zoltán. 2010. A new look at metaphorical creativity in cognitive linguistics. Cognitive Linguistics 21 (4): 663–697. Kuhn, Thomas S. 1993. Metaphor in science. In A. Ortony, ed., Metaphor and Thought, 2nd edition, 533–542. New York: Cambridge University Press. Larsen-Freeman, Diane. 2006. The emergence of complexity, fluency, and accuracy in the oral and written production of five Chinese learners of English. Applied Linguistics 27 (4): 590–619. Leezenberg, Michiel. 2001. Contexts of Metaphor. Oxford: Elsevier. Littlemore, Jeannette Graham Low. 2006. Figurative Thinking and Foreign Language Learning. Basingstoke: Palgrave Macmillan. López Rodríguez, Irene. 2009. Of women, bitches, chickens and vixens: Animal metaphors for women in English and Spanish. Cultura, lenguaje y representación 7: 77–111. MacArthur, Fiona. 2005. Whose hand and what is it holding? Hand for control revisited. In A. Wallington, J. Barnden, S. Glasbey, M. Lee, L. Zhang, eds., Proceedings of the Third Inter- disciplinary Workshop on Corpus-Based Approaches to Figurative Language, 35–41. Birmingham: University of Birmingham. Morgan, Jerry L. 1993. Observations on the pragmatics of metaphor. In A. Ortony, ed., Metaphor and Thought, 2nd edition, 125–134. Cambridge: Cambridge University Press. Nerlich, Brigitte, David D. Clarke, Zazie Todd. 1999. Mummy, I like being a sandwich. In K. U. Panther G. Radden, eds., Metonymy in Language and Thought, 361–383. Amsterdam Philadelphia: Benjamins. Odlin, Terence. 1991. Irish English idioms and Language Transfer. World-Wide Words 12 (2): 175–193. Oncins-Martínez, José L. 2006. Notes on the metaphorical basis of sexual language in Early Modern English. In J. G. Vázquez González, M. Martínez Vázquez, P. Ron Vaz, eds., The Historical Linguistics-Cognitive Linguistics Interface, 205–224. Huelva: Universidad de Huelva. Piaget, Jean. 1962. Play, Dreams and Imitation in Childhood. New York: Norton. Polzenhagen, Frank Hans-Georg Wolf. 2007. Culture-specific conceptualizations of corruption in African English: Linguistic analyses and pragmatic applications. In F. Sharifian G. B. Palmer, eds., Applied Cultural Linguistics, 125–168. Amsterdam Philadelphia: Benjamins. Pragglejaz Group. 2007. MIP: A method for identifying metaphorically used words in discourse. Metaphor and Symbol 22 (1): 1–39.
- 32. Introduction Ritchie, L. David 2003. Argument is war – Or is it a game of chess? Multiple meanings in the analysis of implicit metaphors. Metaphor and Symbol 18 (2): 125–146. Semino, Elena. 2005. The metaphorical construction of complex domains: The case of speech activity in English. Metaphor and Symbol 20 (1): 35–70. Sharifian, Farzad. 2010. Semantic and pragmatic conceptualizations within an emerging variety. In A. Kirkpatrick, ed., The Routledge Handbook of World Englishes, 442–457. Abingdon: Routledge. Sharifian, Farzad. 2011. Cultural Conceptualisations and Language. Amsterdam Philadelphia: Benjamins. Sperber, Dan. 1996. Explaining Culture: A Naturalistic Approach. Oxford: Blackwell. Stefanowitsch, Anatol. 2006. Corpus-based approaches to metaphor and metonymy. In A. Stefanowitsch S. T. Gries, eds., Corpus-Based Approaches to Metaphor and Metonymy, 1–16. Berlin: Mouton de Gruyter. Steen, Gerard, Aletta G. Dorst, Berenike Herrmann, Anna A. Kaal. 2010. A Method for Linguis- tic Metaphor Identification: From MIP to MIPVU. Amsterdam Philadelphia: Benjamins. Sweetser, Eve. 1990. From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure. Cambridge: Cambridge University Press. Talebinejad, M. Reza H. Vahid Dastjerdi. 2005. A cross-cultural study of animal metaphors: When owls are not wise! Metaphor and Symbol 20 (2): 133–150. Tomasello, Michael. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Trim, Richard. 2007. Metaphor Networks. The Comparative Evolution of Figurative Language. Basingstoke: Palgrave Macmillan. Turner, Terence. 1991. ‘We are parrots, twins are birds’. Play of tropes as operational structures. In J. Fernández, ed., Beyond Metaphor: The Theory of Tropes in Anthropology, 121–158. Stanford, CA: Stanford University Press. Vosniadou, Stella Andrew Ortony. 1983. Intelligence of the literal-metaphorical-anomalous distinction in young children. Child Development 54: 154–161. Wierzbicka, Anna. 1992. Semantics, Culture and Cognition. Oxford: Oxford University Press. Winner, Ellen. 1988. The Point of Words: Children’s Understanding of Metaphor and Irony. Cambridge, MA: Harvard University Press. Yu, Ning. 2000. Figurative uses of finger and palm in Chinese and English. Metaphor and Symbol 15 (3): 159–175. Yu, Ning. 2007. Heart and cognition in ancient Chinese philosophy. Journal of Cognition and Culture 7 (1/2): 27–47. Yu, Ning. 2009. The Chinese HEART in a Cognitive Perspective: Culture, Body, and Language. Berlin: Mouton de Gruyter.
- 34. part 1 Contexts of research
- 36. chapter 1 An assessment of metaphor retrieval methods* Tony Berber Sardinha São Paulo Catholic University, Brazil This chapter offers a quantitative assessment of different techniques and tools for retrieving metaphors from large electronic corpora. These are (i) reading parts of a larger corpus in order to find candidates that are then sought for in the whole corpus; (ii) searching for metaphors using different search terms; (iii) looking for metaphor clusters; (iv) finding metaphor candidates through keywords; (v) finding metaphor candidates through the Metaphor Candidate Identifier; and (vi) finding metaphor candidates by computing semantic relatedness between neighbouring words. Keywords: automatic retrieval, corpora, metaphor identification, procedures, tools 1. Introduction Metaphor identification is a vast field that encompasses a large array of procedures, techniques, and tools. It involves at least two distinct phases: retrieval and analysis. During retrieval, occurrences of potentially metaphorical strings are extracted from the corpus and stored, and during analysis, these occurrences are actually evaluated in terms of whether they are cases of metaphor or not. Hence, when researchers refer to identifying metaphors, they normally mean determining which textual units (usually words) are metaphors and which are not, and not simply finding candidates, or possible metaphors. In this chapter, the focus is on the retrieval part of metaphor identification. We can further break down retrieval procedures into two basic groups: sampling techniques and census techniques. Sampling is “the selection of a fraction of the total * The author is grateful to CNPq (Brasília, Brazil) for grants # 307307/2006-9, 400574/2007- 1, 450239/2006-3, 350455/2003-1, and Capes for grant # 0397/04-0, as well as the Researching and Applying Metaphor (RaAM) International Association for their support. I’d also like to thank the two anonymous reviewers for their thorough revision and helpful comments.
- 37. Tony Berber Sardinha number of units of interest to decision makers for the ultimate purpose of being able to draw general conclusions about the entire body of units” (Parasuraman et al. 2004: 333). A sampling technique for corpus-based metaphor research would then involve selecting a pool of units (normally word types or lemmas) to represent the to- tality of words in the corpus. Census techniques, on the other hand, are those in which “every population unit is examined” (Parasuraman et al. 2004: 359), and therefore in metaphor research this would ultimately mean that researchers would have to analyse each token in the corpus. Sampling techniques are more common in corpus-based metaphor research than census ones, probably because of the fact that current electronic corpora are normally too large to analyse unit by unit (e.g. word by word). Examples of sampling techniques include determining search strings ahead of time, using a corpus to determine a pool of strings, choosing keywords or words with marked frequency, choosing words as- sociated with a particular semantic field via automatic semantic tagging, focusing on metaphor clusters or words near a previously-identified metaphor, obtaining a list of candidates through specialized metaphor detecting software, and selecting words based on semantic distance, among others. Census techniques include the Metaphor Identification Procedure (MIP) (Pragglejaz Group 2007), its variant MIPVU (referring to Vrije University, where it was developed) (Steen et al. 2010), and Metaphor Identification through Vehicle Terms (MIV) (Cameron and Maslen 2010). Each of these encompasses a number of specific steps for metaphor identification, which are detailed in their respective publications. These are not reviewed here, as they are not relevant to sampling and accuracy issues, which are the focus of this chapter. The promise offered by machine identification of metaphor is that computers will take a census of the metaphor population in a corpus and present researchers with only and all of the metaphors in the corpus. In this way, the burden of analysing each word token is lifted off the researchers’ shoulders, and the whole issue of sampling adequacy is gone, since the output would be the actual set of metaphorically used units in the data. However, machines do make mistakes – even part-of-speech tagging in- evitably incurs error, even though assigning parts of speech is far more straightforward than spotting metaphor uses. Consequently what metaphor retrieval software can do in reality is to provide a sample of the data that will hopefully be as comprehensive and precise as possible, containing most of the metaphors and few non-metaphors. Therefore the issue of sampling adequacy remains. The focus in this chapter is on sampling techniques because error is inherent in all of them. Just as all words in a corpus are not metaphorically used, all words in a sample are not metaphors either. Ideally, in a sample, only metaphors and all the metaphors in the source corpus will be included. In reality, though, error is introduced in samples, and so a larger proportion of non-metaphors may be included as a result of sampling error. The question is then: how reliable are samples obtained by different techniques? In this chapter a partial answer will be provided by comparing different sampling
- 38. Chapter 1.╇ An assessment of metaphor retrieval methods techniques with respect to how accurate they are in terms of offering good samples (i.e. with minimal error) for researchers. Unlike sampling techniques, with census techniques a question that arises is how reliable the identification is, that is, whether the steps were correctly followed by all coders involved. In census projects, usually more than one person is responsible for doing the coding, and in order for the coding to be reliable, there must be a high degree of agreement among coders (see Dorst and Kaal, this volume; Chapetón -Castro and Verdaguer-Clavera, this volume). As can be seen, the methodological issues surrounding sampling and census tech- niques are markedly different. Because census techniques involve questions related to coder agreement and not sample adequacy, they are not examined here. Pre-defining a search string or pool of string is perhaps the most frequently used sampling technique in corpus-based metaphor research. It consists in determining ahead of time one or more search strings based on particular research goals. A number of different researchers have applied this technique. For instance, Deignan (2005) chose words such as ‘hunt’ (8) and ‘warm’ (68), and expressions like ‘hot under the collar’ (21) and ‘in the running’ (28), and then searched for them in a large corpus (Bank of English). Her choice of each of these terms was motivated by theoretical con- siderations, including the link between linguistic and conceptual metaphor, the rela- tionship between metaphor and metonymy, patterning of linguistic metaphor, and frequency of metaphor compared to literal senses. Stefanowitsch (2006) also defined a set of words to focus on, independent of the corpus to be analysed, with the primary purpose of detailing the procedure known as Metaphor Pattern Analysis (MPA), which is aimed at finding metaphorical expres- sions in corpora. A metaphorical pattern was defined as “a multi-word expression from a given source domain (SD) into which one of more specific lexical items from a given target domain (TD) have been inserted” (66), and lexical items included content words (nouns, verbs, adjectives, adverbs). In Stefanowitsch (2006), MPA was used to identify metaphorical conceptualizations of emotions, such as anger, happiness and sadness. It involved the following steps. First, a target domain, for example, anger, was selected based on the previous literature on metaphors and emotions. Second, a lexical item was chosen to represent that domain, for instance ‘angry’. Third, a corpus (the British National Corpus [BNC]) was searched for that lexical item and a sample of up to 1,000 concordance lines was retrieved. Fourth, these lines were analysed by hand to determine whether each occurrence was a metaphor or not, and if so, what concep- tual metaphor motivated it; a metaphor was counted when the chosen lexical item and other lexical items nearby expressed a source-target domain mapping, for example, ‘angry’ and ‘boiling’ were considered to express the conceptual metaphor anger is hot fluid in a container. MPA and introspection were compared as to their ability to find metaphorical mappings. Results showed that MPA found well over 90% of the mappings identified by introspection, and that it also spotted mappings that were not arrived at introspectively.
- 39. Tony Berber Sardinha In terms of the techniques involved, MPA is a concordance-based procedure that relies on the choice of appropriate candidates: (“choose the lexical items wisely” [Stefanowitsch 2006: 66]), which in turn depends on the selection of particular do- mains that are of interest to a researcher. Hence, it is not suited for the analysis of a whole corpus, because it would mean having to analyse each word in the corpus, and therefore was not included for testing here. It must be said, though, that the idea of metaphorical patterns can be used to automate metaphor retrieval in corpora, accord- ing to Stefanowitsch (2006: 102–103): [W]e might even envision a lexical database containing a large number of lexical items and the metaphorical patterns they occur with (analogous to the FrameNet project at the UC Berkeley), which would allow easy retrieval of all metaphors as- sociated with a particular lexical item (or semantic field) and vice versa. In fact, such databases were used by the Metaphor Candidate Identifier (MCI), the metaphor detecting tool reviewed in Section 7. In both of these techniques, the set of strings to be searched for was defined in top-down mode, that is, the selection arose from theoretical or methodological con- cerns. Some researchers, though, have used bottom-up approaches as means for deter- mining which words to investigate. One such procedure is that developed by Cameron and Deignan (2003). These researchers point out problems with both small and large corpora. With small corpora, the main issue is a lack of generalizability: “the frequency and metaphorical use of a particular word form is inevitably influenced by the collection of data from a limited number of discourse events” (Cameron and Deignan 2003: 151). And with large corpora, there are two main problems, the first being a lack of information about the context, which may make at least some of the data difficult to interpret, and the sec- ond, the very issue being discussed in this chapter: [A] problem in searching large corpora is that patterns may be missed, because the researcher usually begins by searching for particular linguistic forms. If he or she has not identified a particular form as worthy of study, it may not emerge from the data during the analysis, and an important metaphorical use may be missed. This reflects a fundamental difficulty in researching linguistic metaphors through a corpus: We are trying to trace patterns of meaning but can only begin our analy- sis by looking at forms. (Cameron and Deignan 2003: 151) Their goal was to find “tuning devices” (‘just’, ‘sort of’, ‘actually’, etc.), or words and expressions normally referred to as hedges or vague language, and to look at how they were used in conjunction with metaphor. In order to find a good sample of tuning devices to search for in the large corpus, without predicting them by either introspec- tion or examples from the previous literature, they decided to read a small corpus (28,285 words) and note down any relevant search terms. These were then searched for in the large corpus (9 million words of the Bank of English), which contributed further
- 40. Chapter 1.╇ An assessment of metaphor retrieval methods information about frequency and patterning that illuminated several aspects of the use of tuning devices in English. Charteris-Black (2004) identified metaphors in a number of different corpora with the aim of developing Critical Metaphor Analysis, an approach that “supplements the cognitive semantic view by accounting for particular metaphor choices in different types ofdiscourseleadingtoadiscoursemodelofmetaphor”(243).LikeCameronandDeignan (2003), search terms were not chosen ahead of the actual corpus analysis, but, unlike them, he did not use a different corpus as source. Instead, search words were chosen through “extensive reading” (117, 178), that is, by reading a number of texts in the corpus and selecting any relevant terms that were then searched for across the same corpus. In her analysis of metaphors in business media discourse, Koller (2004: 48) de- fined search terms by combining a number of sources, including her previous knowledge of the field, reading some texts in the corpus, and looking up thesauri and glossaries as means to corroborate the relevance of the terms to the field of business. A further technique involves sampling based on word frequency. This can be put in place by simply choosing from the most frequent words in the corpus, or by choos- ing from words with marked frequency, or keywords. Keywords are words whose relative frequency is statistically higher in the corpus in comparison with a reference corpus, and software programmes such as WordSmith Tools and WMatrix calculate these (see Philip, this volume). Notice that in both these cases, sampling is initially carried out by machine, and subsequently by hand and eye, as the researchers pick some of the words out of the computer generated lists. Deignan and Semino (2010) used both methods as entry points into a corpus of speeches by former British Prime Minister Tony Blair. They examined the most frequent words in the corpus and then chose some that were “potentially of interest” (165), such as ‘back’, ‘forward’, and ‘cuts’. They then perused the keyword listing and noticed some overlap with the word fre- quency list, but they also spotted less frequent keywords such as ‘backward’, ‘delivered’, and ‘fight’ that seemed worth investigating for metaphor. Semantic tagging is another sampling technique used in the literature. It consists in using specialized software that adds a code (a tag) to each word in the corpus that identifies the semantic field to which each word belongs. After that, researchers select one or more particular semantic field(s) and retrieve all occurrences of words tagged for that semantic field. One advantage of this technique over string based sampling is that with semantic tagging a pool of different words related by a common sense field can be automatically identified. This technique has been used by Deignan and Semino (2010) in the analysis of the Blair speech corpus, and it was put in place through WMatrix, which incorporates USAS, a semantic tagger developed by Rayson (2008). Their analysis revealed a number of semantic fields that seemed worthy of closer inves- tigation, like “Movement, location, travel and transport” (176). This field was immedi- ately noticeable given its frequency and incongruence with the general topic of the speeches, which is politics and not movement. Several semantically related words were
- 41. Tony Berber Sardinha part of this field, including ‘route’, ‘journey’, and ‘gone’, which were then searched for and analysed for metaphor. 2. Metaphor retrieval procedures examined in this chapter There has been growing interest in using corpora in metaphor research in recent years, and as a result a number of tools and techniques have been proposed and used for metaphor identification. However, very little is known about their ability to retrieve all and only metaphors from corpora. The aim of this chapter is to report on a quantitative assessment of methods for metaphor retrieval. Out of the many different techniques and instruments reported in the literature on metaphor, corpus linguistics and Natural Language Processing, three procedures and three computer tools have been selected for assessment. The procedures are: (i) reading parts of a larger corpus in order to find candidates that are then sought for in the whole corpus through a concordancer; (ii) searching for metaphors using different search terms, such as single words, collocates and lexical bundles; and (iii) looking for metaphor clusters. The second item requires a concor- dancer, which is a computer tool, but it was classified as a procedure because the point of the section is not to discuss concordancing per se, but the effect of different search term types (used not only with concordancers, such as grep1, but also with tools) on metaphor retrieval. Just as with the first procedure, a computer tool of some sort is as- sumed, but the tool itself is not the focus. The three computer tools are: (i) finding metaphor candidates through keywords, or words whose frequencies are statistically higher in a corpus than in a comparable reference corpus; (ii) finding metaphor candidates through the Metaphor Candidate Identifier, an online tool that looks for metaphorically used words by matching single words and patterns drawn from hand-coded training data; and (iii) finding metaphor candidates by computing semantic relatedness, more specifically, by computing a mea- sure of the difference in meaning between neighbouring words. These tools were cho- sen because they are free and publicly available2. Another tool that has been used in the literature for choosing metaphor candidates is WMatrix (Rayson 2008), but it requires a paid subscription (even though a free password for research purposes can be obtained for a limited period of time), and that is why it was not included in this assessment. 1. Grep is a command line utility that enables users to search text material. It is widely avail- able on Unix, Linux, and Mac systems. 2. WordSmith Tools 3.0 is free from Mike Scott’s website at www.lexically.net; AntConc is freely available on Laurence Anthony’s webside at http://www.antlab.sci.waseda.ac.jp/software. html; the MCI is a free online tool at www2.lael.pucsp.br and www.corpuslg.org/tools; and se- mantic relatedness is implemented in the free Perl package WordNet::Similarity, available at http://wn-similarity.sourceforge.net.
- 42. Chapter 1.╇ An assessment of metaphor retrieval methods Other tools such as Cormet (Mason 2004) and TroFi (Birke 2005), which are reported in the Natural Language Processing literature, are not available for installation. The order of presentation of procedures is from most to least conventional, with partial corpus reading as arguably the most traditional technique, and clustering as the most experimental. For computer tools, the order of presentation is from least to most demanding of computer and programming skills. Keywords is the least demanding because it is implemented in relatively easy to use, point-and-click programmes with graphic interfaces (such as WordSmith Tools and AntConc). The MCI is much simpler to get started with than either WordSmith Tools or AntConc, but it is more challenging because it requires some understanding of how it operates “under the hood” in order for researchers to make sense of its output. And WordNet::Similarity is undoubtedly the most difficult tool to install and operate, as it has no graphic interface and requires programming skills and familiarity with command line interfaces. Most methods tested here are bottom-up, because they are meant to mine corpora for metaphor candidates, rather than seeking predefined candidates. The exception comes under our assessment of search terms for concordancing, which presupposes that a set of candidates has already been determined, and therefore may be a case of top-down methodology. As regards the corpus-driven/corpus-based dichotomy (Tognini-Bonelli 2001), these methods can be either, because researchers may use them to test particular theories of metaphor, in which case they may be classed as corpus-based, or they may be used to explore how metaphors present themselves lexi- cally in corpora, in which case they may be seen as corpus-driven. It must be stressed that this assessment is not a final evaluation, since performance of any one of these methods may be altered by different test corpora. The data used here were: – Conference Call Corpus: A corpus of conference calls, or meetings held over the phone, between investment banks, shareholders, and the press, in Brazil, in Portuguese. It contains 14 different conference calls, 82,881 tokens, and was fully annotated for metaphor by hand. It is a slightly modified version of the corpus used in Berber Sardinha (2008). It was selected because it was the only metaphor annotated corpus available at the time of writing. Recently, the English MIPVU corpus has been made available, containing excerpts from the BNC Baby that were fully coded for metaphor by hand. The Conference Call corpus was used to exam- ine the following procedures: reading portions of the corpus, search term choice, clustering, and keywords. – MCI test corpus. An English corpus containing five texts and 1,313 tokens, all hand-coded for metaphor, used to test the Metaphor Candidate Identifier. More details in Table 9. – BNC Concordance. A set of 7,524 concordance lines drawn from the BNC and hand-coded for metaphor used to test the semantic relatedness procedure.
- 43. Tony Berber Sardinha Metaphors were identified in the data broadly following the Metaphor Identification Procedure (MIP) (Pragglejaz Group 2007): 1. The whole corpus was read to gather an understanding of the topics covered in the texts. 2. For each word in the text, both its contextual and basic meanings were established. 3. If the word had a more basic current-contemporary meaning, a decision was made as to whether the basic meaning contrasted with and contributed to the meaning of the word in the text. 4. If it did, then the word was marked as metaphorical; if not, then it was not. The most important differences between the identification procedure applied to the data in this study and that proposed by MIP were: – MIP recommends that the texts be segmented in terms of lexical units, and that decisions be made in each case as to whether a word should be analysed on its own or as part of a larger lexical unit. In this study, segmentation was at word level, and so metaphor coding was done word by word. A word was defined as a string of at least one letter surrounded by regular delimiters such as blank spaces, line end- ings, and punctuation marks. – MIP suggests a corpus-based dictionary and an etymological dictionary be used to aid “researchers’ intuitions about any difficult cases” (17). In the case study pre- sented in the Pragglejaz Group article (2007), out of 28 lexical units, four were looked up in dictionaries (representing 14% of the total units, or 1 every 7 units), which can be rather time-consuming with larger datasets. In the data analysis re- ported here, no dictionary was consulted, and both basic and contextual meanings were determined by the researcher using his own background knowledge. The following sections focus on the techniques and tools examined. 3. Reading portions of the corpus for candidates As has been said, one technique commonly employed by metaphor researchers is to read a sample of the corpus texts, noting down any metaphors encountered and then searching the corpus for these. There are a number of questions raised by this meth- od, motivated by the concern that there might be a substantial number of metaphors left undetected in the corpus because they did not occur in the sample that was read. The main questions seem to be then of whether one can retrieve the totality of meta- phors from the corpus by reading just a portion of it, and if not, what is the propor- tion of metaphors retrieved, and whether this proportion rises as the amount of text read increases. The corpus used was the conference call corpus. In order to put this technique to the test, different size samples were experimented with. For sample size 1, the
- 44. Chapter 1.╇ An assessment of metaphor retrieval methods texts in the sample were each an individual text (1, 2, 3, etc. up to 14). From then on, each sample size was made up of all possible text combinations for that particu- lar sample size. Therefore, for sample size 2, the texts were pairs of individual texts (1 and 2, 2 and 3, 3 and 4, etc. up to 13 and 14). For sample size 3, the texts were triplets (1, 2, 3; 2, 3, 4; 3, 4, 5; etc.). And so on, until sample size 13, in which case the texts in the sample were a group of 13 texts (1 through 13, 2 through 14, 3 through 14 plus text 1, 4 through 14 plus texts 1 and 2, etc.). These combinations were used in order to prevent bias, which might occur if particular texts were read that had far more metaphor cases than the others. In this way, all texts are consid- ered for reading. For each of these situations, recall was computed. In this investigation, recall is the number of metaphor types in the corpus retrieved by reading any one sample size. It was computed by dividing the number of metaphor types found in a text portion by the total number of metaphor types found in the corpus (multiplied by 100). By meta- phor type is meant a unique instance of a metaphorically used word; subsequent appearances of the same metaphorically used word were not computed. The higher the recall, the more metaphors were retrieved by reading a particular portion of the cor- pus. Afterward, the average recall was calculated for the whole sample size. To illustrate, Table 1 shows the figures for text portion 1. Table 2 shows, for each size sample, the average recall, recall increase and the ratio of recall to sample size (as a percentage). This ratio is a basic measure of effectiveness: the higher the number, the more effective the sample is, in the sense that more Table 1.╇ Recall for text 1 Texts in sample Metaphors retrieved (A) Metaphors in corpus (B) Recall (A/B * 100) â•⁄ 1 123 414 29.7% â•⁄ 2 â•⁄ 95 414 22.9% â•⁄ 3 106 414 25.6% â•⁄ 4 134 414 32.4% â•⁄ 5 â•⁄ 74 414 17.9% â•⁄ 6 125 414 30.2% â•⁄ 7 â•⁄ 95 414 22.9% â•⁄ 8 106 414 25.6% â•⁄ 9 â•⁄ 43 414 10.4% 10 105 414 25.4% 11 132 414 31.9% 12 109 414 26.3% 13 â•⁄ 64 414 15.5% 14 105 414 25.4% Average recall for sample size 1 24.4%
- 45. Tony Berber Sardinha Table 2.╇ Recall for reading portions of corpus Sample size Average recall Average increase Recall/sample size â•⁄ 1 (7%) 24.4% – 3.4 â•⁄ 2 (14%) 37.8% 13.3% 2.6 â•⁄ 3 (21%) 47.4% â•⁄ 9.7% 2.2 â•⁄ 4 (29%) 55.2% â•⁄ 7.7% 1.9 â•⁄ 5 (36%) 61.6% â•⁄ 6.4% 1.7 â•⁄ 6 (43%) 67.3% â•⁄ 5.7% 1.6 â•⁄ 7 (50%) 72.4% â•⁄ 5.1% 1.4 â•⁄ 8 (57%) 77.2% â•⁄ 4.8% 1.4 â•⁄ 9 (64%) 81.7% â•⁄ 4.5% 1.3 10 (71%) 85.8% â•⁄ 4.1% 1.2 11 (79%) 89.7% â•⁄ 3.9% 1.1 12 (86%) 93.3% â•⁄ 3.7% 1.1 13 (93%) 96.7% â•⁄ 3.4% 1.0 14 (100%) 100% 0% 1.0 metaphors will have been retrieved with less reading input. On the other hand, if the ratio is low (the minimum is 1), then more effort will have been spent by going through a large reading sample to find metaphors. These figures show that recall increases as more texts are added to the reading sample, but the increase is not steady: the effect of adding more texts to a smaller sample is more striking than adding to a larger sample. If recall increased at a steady rate, it would increase by 7.1% with each portion (since 100/14 = 7.1). The point of diminishing returns for recall is where the expected average increase drops below 7.1%, which is at sample size 5. This is also the point at which more than half of all the metaphors will have been found. This suggests that a corpus portion consisting of four texts (or 29% of the whole corpus) would be the optimal sample size, beyond which the rate of finding new metaphors would perhaps not justify the effort involved in reading more texts. The effectiveness of the technique, as measured by the ratio recall/ sample size decreases as samples get larger. Effectiveness seems to have been undercut after sample size 3, or 21% of the whole corpus, since up to that point the ratio of metaphor retrieval was over 2, meaning twice as many metaphors were found than text material was read. However, these figures show that there are new metaphors in each text, no matter how big a reading sample is. Even a reading sample consisting of all texts but one (13) does not yield all of the metaphors in the corpus. On the whole, these results indicate that reading a few texts of the corpus for can- didates is an effective sampling technique, which enables researchers to retrieve a large portion of the metaphors present across the whole corpus. Reading just 7% (1 text) of
- 46. Chapter 1.╇ An assessment of metaphor retrieval methods the corpus retrieves about a quarter of the metaphors. The majority of the metaphors are found by reading 29% (4 texts) of the corpus. Again, this conclusion is based on the rationale that researchers will not read an entire corpus in the first place, and that they give some consideration to the amount of text that they will read. The practical advice drawn from these results would then be that researchers should strive to read all of the texts in their corpus, but if that is not possible (as is often the case with electronic corpora), then they should read at least about 30% of them. 4. Concordancing: Search term choice Techniques such as the previous one generally presuppose researchers will depend on a concordancer in order to search for the candidates noted during reading. But there are different kinds of search terms that can be used, such as single words, multiple word sequences, and word plus a collocate, to mention a few. The question then arises as to whether different kinds of search words are more reliable than others in retriev- ing metaphors. In this section, answers to this question will be pursued, but this ex- periment rests on the assumption that researchers would have an attested set of search terms, obtained, for instance, by reading portions of the corpus. In other words, the results presented here do not apply to situations in which researchers make up a list of search terms by guesswork, intuition, or similar methods. Different search term types have distinct advantages and disadvantages. Single words are an obviously easy search term to formulate, but they can be ambiguous and therefore retrieve instances of non-metaphor along with metaphors (‘waste’ would pick up both ‘waste time’, which is metaphorical, and ‘waste money’, which is not). Word sequences, on the other hand, can be trusted to retrieve more unambiguous cases of metaphor (‘waste time’, ‘waste efforts’, ‘waste our lives’, all of which are meta- phorical uses of ‘waste’), but they can be difficult to formulate, because the exact word sequences that appear in the corpus may be hard to predict. Node plus collocate search- ing may be seen as having an advantage over single word searching (‘waste’ followed by ‘time’ at two words to the right will probably not retrieve any cases of non-meta- phor), but it also has the major drawback of predicting collocates. Given the problems associated with formulating both bundles and collocations, then it is likely that most researchers would prefer to search their corpora for single words anyway, at least at first, and then probably move on to bundles or collocations, when they have a better idea of the linguistic metaphors in the corpus. But the issue still remains of how reli- able single words are as search terms. Less reliable search terms mean extra work for researchers, who will have to read and judge more cases, a situation which may be critical when dealing with large corpora yielding thousands of citations of particular search terms.
- 47. Tony Berber Sardinha The main variable in this investigation is search term type, which is one of the fol- lowing: single word, bundle, or collocation. For bundles, three subtypes were identi- fied, depending on how many words are in the bundle: two words, three words, and four words. For collocations, the following subtypes were determined, depending on the position of the node: node + 5L (five words to the left of the node), node + 4L (four words to the left of the node), and so on up to node + 5R (five words to the right of the node). The position in which the metaphorically used word occurred did not matter. For bundles, the metaphorically used word(s) could be any of the words com- prising the bundle. For collocation, the metaphorically used word(s) could be either the node or the collocate. The question addressed in this section is how precise each of these search term types is when used to retrieve metaphors from the corpus. Precision was calculated by dividing the number of metaphors retrieved by the total number of instances retrieved (multiplied by 100). For instance, if a word retrieved 100 citations from the corpus, and 50 of those were metaphors, then precision would be 50% (50/100 * 100). This investigation was carried out as follows. First, all instances of metaphor from one text in the corpus were retrieved and turned into single words, bundles (formed by sequences of two, three or four words) and collocations (node plus collocates at positions five, four, three, two and one words to the right and left of the node). These were not mutually exclusive: the same single word was part of a bundle and of a collocation, and collocations of the kind node + 1L and node + 1R were both two-word bundles. The deci- sion to extract the search terms from the corpus itself and not to make up the search terms was taken because the intention was not to test our intuition but rather to test the retrieving power of real search terms. If we had made up a list of search terms, some of them might not match any metaphors in the corpus, thus interfering with the results. By drawing the search terms from the corpus, we ensured a level playing field for all search terms, making sure all of them could achieve 100% precision. Secondly, all of these in- stanceswerematchedagainstalloftheirrespectivemetaphorunits(singlewords,bundles and collocations) in the corpus; each time a match was found, a hit was scored. If more than one metaphorically used word occurred in a bundle or collocation, then hits were scored accordingly (a bundle with two metaphorically words received two hits, etc.). Finally, all hits were computed and precision was calculated for each search term type. Table 3 shows the results for precision for each search word type and subtype. The figures show the most precise search units are fixed word sequences, such as bundles and collocations formed by neighbouring collocates, which is not surprising, since fixed patterns normally express a specific meaning. They also show there was no dif- ference among the subtypes of bundles, all of which were 100% precise, unlike colloca- tions, which varied from 97.1% to 100%. The least precise search term type was the single word, as predicted, at 73.2%. This study suggested a number of interesting findings. Firstly, single words were surprisingly precise, yielding only about one quarter of false positives (non-metaphors
- 48. Chapter 1.╇ An assessment of metaphor retrieval methods Table 3.╇ Precision for different search terms Search term type Precision single word 73.2% 2-word bundle 100% 3-word bundle 100% 4-word bundle 100% node + 5L 97.9% node + 4L 97.3% node + 3L 97.2% node + 2L 97.7% node + 1L 100% node + 1R 100% node + 2R 98.6% node + 3R 98.0% node + 4R 97.0% node + 5R 97.1% instead of metaphors). This is probably due to the fact that the Conference Call Corpus is highly controlled for genre (conference calls) and topic (investments), and so from a probabilistic standpoint, metaphorically used words are generally used to express that one sense only, in a particular phraseology (Berber Sardinha 2008; Philip, this volume). With genre, register and/or topic diversified corpora, this figure would prob- ably be lower, as single words take on different meanings in different contexts, express- ing a metaphorical use in one context and a non-metaphorical use in another. The practical advice that emerges from this is that starting with single words is probably a good working strategy for researchers. Later on, as they become acquainted with the phraseology of metaphors in the corpus, they may formulate more precise searches with either bundles or collocations. Secondly, bundle subtypes were equally precise, at 100%, which in practical terms means that with a corpus like this researchers do not need to worry about predicting long fixed word sequences to make precise searches, as a simple two-word sequence will retrieve metaphors only. This again may be a conse- quence of the tightly controlled vocabulary used in the corpus, and this is expected to change somewhat with diversified corpora. Overall, these results corroborate Deignan’s (2005) findings that indicated that metaphorically used language tends to exhibit a tight phraseology, whereas non-metaphoric language is more freely combin- ing. Finally, there was not much difference among collocate positions, all of which scored above 97%. One might have expected collocates to become less precise the fur- ther away they were from the node, but this was not corroborated here. The practical suggestion arising here is that with corpora like this, researchers should not restrict
- 49. Tony Berber Sardinha searches to patterns formed with near collocates, since metaphor phraseology often stretches a long way away from the node. 5. Clustering Clustering is a property of metaphor distribution in texts, according to which meta- phors are distributed unevenly within texts, in such a way that many form groups of metaphorical units occurring near each other. A number of studies have shown clus- tering as a feature of metaphor distribution. According to Cameron (2008), one of the reasons for clustering is topical, since developing a topic in discourse sometimes requires users to repeat metaphors that are being employed to express a particular topic. Another reason for clustering in speech has to do with the tendency for speakers to repeat, reformulate and pick up on each others’ points, thus re-using groups of words within a short period of time. To my knowledge, clustering has not been em- ployed so far as a technique for retrieving metaphors. However, it appears as though it could be, perhaps as an awareness raising tool for researchers to apply during meta- phor coding. If researchers become aware of clustering, once they spot one case of metaphor in a text, for instance, they may decide to look more closely for other in- stances of metaphor nearby. The aim here is to assess clustering from a quantitative standpoint. A metaphor cluster is defined here simply as an occurrence of two meta- phors within a variable stretch of text. In order to explore clustering quantitatively, the starting point is to assume that there is a textual window around a metaphor where one can find other instances of metaphor, thus forming a cluster. The problem, of course, lies in determining the extent of that window. In this investigation we then look at the issue of finding an optimal window that would allow us to retrieve as many metaphors as possible from the corpus. The first step was to determine a figure that represented the average distance be- tween metaphors in the corpus. This average distance was calculated by dividing the number of word tokens (82,881) by the number of metaphor tokens (3,800), yielding 21.8, meaning that metaphors are on average about 22 words away from each other. This figure represents the expected distance between metaphors if they were distrib- uted evenly across the corpus. Therefore, a criterion for clustering was set according to which the maximum window size would not exceed the average distance between metaphors across the corpus. Next, the following window sizes were tested: 5, 10, 15 and 20 words, and recall was calculated for each window size. Recall was computed for each text by dividing the number of metaphor tokens occurring within the window by the total number of met- aphor tokens in the text multiplied by 100. Finally, mean recall for each window size was computed by averaging out the individual recall figures for each text. To illustrate, Table 4 shows the results for window size = 5.
- 50. Chapter 1.╇ An assessment of metaphor retrieval methods Table 4.╇ Clustering retrieval for window size = 5 Text Metaphor tokens within window Metaphor tokens in text Retrieval â•⁄ 1 106 395.0 26.84% â•⁄ 2 â•⁄ 68 254.0 26.77% â•⁄ 3 â•⁄ 65 274.0 23.72% â•⁄ 4 â•⁄ 77 383.0 20.10% â•⁄ 5 â•⁄ 27 210.0 12.86% â•⁄ 6 â•⁄ 85 357.0 23.81% â•⁄ 7 â•⁄ 59 285.0 20.70% â•⁄ 8 â•⁄ 62 256.0 24.22% â•⁄ 9 â•⁄ 12 â•⁄ 75.0 16.00% 10 â•⁄ 50 293.0 17.06% 11 â•⁄ 63 289.0 21.80% 12 â•⁄ 62 308.0 20.13% 13 â•⁄ 28 133.0 21.05% 14 â•⁄ 66 288.0 22.92% Average 21.28% Results indicate that with a window of size 5, an average 21% of the metaphors fall within a cluster. This was repeated with the other window sizes, and the results appear in Table 5. Table 5 presents a couple of interesting findings. The first is that, as would be ex- pected, recall rises as the window size expands. Wider windows pick up more meta- phors, whereas narrower windows miss out on more metaphors. The second is that none of the window sizes returned recall rates near 100%; even a generous window size such as 20, which is near the average distribution (22), recall is only about 2/3 of all metaphors. With a window size this wide, there is not much point in looking for meta- phors within clusters, as windows would be so large that there would be very few gaps between them, thus essentially forcing researchers to read the whole corpus. The practical advice that could be gleaned from this would be to stick to narrow window sizes such as 5 and 10, which, in corpora similar to ours, would help retrieve up to 40% of the metaphors. In addition, window sizes such as these normally fit Table 5.╇ Clustering recall Window size Average recall â•⁄ 5 21.28% 10 41.14% 15 55.83% 20 65.25%
- 51. Tony Berber Sardinha within the length of most concordances. This may enable researchers to spot meta- phors in the vicinity of node words on concordance lines. 6. WordSmith Tools keywords Keywords are words whose frequencies are statistically higher in a corpus in compari- son to a reference corpus. Keywords is also the name of an application that is part of the corpus analysis package WordSmith Tools (Scott 1997) that extracts keywords au- tomatically. Keywords can be extracted by a number of different tools besides WordSmith Tools, including AntConc, WMatrix, and the CEPRIL Keyword Tool (www2.lael.pucsp.br/corpora).Keywordshavebeenusedinmetaphorresearch(Berber Sardinha 2009, Partington 2006, Philip 2008, this volume) for the general purpose of selecting candidates for close inspection. As with the other techniques, there are questions surrounding the reliability of key- words as a means of metaphor retrieval, not least because little is known about the rela- tionship between metaphor and marked lexical frequency, the guiding principle behind keywords. The specific goals here are to find out what proportion of metaphors can be retrieved through keyword extraction, and how precise this method is. These seem im- portant issues surrounding keywords, even if metaphor researchers employ keywords for purposes other than retrieving the majority of metaphors from their corpora. To investigate this issue, the following procedures were followed. First, the key- words were extracted in WordSmith Tools version 3, by comparing the word frequency of the corpus to that of the Banco do Português (version 1), a large register-diversified corpus of Brazilian Portuguese comprising over 230 million words of spoken and writ- ten language. The settings for keywords were as follows: max keywords 500,000, max p. value .05, keywords procedure log-likelihood. These settings enabled all keywords to be extracted, and not just the default 500. A total of 2,532 keywords were produced, including both positive and negative ones. Secondly, all metaphorically used words in the corpus were listed. Thirdly, positive keywords were separated from negative words. Positive keywords are the default keywords, that is, their frequency is marked in the main corpus; negative words are the reverse of these, in the sense that their frequencies are statistically higher in the reference corpus. Negative keywords, if available in a particular corpus, appear in red at the bottom of the screen in WordSmith Tools ver- sion 3. The keyword lists were split into samples that started with the top 100 keywords and were incremented by 100 keywords; samples were then 100, 200, 300, and so on up to 2,044 for the positive keywords and up to 488 for the negative ones. Finally, meta- phorically used words were then matched against the keywords, the number of exact matches was recorded, and performance metrics were computed (precision and recall). Precision was calculated by dividing the total matches for a particular sample by the size of that sample; recall was computed by dividing the total matches for a particular
- 52. Chapter 1.╇ An assessment of metaphor retrieval methods sample by the total metaphorically used words in the corpus (414). Results for positive keywords appear in Table 6. As can be seen in Table 6, the best precision score was for the 600 keyword sample, which amounts to 29% of the keyword output. The best recall mark was for the whole list, with 42% of the total metaphors retrieved. Results for negative keywords appear in Table 7. Table 6.╇ Precision and recall for positive keywords Sample Matches Precision Recall 100 â•⁄â•⁄ 7 â•⁄ 7% â•⁄ 2% 200 â•⁄ 16 â•⁄ 8% â•⁄ 4% 300 â•⁄ 26 â•⁄ 9% â•⁄ 6% 400 â•⁄ 38 10% â•⁄ 9% 500 (default) â•⁄ 46 â•⁄ 9% 11% 600 â•⁄ 64 11% 15% 700 â•⁄ 70 10% 17% 800 â•⁄ 75 â•⁄ 9% 18% 900 â•⁄ 85 â•⁄ 9% 21% 1000 â•⁄ 88 â•⁄ 9% 21% 1100 â•⁄ 92 â•⁄ 8% 22% 1200 103 â•⁄ 9% 25% 1300 110 â•⁄ 8% 27% 1400 119 â•⁄ 9% 29% 1500 125 â•⁄ 8% 30% 1600 129 â•⁄ 8% 31% 1700 142 â•⁄ 8% 34% 1800 147 â•⁄ 8% 36% 1900 160 â•⁄ 8% 39% 2000 170 â•⁄ 9% 41% Whole list (2044) 172 â•⁄ 8% 42% Table 7.╇ Precision and recall for negative keywords Sample Matches Precision Recall 100 â•⁄ 8 8% 2% 200 13 7% 3% 300 19 6% 5% 400 23 6% 6% 500 24 5% 6% Whole list (588) 24 4% 6%
- 53. Tony Berber Sardinha Table 8.╇ Overall recall by keywords Total metaphors retrieved Whole list Portion of list Corresponding to highest precision Default 500 keywords By positive keywords 172 42% 64 15% 46 11% By negative keywords â•⁄ 24 â•⁄ 6% â•⁄ 8 â•⁄ 2% – – Total 196 47% 72 17% 46 11% Results indicate that the best precision score is for the top 100 negative keywords, at 8%, and the best recall is for the whole list, at 6%. Topmost negative keywords are those bearing the most marked frequencies, meaning they are the rarest words in the corpus. This suggests some metaphorically used words are unusual in the corpus. Table 8 shows the overall recall achieved by the keywords procedure. The key- words procedure retrieved less than half of the metaphors, if we include both positive and negative keywords. About 53% of the metaphorically used words were not key- words at all, that is, their frequency was statistically similar in the comparison corpus. This suggests metaphorically used words are neither particularly frequent nor rare, otherwise they would have been keywords, positive or negative. The highest recall was reached with the whole list of keywords (including positive and negative), but it would be unusual for researchers to consider the full list of keywords in their analysis, not least because the list extracted here was obtained with the least stringent criteria pos- sible for keyword extraction in WordSmith Tools. Normally, researchers use the de- fault criteria, which produce a 500-keyword list, and for that list, recall was only 11%. Recall for the point on the list where precision was highest was slightly better at 15%, but in practice such a point is hard if not impossible to determine, given that research- ers will not know which keywords are metaphorically used before running Keywords. In conclusion, keywords do not seem to be a particularly effective retrieval tech- nique, at least with the data used here. That does not mean, however, that selecting words with keyword status is not relevant for metaphor research. The fact that words have a marked frequency may be important in a number of ways, as pointed out in the literature, as keywords may signal important textual properties such as aboutness, style, and textual salience, among other attributes, all of which may be relevant to particular metaphor research projects. These findings pertain to metaphor retrieval only, and not to the relevance of keywords per se. One further point that we must re- mind ourselves of is that Keywords was not designed to retrieve metaphors, and therefore cannot be criticized for not doing particularly well at a job it was not intended to do.
- 54. Chapter 1.╇ An assessment of metaphor retrieval methods 7. Metaphor Candidate Identifier The Metaphor Candidate Identifier (MCI) is a computer programme developed by Berber Sardinha (2007), which aims specifically at retrieving metaphorically used words from corpora. It works by matching each word in a corpus, its patterns and its part of speech to a set of five metaphor databases, and then calculating the average probability of that word being metaphorically used. These databases were compiled from hand-coded concordances (the “training data”), where each node word was judged as metaphorical or not based on principles similar to those proposed in MIP (Pragglejaz Group 2007). Each database holds specific information about single words, 3-word bundles preceding and following each word, the immediate collocates to the left and right of the word (called ‘framework’), and the part of speech assigned to that word by a tagger (Tree-Tagger). The output of the programme is an ordered list of candidate words, sorted by its probability of metaphorical use. The MCI is an online tool that is available in two versions, one for analysing Portuguese corpora and an- other for English corpora; both versions can be accessed for free on the web at the CEPRIL (Centre for Research, Resources and Information on Language, Sao Paulo Catholic University) website at www2.lael.pucsp.br. To illustrate how the programme goes about identifying metaphor candidates, let’s take the following sentence from the “Ozone” text in Cameron (2003: 168), where ‘made’ is a metaphorically used word: (1) But not all the energy made by the Sun is safe. The MCI would check each word in that sentence, and for ‘made’, processing would be carried out in the following way: – made: – Check single word database, which stores each word that was found to be metaphorically used in the training data, together with its probability of met- aphor use. ‘Made’ is found on the database, with a probability of .6000. This value is grabbed and stored in the programme’s memory. If this word were not on the database, this would mean it was never found in the previously hand- coded texts to be metaphorically used, either because it appeared in the train- ing data in its basic sense or it never appeared at all in the texts. Either way, the programme would store the value of .00001 for it. – Extract 3-word bundle preceding it: “all the energy” – Check that bundle in the ‘left bundle database’, which stores all 3-word bun- dles that preceded each metaphorically used word in the training data, to- gether with its probability of metaphor use. The bundle is not found there, and so the programme stores the value of .00001 for it. – Extract 3-word bundle following it: ‘by the Sun’
- 55. Another Random Scribd Document with Unrelated Content
- 56. with the cipher by its side, it becomes ten. She was the wife of John Flaxman, the Sculptor. Down with the Austrian woman, shouted the infuriated mob of Paris, supposing that they saw before them the ill-fated Marie Antoinette. An officer corrected their mistake, and the lady, just rescued from the most terrible of deaths,—that of being torn to pieces by savages,—said to him, Why undeceive them? You might have spared them a greater crime. She was the same, who, when asked her name and rank before the revolutionary tribunal, replied, with dignity, I am Elizabeth of France, the aunt of your king. She was compelled to witness the execution of twenty-four of her fellow-prisoners, and then met her own death without a complaint. Among savage nations what could be more terrific than a volcano? And when, in addition to its natural mysteries, a cunning priesthood has invested it with the attributes of a malignant and revengeful deity, who but an enlightened and civilized person would dare to approach it? It was tabooed, and whoever insulted it, would be destroyed by its shower of liquid fire. It is hard to shake off the prejudices and superstitions of a life- time. Yet Kapiolani, a woman of Hawaii, who had already done much to raise the character of her countrymen, set the heathen priests at defiance, declared the volcano to be the work of a merciful God, and boldly descended some distance into its crater. There she composedly praised the Lord in the midst of one of His wonderful
- 57. works. The effect of her faith upon the minds of her countrymen was wonderful. In all that is known of Assyria, the most ancient empire of the earth, every extant fragment, moral or material, bears evidence of a sex to which that land of wonders owes the immortality of its grandeur. The name of Semiramis has preserved (what Sardanapalus could not destroy, nor Cyrus bury under the ruins of Babylon,) the memory of the greatest combination of wealth, power, art, and magnificence, which the world had till then witnessed, or has since conceived. For the greatest capitals of the most powerful and refined of modern states, supposed to have reached the acme of civilization, have but one epithet to mark their supereminence; and Rome and London (in boast, or in reproach,) have each been called the Babylon of their own proudest times. Babylon, with its hundred gates and towers, was founded by a woman of low origin and destitute youth, who attained to supreme power by her genius alone; and though all that has been ascribed to her may not be strictly true, though Diodorous Siculus in his enthusiasm may have exaggerated, and Ctesias may have too vividly colored his brilliant delineations of her greatness, yet that such a woman lived and reigned in Assyria, that she founded its capital, and influenced her age by her works and her talents, that she built cities, raised aqueducts, constructed roads, commanded great armies in person, and, both as conqueror and legislator, was among the earliest agents of Asiatic civilization, there remains no room for historic doubt. Her passage over the Indus, her conquests on its shores, the brilliant triumphs she obtained abroad, the astute wisdom with
- 58. which she met conspiracy at home, and the bold confidence she expressed in the decisions of posterity, are stubborn facts. These obtained for her the sympathy of the greatest character and conqueror of a nearer antiquity; but Alexander, taking Semiramis for his model, vainly tried to restore her gorgeous city, on her own plans, and with her own views. Posterity has nobly ratified the appeal of Semiramis to its verdict. At the end of three thousand years, her life and character have been taken as the inspiration of its genius, and the spell of its attraction. Semiramis, however, has paid the penalty of her sex's superiority, and has been the mark of calumnious pedantry through succeeding ages. *Since the above was in type, Mlle. Nilsson has several times sung Way down upon the Swanee River at her concerts.
- 59. *** END OF THE PROJECT GUTENBERG EBOOK OUR GIRLS *** Updated editions will replace the previous one—the old editions will be renamed. Creating the works from print editions not protected by U.S. copyright law means that no one owns a United States copyright in these works, so the Foundation (and you!) can copy and distribute it in the United States without permission and without paying copyright royalties. Special rules, set forth in the General Terms of Use part of this license, apply to copying and distributing Project Gutenberg™ electronic works to protect the PROJECT GUTENBERG™ concept and trademark. Project Gutenberg is a registered trademark, and may not be used if you charge for an eBook, except by following the terms of the trademark license, including paying royalties for use of the Project Gutenberg trademark. If you do not charge anything for copies of this eBook, complying with the trademark license is very easy. You may use this eBook for nearly any purpose such as creation of derivative works, reports, performances and research. Project Gutenberg eBooks may be modified and printed and given away—you may do practically ANYTHING in the United States with eBooks not protected by U.S. copyright law. Redistribution is subject to the trademark license, especially commercial redistribution. START: FULL LICENSE
- 60. THE FULL PROJECT GUTENBERG LICENSE
- 61. PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK To protect the Project Gutenberg™ mission of promoting the free distribution of electronic works, by using or distributing this work (or any other work associated in any way with the phrase “Project Gutenberg”), you agree to comply with all the terms of the Full Project Gutenberg™ License available with this file or online at www.gutenberg.org/license. Section 1. General Terms of Use and Redistributing Project Gutenberg™ electronic works 1.A. By reading or using any part of this Project Gutenberg™ electronic work, you indicate that you have read, understand, agree to and accept all the terms of this license and intellectual property (trademark/copyright) agreement. If you do not agree to abide by all the terms of this agreement, you must cease using and return or destroy all copies of Project Gutenberg™ electronic works in your possession. If you paid a fee for obtaining a copy of or access to a Project Gutenberg™ electronic work and you do not agree to be bound by the terms of this agreement, you may obtain a refund from the person or entity to whom you paid the fee as set forth in paragraph 1.E.8. 1.B. “Project Gutenberg” is a registered trademark. It may only be used on or associated in any way with an electronic work by people who agree to be bound by the terms of this agreement. There are a few things that you can do with most Project Gutenberg™ electronic works even without complying with the full terms of this agreement. See paragraph 1.C below. There are a lot of things you can do with Project Gutenberg™ electronic works if you follow the terms of this agreement and help preserve free future access to Project Gutenberg™ electronic works. See paragraph 1.E below.
- 62. 1.C. The Project Gutenberg Literary Archive Foundation (“the Foundation” or PGLAF), owns a compilation copyright in the collection of Project Gutenberg™ electronic works. Nearly all the individual works in the collection are in the public domain in the United States. If an individual work is unprotected by copyright law in the United States and you are located in the United States, we do not claim a right to prevent you from copying, distributing, performing, displaying or creating derivative works based on the work as long as all references to Project Gutenberg are removed. Of course, we hope that you will support the Project Gutenberg™ mission of promoting free access to electronic works by freely sharing Project Gutenberg™ works in compliance with the terms of this agreement for keeping the Project Gutenberg™ name associated with the work. You can easily comply with the terms of this agreement by keeping this work in the same format with its attached full Project Gutenberg™ License when you share it without charge with others. 1.D. The copyright laws of the place where you are located also govern what you can do with this work. Copyright laws in most countries are in a constant state of change. If you are outside the United States, check the laws of your country in addition to the terms of this agreement before downloading, copying, displaying, performing, distributing or creating derivative works based on this work or any other Project Gutenberg™ work. The Foundation makes no representations concerning the copyright status of any work in any country other than the United States. 1.E. Unless you have removed all references to Project Gutenberg: 1.E.1. The following sentence, with active links to, or other immediate access to, the full Project Gutenberg™ License must appear prominently whenever any copy of a Project Gutenberg™ work (any work on which the phrase “Project Gutenberg” appears, or with which the phrase “Project Gutenberg” is associated) is accessed, displayed, performed, viewed, copied or distributed:
- 63. This eBook is for the use of anyone anywhere in the United States and most other parts of the world at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org. If you are not located in the United States, you will have to check the laws of the country where you are located before using this eBook. 1.E.2. If an individual Project Gutenberg™ electronic work is derived from texts not protected by U.S. copyright law (does not contain a notice indicating that it is posted with permission of the copyright holder), the work can be copied and distributed to anyone in the United States without paying any fees or charges. If you are redistributing or providing access to a work with the phrase “Project Gutenberg” associated with or appearing on the work, you must comply either with the requirements of paragraphs 1.E.1 through 1.E.7 or obtain permission for the use of the work and the Project Gutenberg™ trademark as set forth in paragraphs 1.E.8 or 1.E.9. 1.E.3. If an individual Project Gutenberg™ electronic work is posted with the permission of the copyright holder, your use and distribution must comply with both paragraphs 1.E.1 through 1.E.7 and any additional terms imposed by the copyright holder. Additional terms will be linked to the Project Gutenberg™ License for all works posted with the permission of the copyright holder found at the beginning of this work. 1.E.4. Do not unlink or detach or remove the full Project Gutenberg™ License terms from this work, or any files containing a part of this work or any other work associated with Project Gutenberg™. 1.E.5. Do not copy, display, perform, distribute or redistribute this electronic work, or any part of this electronic work, without prominently displaying the sentence set forth in paragraph 1.E.1
- 64. with active links or immediate access to the full terms of the Project Gutenberg™ License. 1.E.6. You may convert to and distribute this work in any binary, compressed, marked up, nonproprietary or proprietary form, including any word processing or hypertext form. However, if you provide access to or distribute copies of a Project Gutenberg™ work in a format other than “Plain Vanilla ASCII” or other format used in the official version posted on the official Project Gutenberg™ website (www.gutenberg.org), you must, at no additional cost, fee or expense to the user, provide a copy, a means of exporting a copy, or a means of obtaining a copy upon request, of the work in its original “Plain Vanilla ASCII” or other form. Any alternate format must include the full Project Gutenberg™ License as specified in paragraph 1.E.1. 1.E.7. Do not charge a fee for access to, viewing, displaying, performing, copying or distributing any Project Gutenberg™ works unless you comply with paragraph 1.E.8 or 1.E.9. 1.E.8. You may charge a reasonable fee for copies of or providing access to or distributing Project Gutenberg™ electronic works provided that: • You pay a royalty fee of 20% of the gross profits you derive from the use of Project Gutenberg™ works calculated using the method you already use to calculate your applicable taxes. The fee is owed to the owner of the Project Gutenberg™ trademark, but he has agreed to donate royalties under this paragraph to the Project Gutenberg Literary Archive Foundation. Royalty payments must be paid within 60 days following each date on which you prepare (or are legally required to prepare) your periodic tax returns. Royalty payments should be clearly marked as such and sent to the Project Gutenberg Literary Archive Foundation at the address specified in Section 4, “Information
- 65. about donations to the Project Gutenberg Literary Archive Foundation.” • You provide a full refund of any money paid by a user who notifies you in writing (or by e-mail) within 30 days of receipt that s/he does not agree to the terms of the full Project Gutenberg™ License. You must require such a user to return or destroy all copies of the works possessed in a physical medium and discontinue all use of and all access to other copies of Project Gutenberg™ works. • You provide, in accordance with paragraph 1.F.3, a full refund of any money paid for a work or a replacement copy, if a defect in the electronic work is discovered and reported to you within 90 days of receipt of the work. • You comply with all other terms of this agreement for free distribution of Project Gutenberg™ works. 1.E.9. If you wish to charge a fee or distribute a Project Gutenberg™ electronic work or group of works on different terms than are set forth in this agreement, you must obtain permission in writing from the Project Gutenberg Literary Archive Foundation, the manager of the Project Gutenberg™ trademark. Contact the Foundation as set forth in Section 3 below. 1.F. 1.F.1. Project Gutenberg volunteers and employees expend considerable effort to identify, do copyright research on, transcribe and proofread works not protected by U.S. copyright law in creating the Project Gutenberg™ collection. Despite these efforts, Project Gutenberg™ electronic works, and the medium on which they may be stored, may contain “Defects,” such as, but not limited to, incomplete, inaccurate or corrupt data, transcription errors, a copyright or other intellectual property infringement, a defective or
- 66. damaged disk or other medium, a computer virus, or computer codes that damage or cannot be read by your equipment. 1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except for the “Right of Replacement or Refund” described in paragraph 1.F.3, the Project Gutenberg Literary Archive Foundation, the owner of the Project Gutenberg™ trademark, and any other party distributing a Project Gutenberg™ electronic work under this agreement, disclaim all liability to you for damages, costs and expenses, including legal fees. YOU AGREE THAT YOU HAVE NO REMEDIES FOR NEGLIGENCE, STRICT LIABILITY, BREACH OF WARRANTY OR BREACH OF CONTRACT EXCEPT THOSE PROVIDED IN PARAGRAPH 1.F.3. YOU AGREE THAT THE FOUNDATION, THE TRADEMARK OWNER, AND ANY DISTRIBUTOR UNDER THIS AGREEMENT WILL NOT BE LIABLE TO YOU FOR ACTUAL, DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE OR INCIDENTAL DAMAGES EVEN IF YOU GIVE NOTICE OF THE POSSIBILITY OF SUCH DAMAGE. 1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you discover a defect in this electronic work within 90 days of receiving it, you can receive a refund of the money (if any) you paid for it by sending a written explanation to the person you received the work from. If you received the work on a physical medium, you must return the medium with your written explanation. The person or entity that provided you with the defective work may elect to provide a replacement copy in lieu of a refund. If you received the work electronically, the person or entity providing it to you may choose to give you a second opportunity to receive the work electronically in lieu of a refund. If the second copy is also defective, you may demand a refund in writing without further opportunities to fix the problem. 1.F.4. Except for the limited right of replacement or refund set forth in paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
- 67. INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABILITY OR FITNESS FOR ANY PURPOSE. 1.F.5. Some states do not allow disclaimers of certain implied warranties or the exclusion or limitation of certain types of damages. If any disclaimer or limitation set forth in this agreement violates the law of the state applicable to this agreement, the agreement shall be interpreted to make the maximum disclaimer or limitation permitted by the applicable state law. The invalidity or unenforceability of any provision of this agreement shall not void the remaining provisions. 1.F.6. INDEMNITY - You agree to indemnify and hold the Foundation, the trademark owner, any agent or employee of the Foundation, anyone providing copies of Project Gutenberg™ electronic works in accordance with this agreement, and any volunteers associated with the production, promotion and distribution of Project Gutenberg™ electronic works, harmless from all liability, costs and expenses, including legal fees, that arise directly or indirectly from any of the following which you do or cause to occur: (a) distribution of this or any Project Gutenberg™ work, (b) alteration, modification, or additions or deletions to any Project Gutenberg™ work, and (c) any Defect you cause. Section 2. Information about the Mission of Project Gutenberg™ Project Gutenberg™ is synonymous with the free distribution of electronic works in formats readable by the widest variety of computers including obsolete, old, middle-aged and new computers. It exists because of the efforts of hundreds of volunteers and donations from people in all walks of life. Volunteers and financial support to provide volunteers with the assistance they need are critical to reaching Project Gutenberg™’s goals and ensuring that the Project Gutenberg™ collection will
- 68. remain freely available for generations to come. In 2001, the Project Gutenberg Literary Archive Foundation was created to provide a secure and permanent future for Project Gutenberg™ and future generations. To learn more about the Project Gutenberg Literary Archive Foundation and how your efforts and donations can help, see Sections 3 and 4 and the Foundation information page at www.gutenberg.org. Section 3. Information about the Project Gutenberg Literary Archive Foundation The Project Gutenberg Literary Archive Foundation is a non-profit 501(c)(3) educational corporation organized under the laws of the state of Mississippi and granted tax exempt status by the Internal Revenue Service. The Foundation’s EIN or federal tax identification number is 64-6221541. Contributions to the Project Gutenberg Literary Archive Foundation are tax deductible to the full extent permitted by U.S. federal laws and your state’s laws. The Foundation’s business office is located at 809 North 1500 West, Salt Lake City, UT 84116, (801) 596-1887. Email contact links and up to date contact information can be found at the Foundation’s website and official page at www.gutenberg.org/contact Section 4. Information about Donations to the Project Gutenberg Literary Archive Foundation Project Gutenberg™ depends upon and cannot survive without widespread public support and donations to carry out its mission of increasing the number of public domain and licensed works that can be freely distributed in machine-readable form accessible by the widest array of equipment including outdated equipment. Many
- 69. small donations ($1 to $5,000) are particularly important to maintaining tax exempt status with the IRS. The Foundation is committed to complying with the laws regulating charities and charitable donations in all 50 states of the United States. Compliance requirements are not uniform and it takes a considerable effort, much paperwork and many fees to meet and keep up with these requirements. We do not solicit donations in locations where we have not received written confirmation of compliance. To SEND DONATIONS or determine the status of compliance for any particular state visit www.gutenberg.org/donate. While we cannot and do not solicit contributions from states where we have not met the solicitation requirements, we know of no prohibition against accepting unsolicited donations from donors in such states who approach us with offers to donate. International donations are gratefully accepted, but we cannot make any statements concerning tax treatment of donations received from outside the United States. U.S. laws alone swamp our small staff. Please check the Project Gutenberg web pages for current donation methods and addresses. Donations are accepted in a number of other ways including checks, online payments and credit card donations. To donate, please visit: www.gutenberg.org/donate. Section 5. General Information About Project Gutenberg™ electronic works Professor Michael S. Hart was the originator of the Project Gutenberg™ concept of a library of electronic works that could be freely shared with anyone. For forty years, he produced and distributed Project Gutenberg™ eBooks with only a loose network of volunteer support.
- 70. Project Gutenberg™ eBooks are often created from several printed editions, all of which are confirmed as not protected by copyright in the U.S. unless a copyright notice is included. Thus, we do not necessarily keep eBooks in compliance with any particular paper edition. Most people start at our website which has the main PG search facility: www.gutenberg.org. This website includes information about Project Gutenberg™, including how to make donations to the Project Gutenberg Literary Archive Foundation, how to help produce our new eBooks, and how to subscribe to our email newsletter to hear about new eBooks.
- 71. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com