Confronting Metaphor in Use
Pragmatics & Beyond New Series (P&BNS) Pragmatics & Beyond New Series is a continuation of Pragmatics & Beyond and its Companion Series. The New Series offers a selection of high quality work covering the full richness of Pragmatics as an interdisciplinary field, within language sciences.
Editor Andreas H. Jucker
University of Zurich, English Department Plattenstrasse 47, CH-8032 Zurich, Switzerland e-mail:
[email protected]
Associate Editors Jacob L. Mey
University of Southern Denmark
Herman Parret
Jef Verschueren
Susan C. Herring
Emanuel A. Schegloff
Belgian National Science Foundation, Universities of Louvain and Antwerp
Belgian National Science Foundation, University of Antwerp
Editorial Board Shoshana Blum-Kulka Hebrew University of Jerusalem
Jean Caron
Université de Poitiers
Indiana University
Masako K. Hiraga
St.Paul’s (Rikkyo) University
University of California at Los Angeles
Deborah Schiffrin
David Holdcroft
Georgetown University
Sachiko Ide
Kobe City University of Foreign Studies
Sandra A. Thompson
Thorstein Fretheim
Catherine KerbratOrecchioni
John C. Heritage
Claudia de Lemos
Teun A. van Dijk
Marina Sbisà
Richard J. Watts
Robyn Carston
University College London
Bruce Fraser
Boston University University of Trondheim University of California at Los Angeles
University of Leeds Japan Women’s University
University of Lyon 2 University of Campinas, Brazil University of Trieste
Paul Osamu Takahara
University of California at Santa Barbara Pompeu Fabra, Barcelona University of Berne
Volume 173 Confronting Metaphor in Use: An applied linguistic approach Edited by Mara Sophia Zanotto, Lynne Cameron and Marilda C. Cavalcanti
Confronting Metaphor in Use An applied linguistic approach
Edited by
Mara Sophia Zanotto Pontifical Catholic University of São Paulo, Brazil
Lynne Cameron The Open University, UK
Marilda C. Cavalcanti State University of Campinas, Brazil
John Benjamins Publishing Company Amsterdam / Philadelphia
8
TM
The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.
Library of Congress Cataloging-in-Publication Data Confronting metaphor in use : an applied linguistic approach / edited by Mara Sophia Zanotto, Lynne Cameron, Marilda C. Cavalcanti. p. cm. (Pragmatics & Beyond New Series, issn 0922-842X ; v. 173) Includes bibliographical references and index. 1. Metaphor. I. Zanotto, Mara Sofia. II. Cameron, Lynne. III. Cavalcanti, Marilda do Couto. P301.5.M48C655 2008 418--dc22 isbn 978 90 272 5417 7 (Hb; alk. paper)
2007048315
© 2008 – 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
Table of contents
Acknowledgements Introduction
vii 1
Applied linguistic approaches to metaphor Mara Sophia Zanotto, Lynne Cameron and Marilda C. Cavalcanti
Part I Investigating the nature of metaphor in use
9
1.
Opening Pandora’s box: Multiple readings of ‘a metaphor’ Mara Sophia Zanotto and Dieli Vesaro Palma
11
2.
Metaphor shifting in the dynamics of talk Lynne Cameron
45
3. Adding sound to the picture: Motivating the lexical composition of metaphorical idioms in English, Dutch, and Spanish Frank Boers and Hélène Stengers 4.
Metaphor and positioning in academic book reviews Graham Low
Part II Examining metaphor in corpora 5. Brothers in arms: Contradictory metaphors in contemporary marketing discourse Veronika Koller
63 79
101 103
6.
Metaphor probabilities in corpora Tony Berber Sardinha
127
7.
Corpus linguistic data and conceptual metaphor theory Alice Deignan
149
8. Exploring metaphors in corpora: A study of ‘war’ in corpus generated data Solange Vereza
163
Table of contents
Part III Understanding metaphor in language education 9.
Young learners’ understanding of figurative language Ana M. Piquer-Piriz
10. Th e relationship between associative thinking, analogical reasoning, image formation and metaphoric extension strategies Jeannette Littlemore Part IV Using metaphor as a tool in professional development
181 183
199
223
11. The awakening of Sleeping Beauty: A teacher’s metaphor of professional development and language teaching João A. Telles
225
12. En route through metaphors: Chatrooms as safe places to deal with difficulties in an online course Marilda C. Cavalcanti and Ana Cecilia Bizon
243
13. School teachers in favela contexts: Metaphors they live by Fernanda Coelho Liberali 14. Professional knowledge landscape: A metaphor to conceive of e-mail practices in business settings Maximina M. Freire
267
287
Index of subjects
311
Index of names
313
Acknowledgements
Mara Sophia Zanotto would like to thank Jacob Mey for his acknowledgement of the importance of the conference Metaphor in Language and Thought held in Sao Paulo, Brazil in 2002 which originated this volume, and for his encouragement towards this publication. The editors would also like to thank Carol Johns-MacKenzie of the Open University for her expert help with preparing the manuscript, and Isja Conen of John Benjamins for seeing the manuscript through to publication.
Confronting metaphor in use An applied linguistic approach Introduction People use metaphor in the ordinary moments of their lives. In family and work situations, they use metaphor to explain their thoughts and ideas to other people, and to express delight, caring, approval, as well as their more negative counterparts. Metaphor surrounds us as we go about our daily business, impacting on how we understand ourselves and others. It is timely for researchers to approach metaphor as social and situated, as a matter of language and discourse, and not just as a matter of thought. Over the last twenty five years, scholars have come to appreciate in depth the cognitive, motivated and embodied nature of metaphor, but have tended to background the linguistic form of metaphor and have largely ignored how this connects to its role in the discourses in which our lives are constructed and lived. This book brings language and social dimensions into the picture, offering snapshots of metaphor use in real language and in real lives across the very different cultures of Europe and Brazil and contributing to the theorising of metaphor in discourse. It originated in the 2002 conference Metaphor in Language and Thought held in São Paulo, Brazil. In addition, some supplementary papers were specially commissioned. This conference, and others such as the series Researching and Applying Metaphor, showed that many applied linguists are employing a range of methods to investigate metaphor in real world situations. The collection of papers in the book makes an important contribution to the development of an applied linguistic approach to metaphor by bringing together a wide-ranging collection of empirical studies. Between them, these studies demonstrate how metaphor can be taken as a research focus in fields of language use and can be used itself as an “analytic tool” (Zanotto & Cavalcanti 2003; http://creet.open.ac.uk/projects/metaphor-analysis/ index.cfm) to understand human interaction and thinking. This introductory chapter sets out what we mean by an applied linguistic approach and the implications of adopting such an approach for the study of metaphor. It outlines how this approach builds on conceptual metaphor theory, while also challenging it through its position that the language of metaphor must be integral to theory and method.
What is an applied linguistic approach to metaphor? There are various interpretations of ‘applied linguistic’ but central to how we intend the term here is language in use – our problem space is language used by people in
Introduction: Confronting metaphor in use
particular contexts. In this we follow the British applied linguist, Brumfit who defines applied linguistics as “the theoretical and empirical study of real world problems in which language plays a central role” (Brumfit 1995: 27). The real world for applied linguistics has historically been the teaching and learning of foreign languages, but nowadays applied linguistics goes beyond education to encompass language use in a range of workplace situations and contexts, including health, the law, forensic studies, development and human rights. In this book, we encounter metaphor at work in English, Portuguese, Spanish, and Dutch; in talk, e-mail, classroom interaction, poetry, book reviews, and advertisements; in schools, in training and in the globalised commercial world. Importantly, the focus on the real world means that applied linguistic researchers are committed to considering the implications of their studies – research often includes reflexivity and / or intervention that aims to implement or evaluate improvements in practice, to change as well as to understand. It will be clear that this interpretation of applied linguistics is very different from its more simplistic interpretation as ‘the application of linguistics’. Contemporary applied linguistics has developed theoretical and research tools suited to the exploration and understanding of contextualised language use, and finds the abstracted, decontextualised systems of formal linguistics seldom adequate for this work. While some academic departments of applied linguistics may have originated in departments of linguistics, the maturing of applied linguistics has led to a great deal of separation between the disciplines over the last decades (Cavalcanti 2004). Applied linguistic approaches to metaphor do not seek a linguistic theory of metaphor to apply; rather they draw together aspects of theory and methodology that serve to address research problems. In focusing on language used in real-world contexts, this often leads to work that crosses disciplines, engaging with theory around social, historical and cultural factors. In line with Candlin & Sarangi (2004: 2): applied linguistics [is seen] as a many centred and interdisciplinary endeavour whose coherence is achieved in purposeful, mediated practice by its practitioners. Achieving such focused principled action will allow us to accommodate perhaps unexpected theoretical and methodological bedfellows.
With a problem space that is real world language use, applied linguistic approaches to metaphor must consider both linguistic metaphor and conceptual metaphor. We reject the backgrounding of language that has sometimes appeared to be inherent in cognitive approaches; not only do we need to consider metaphor in its ‘natural environment’ of discourse when tackling applied linguistic problems, but we also suggest that an understanding of the multi-faceted phenomenon that is metaphor would be incomplete without knowledge of how metaphor operates in its discourse environment. While cognitive approaches to metaphor are characterised by a commitment to seek general principles, what Gibbs has called the ‘generalization commitment’ (Gibbs 1994: 15), applied linguistic approaches must pay attention to the specifics of
Introduction: Confronting metaphor in use
the contexts in which metaphor is used and the people who use it. This ‘contextualisation commitment’ leads to a requirement that we deal with both variation and stability in metaphor. While variation of metaphor with culture is a relatively recent concern for cognitive approaches (Kövesces 2005), variation is expected and central when, as in applied linguistic approaches, we work with contextualised use of metaphor. Variation in both conceptual and linguistic metaphor use is a key theme of this book: variation in choice of metaphors and in the sense made of other people’s metaphors; variation resulting from the gender, the history or social position of speakers; variation resulting from the purpose of discourse and from the organisation of discourse. All this variation presents challenges for a generalised theory of metaphor, challenges that we believe to be healthy for the field. We are forced away from a single, and, it has to be said, sometimes hegemonic, theory of metaphor as conceptual, towards a more pluralistic, multi-disciplinary perspective on metaphor in use. The idea of stability combined with variation emerges as a theme in this book as a direct result of our view of metaphor as dynamic, rather than static. Metaphor in use is dynamic because it is constructed by people as they talk or write, to suit their evolving purposes and goals (Cameron, 2007). The use of metaphor in a first language is learnt by children in the course of their interaction with adults and other children, in the context of their families and other socio-cultural groups. As it is used and re-used in new communicative contexts, metaphor in language and in thinking is continually changed and adapted. Out of this change, stabilities of various sorts emerge that we may label, for example, idiom or conceptual metaphor. When we analyse metaphor across a range of contexts, we can focus on what is stable, leading to generalisations, or we can focus on what changes, leading to contextualisations. We suggest that a focus on both stability and change, on how specific instances relate to longer term stabilities, can illuminate the dynamic nature of metaphor.
Researching metaphor in an applied linguistic approach Adopting an applied linguistic approach to metaphor has the immediate implication that a single research method is unlikely to be adequate for investigating metaphor in real world contexts. This collection of research studies illustrates how diverse methodologies are used and different techniques are combined to produce the rich data that are often needed to answer applied linguistic research questions. Going beyond experimental studies which have been characteristic of metaphor research in the last three decades (see Zanotto & Palma, this volume), this book fills in a gap in methodological diversification to deal with empirical data and with qualitative research in metaphor studies, while also including quantitative and qualitative/quantitative research. There are six qualitative studies (Cameron, Zanotto & Palma, Cavalcanti & Bizon, Liberali, Telles, Freire), four chapters with a primarily quantitative approach (Deignan, Berber Sardinha, Vereza, Boers & Stengers) and three chapters which present a combination of quantitative/qualitative techniques (Low, Piquer-Piriz, Koller).
Introduction: Confronting metaphor in use
At the very least, analysis of language as discourse must be accompanied by contextual information about language users and the situation of use. Contextual information may be generated using techniques such as those based on observation (and recorded in field notes and diary studies), audio and video recordings and interviews. Researchers are often interested in the understandings and assumptions of language users, and must turn to interpretive methodology to investigate “human meaning in social life and (in) its elucidation and exposition” (Erickson 1986, 119). In addition to the techniques mentioned above, think-aloud protocols (Ericsson & Simon 1984) deserve special attention because they can provide relevant insights about reflexive stages which cannot be captured in experiments (see Zanotto & Palma this volume). The technique used to access the perspectives of research participants in metaphor studies is an adaptation of the conventional think-aloud procedure which has been developed within a socio-cognitive paradigm in the work of the editors of this collection (Cameron 2003; Cavalcanti 1983; Cavalcanti & Zanotto 1994; Zanotto 1992, 1998). Cameron (2003) developed a socio-cognitive version of think-aloud for use with children, working in pairs and addressing a task as they read a text sentence by sentence and discuss the meaning of metaphors. Zanotto’s Metaphor and Indeterminacy Group at the Pontifical Catholic University of São Paulo have developed a collaborative think-aloud technique with students which produces group verbal protocols. Again, think-aloud procedure alone does not provide sufficient information, and other techniques are used alongside, including reflection in diaries and retrospective interviews. While combinations of research methods appear to be inevitable in the investigation of metaphor in use, the issue of how techniques are selected and combined requires attention to ‘trustworthiness’ (Lincoln & Guba 1985; Denzin & Lincoln 1998) and to theoretical compatibility. We have selected studies for this volume that we feel meet such criteria, and that can thus illustrate methodological possibilities for other researchers. The book is divided into sections: • • • •
Investigating the nature of metaphor in use Examining metaphor in corpora Understanding metaphor in language education Using metaphor as a tool in professional development.
We now explain this organisation and site the chapters within our applied linguistic approach to metaphor.
Investigating the nature of metaphor in use: stability and variation The first four chapters of the book take metaphor as research object and explore the nature of contextualised metaphor in different types of discourse. This section presents a picture of metaphor in use, in which large scale stabilities emerge from micro-level dynamics and change. The papers of Zanotto & Palma, Cameron, Boers & Stengers,
Introduction: Confronting metaphor in use
and Low show how people adapt their formulations and interpretations of metaphor when they are engaged in talking, reading or writing. Zanotto & Palma and Cameron investigate very specific contexts, in which the messiness of real world metaphor use is displayed and described. The first deals with the multiple readings of metaphor in the context of classrooms. Using interpretive methodology, the study reveals how making decisions about the metaphoricity of a vehicle term is a complex process that may involve interwoven threads of metaphor and metonymy. Cameron´s chapter evidences the shifting nature of metaphor in talk, presenting a categorisation of types of metaphor shifting that can be used in analytic work across contexts. The paper by Boers and Stengers is important in reminding us of the importance of the phonological, much neglected in metaphor studies to date. They investigate lexical idioms, i.e., metaphorical expressions that have stabilised in a language, and suggest that the stabilisation and standardisation of lexical idioms are influenced by patterns of sound. In an interaction across scales of use, the micro-level of phonological realisation contributes to stabilisation of lexical idioms at the macro level. The particularity of stabilised idioms often causes problems for language learners, who may not be aware of exactly which linguistic aspects are frozen and which remain slightly variable (Cameron & Deignan 2006). Low examines the discourse of academic book reviews, combining metaphor analysis with positioning theory. We see how metaphor is used pragmatically as well as conceptually in this genre.
Examining metaphor in corpora The chapters in this section move to a more macro-level, where corpus analytic techniques become a valuable resource for studying stability and variation in large collections of contextualised uses of metaphor. Koller integrates cognitive and discursive approaches into her analysis of the metaphoric conceptualisation of brands, showing that socio-cultural influences are as relevant as embodiment for the construction of cognitive schemata. In Koller’s corpus of marketing discourse, a newly emergent conceptualisation is shown in action. Berber Sardinha’s chapter takes a probabilistic approach to investigate the metaphorical use of words and phrases in two corpora of Brazilian Portuguese, one restricted by genre and one non-specific. The chapter takes inspiration from Halliday’s work on the probability of use of lexicogrammatical items within the language system, and shows how probabilities offer some measure of the extent to which metaphor stabilisation has taken place. The techniques that Berber Sardinha develops can be used with corpora in other languages. Deignan’s chapter pushes further the argument that the linguistic patterns found in corpus data are more complex than conceptual metaphor theory has suggested, and that the nature of social interaction needs to form part of an adequate explanation of the data. Finally in this section, Vereza explores the collocations of metaphors in corpus data, focusing on the word war and using real world data to challenge generalisations made about conceptual metaphors such as argument is war.
Introduction: Confronting metaphor in use
The first sections of the book thus argue that rigorous description of contextualised metaphor use, from the most micro-level of talk to the macro-level of large corpora, is essential as part of developing an adequate explanation of the phenomenon of metaphor.
Understanding metaphor in language education In the third section, the focus moves to the specific context of language education where the goal is to develop learners’ metaphor competence (Littlemore & Low 2006) or metaphor capacity (Cameron 1996). Here attention to the interaction of generalisable stabilities with variation is again of relevance. Learners of a foreign or second language may be able to transfer some competence with metaphor, particularly in comprehension where universal conceptual or primary metaphors may work across languages. The chapters in this section are concerned with what language learners need to learn about metaphor use, and how people, from children to adults, build their metaphorical competence. The chapter by Piquer-Piriz focuses on a younger age group, investigating Spanish children’s comprehension of figurative language in English as a foreign language, combining experimental results with qualitative analysis. In an experimental quantitative study, Littlemore examines a technique to foster the metaphoric competence of university students in England for whom English is a foreign language.
Using metaphor as a tool in professional development The chapters in the fourth section of the book report studies which use metaphor as an analytic tool to explore the construction of professional practice, in the contexts of teacher education (Telles, Cavalcanti & Bizon, Liberali) and business communication (Freire). Each chapter reports an empirical study which identifies and collects metaphor in use, and draws on conceptual metaphor theory to explain the conceptualisation of professional practice. However, cognitive metaphor theory is found not to be sufficient, and each of the papers uses other theoretical frameworks and associated research approaches and techniques to deal with social aspects of metaphor use. Using narrative inquiry as a theoretical approach and hermeneutic phenomenological research procedures, Telles analyses spontaneous metaphors in oral narratives collected as part of an in-depth case study of an individual teacher. The metaphors allow him to (re)construct the teacher’s understandings of herself, of literacy and of her students. In an ethnographic study, Cavalcanti & Bizon use metaphor as a tool to explore computer-mediated interaction between participants in an in-service teacher education course. The metaphors of the teachers, found as they emerged in the chat room interaction, are focused against the theoretical framework of positioning theory. Liberali’s method involves examination of spontaneous metaphors used in oral and written descriptions. Through this process, she uncovers how teachers working in very poor conditions in Brazil think about their schools and students. In a process characteristic of applied linguistic work, the research technique in turn becomes a
Introduction: Confronting metaphor in use
professional development technique to encourage critical reflection by teachers. A further use of hermeneutic or interpretive techniques is found in the paper by Freire who explores the use of computers in the lives of professionals in the business world. Her data are a corpus of e-mail messages, written in English, and interviews with the writers. Again, the research includes reflections on practice by the participants; participation in the research study is designed to be an empowering experience rather than one in which the researcher generates data and then disappears from the scene.
Opening dialogue In this introductory chapter, we have made the case for taking seriously the social and contextual factors that impact on metaphor as a phenomenon of language and of thought, and for doing this through applied linguistic approaches. We have argued that cognitive theory alone cannot adequately serve applied linguistic metaphor research, and that we need to develop a rigorous theoretical and methodological pluralism. We hope that the chapters of the book will serve as evidence for the richness that can be accessed through such pluralism and that needs to be confronted for a full understanding of metaphor. We hope that the collection serves to open up dialogue between researchers that will benefit the field.
References Brumfit, Christopher (1995). Teacher professionalism and research. In G. Cook & B. Seidlhofer (Eds.), Principles and practice in applied linguistics (27–41). Oxford: Oxford University Press. Cameron, Lynne (1996). Discourse context and the development of metaphor in children. Current Issues in Language and Society: Child Language, 3, 49–64 Cameron, Lynne (2003). Metaphor in educational discourse. London: Continuum. Cameron, Lynne, & Alice Deignan (2006). The emergence of metaphor in discourse. Applied Linguistics, 27, 671–690. Cameron, Lynne (2007). Confrontation or complementarity? Metaphor in language use and cognitive metephor theory. Annual Review of Cognitive Linguistics, 5, 107–135. Candlin, Christopher & Srikant Sarangi (2004). Making applied linguistics matter. Journal of Applied Linguistics, 1, 1–8. Cavalcanti, Marilda C. (1983). The pragmatics of F(oreign) L(anguage) reader-text interaction: Key lexical items as source of potential reading problems. Ph.D. thesis. University of Lancaster, Lancaster, England. Cavalcanti, Marilda C. (2004). Applied linguistics: Brazilian perspectives. AILA Review, 17, 23–30. Cavalcanti, Marilda C. & Mara S. Zanotto (1994). Introspection in applied linguistics: Metaresearch on verbal protocols. In L. Barbara & M. Scott (Ed.) Reflections on language learning (148–156). Clevedon, Multilingual Matters. Denzin, Norman. K. & Yvonna Lincoln S. (Eds.) (1998). The landscape of qualitative research. London: Sage.
Introduction: Confronting metaphor in use Erickson, Frederick (1986). Qualitative methods in research on teaching. In M. C. Wittrock (Ed.), Handbook of research on teaching (119–161). New York: MacMillan. Ericsson, K. Anders. & Herbert A. Simon (1984). Protocol analysis. Cambridge, MA: MIT Press. Gibbs, Raymond W. (1994). The poetics of mind: Figurative thought, language and understanding. New York: Cambridge University Press. Kövecses, Zoltán (2005). Metaphor in culture: Universality and variation. Cambridge: Cambridge University Press. Lincoln, Yvonna S. & Egon G. Guba (1985). Naturalistic inquiry. Newbury Park, CA: Sage Publications. Littlemore, Jeannette & Graham D. Low (2006). Figurative thinking and foreign language learning. Basingstoke: Palgrave Macmillan. Zanotto, Mara S. (1992). O processo de compreensão da metáfora na formação dos professores de língua materna (The process of understanding metaphors in teacher education for teachers of the mother tongue) In M. S. Zanotto e M. A. A. Celani (orgs.), Lingüística aplicada: Da aplicação da lingüística à lingüística transdisciplinar (Applied linguistics: From application of linguistics to cross-disciplinary linguistics) (233–246). São Paulo, EDUC. Zanotto, Mara S. (1998). A construção e a indeterminação do significado metafórico no evento social de leitura (The construction and indeterminacy of metaphorical meaning within the social reading event). In V. L. M. O. Paiva, (Ed.), Metáforas do cotidiano (Everyday Metaphors) (13–38). Belo Horizonte: Edit. da UFMG. Zanotto, Mara S. & Marilda C. Cavalcanti (2003). Metaphors in interpretive research in applied linguistics: Focus on education. Paper presented at BAAL 2003, University of Leeds, Leeds, UK, September 4–9.
part i
Investigating the nature of metaphor in use
chapter 1
Opening Pandora’s box Multiple readings of ‘a metaphor’ Mara Sophia Zanotto & Dieli Vesaro Palma This research1 emerged out of a practical need to change reading practices in a classroom environment in order to tune them to a paradigm, which is open to multiple readings, and rejects the absolute interpretive authority of the teacher. When reading was taken as univocal, and the teacher (or the course book author) was considered the authority in interpreting the text, the teacher did indeed feel more secure in carrying out reading tasks in the classroom. However, s/he rarely made room for student subjectivity. In other words, the student’s voice was not taken into account. In fact, it seems that not allowing the student’s voice to be heard has a lot to do with multiple readings, since, if one makes room for students’ subjectivity, and the teacher no longer has the support of the sole reading found in the course book, what occurs may be rather unpredictable and generate considerable uncertainty. Nonetheless, after the paradigmatic changes of the 1970s, studies carried out under different theoretical trends have shown that the traditional view was inadequate, and that a text does in fact allow several readings. Meaning was no longer considered as emanating from the text itself, but as constructed in the interaction of the reader with the text, based on the reader’s context and knowledge – thus the possibility of multiple readings. In the classroom, however, the traditional practice persists – with the teacher’s interpretive authority and the ritual silencing of the student’s voice. Moreover, it is not an easy task to change this practice without providing guidelines to the teacher on how to proceed in the very complex issue of multiple readings. From a theoretical standpoint, multiple readings have been connected with ideas of meaning indeterminacy, which reflect a new approach to meaning. While meaning in the positivist paradigm (at least in terms of literal meaning) was considered something
1. This research is part of the Integrated Project ‘Metaphor, Indeterminacy and Multiple Readings in the Classroom’, developed by GEIM-LA – Metaphor and Indeterminacy Study Group in Applied Linguistics – registered in CNPq (National Council for Scientific and Technological Development), whose members include Palma and whose coordinator is Zanotto.
Mara Sophia Zanotto & Dieli Vesaro Palma
clear, precise and determined, it is now considered inherently indeterminate. As stated by Moura (1998: 107), indeterminacy “occurs when to a linguistic form more than one semantic representation may be attributed, i.e., when the mapping between form and meaning is not univocal.” Therefore, indeterminacy refers to all cases in which meaning is not clearly determined, in other words, it “encompasses a large range of semantic phenomena such as homonymy, polysemy, ambiguity, vagueness, generality, metaphor, and so on.” (Zanotto & Moura 2002: 2). Though metaphor is considered a kind of indeterminacy phenomenon, it is seen, together with presupposition, as “rare and exotic borderline phenomena of natural language.” (Pinkal 1995: 24). For this reason, although recognised by some scholars as inherent to metaphor, metaphorical indeterminacy has not been considered as central focus of research. Indeterminacy scholars, including Martin (1992), Pinkal (1995), Poesio (1995), Wierzbicka (1996) and others, though recognising its existence and complexity, do not attend to this matter. Among metaphor scholars we find a similar picture. Although Black (1979) considered indeterminacy as part of the essence of metaphor, Gibbs (personal communication) has noted that many cognitive linguists seem to ignore this aspect and continue to work with metaphor as if it had a single, determined meaning. This indicates that metaphor indeterminacy needs to be investigated deeply both theoretically and also empirically, because there seems to be no empirical research whose objective is to understand how genuine readers interpreting a text deal with indeterminacy in general, let alone metaphorical indeterminacy. For this reason, we have carried out an empirical investigation using qualitative methodology, in which genuine readers generate (Mason 1996) data in natural contexts, i.e., the classrooms. In earlier research (Zanotto & Palma 2003) we investigated – with several groups of readers – the understanding of a metaphor in an authentic text within a classroom environment. Data analysis led us to conclude that the focused metaphor allowed three compatible readings, probably constituting a type of active polysemy in the context. This study, together with a more recent investigation,2 in which the focused metaphor led to conflicting readings – probably constituting a case of ambiguity – have allowed us to build the hypothesis that perhaps metaphor is not simply another kind of indeterminacy phenomenon with the same status of polysemy, ambiguity, vagueness and generality. Rather, metaphors display different kinds of indeterminacy just like literal language. This means that we can find a metaphor that allows polysemic reading (with compatible readings), or a metaphor that allows ambiguous reading (with conflicting readings), or even vague reading (without definite limits). Thus, it seems that metaphor
2. After the conclusion of the data generation by Zanotto & Palma (2003), Zanotto started another data generation procedure with a different poem by Drummond. The partial analysis of these data was presented in the workshop on Data-based Expert Workshop, as part of the project – MetNet: Metaphor Analysis – Theoretical and Methodological Challenges in Using Discourse Data, coordinated by Lynne Cameron and held in the University of Leeds, in May 2006.
Chapter 1. Opening Pandora’s box
presents the same kinds of indeterminacy as literal language, but perhaps, and most probably, it presents its own specificities. With this hypothesis in mind, we have decided to carry out an empirical3 investigation to map the different kinds of indeterminacy found in metaphors, using data from genuine readers’ interpretations of different metaphors. For each potential type of metaphorical indeterminacy, we will carry out a collective case study (Stake 1998), following interpretive methodology (Erickson 1986; Denzin & Lincoln 1998), asking several groups to read the same text to broaden the possibilities of different readings. The larger project will thus involve several collective case studies (see methodology), with groups of readers reading several texts with potentially different metaphors from the point of view of indeterminacy. This chapter presents the first case study. For this study, we aimed at focusing on the process of co-construction of multiple readings, because the process might shed light on the nature of the undetermined product of interpretation, in other words, the multiple readings. We sought to answer the following questions: a. How do readers co-construct the multiple meanings of the focal metaphorical expression? b. What type of indeterminacy characterises the multiple readings? This chapter is organised in three sections, in addition to the introduction and discussion. In the first section below, we present a synthesis of our theoretical research on indeterminacy and in the second section we explain the research methodology. The third section is dedicated to the data analysis.
Metaphor indeterminacy Indeterminacy and fluidity of metaphorical meaning have, for a long time, constituted an obstacle for those seeking to scientifically focus on metaphor. This is explained by the ‘myth of objectivism’ (Lakoff & Johnson 1980; Johnson 1987), according to which, words must bear clear, precise and determined meanings. As a consequence of this idea, metaphor, as well as other figures of speech, should be avoided when one sought to speak objectively, since their meanings were neither clear, nor precise, nor did they suit reality. Therefore, metaphors were not seen as desirable in scientific discourse, let alone as an object of scientific study.
3. We think that this empirical research, with qualitative methodology, on the indeterminate nature of metaphorical meaning leads us to deal with two of the challenges for metaphor scholars put by Ray Gibbs, in the opening plenary of RaAM 6 (Leeds, April 2006), that are: What do metaphors mean? And how do we count or individuate the meanings of a metaphor?
Mara Sophia Zanotto & Dieli Vesaro Palma
In objectivist semantics, the postulate of meaning determinacy was a key construct. Whenever looked at within this tradition, indeterminacy was seen as subsidiary to the claim of determinacy, as pointed out by Moura (1999): intrinsic indeterminacy found the following alternatives: (a) it was considered a deficiency of natural languages – as opposed to logical languages (Fregean position); (b) if indeterminacy is not considered negative, it should be controlled and eliminated whenever necessary (Quine’s position (1960)); and (c) indeterminacy is controlled within the context, and therefore, does not represent a real problem. (p. 2)
This objectivist understanding of the nature of meaning has been supported since the Pre-Socratic philosophers, and has lasted until recently, dominating Western philosophy and becoming the dominant research program in several subject areas. However, Johnson (1987) points to the seventies as the period in which a crisis seriously challenged this view, theoretically, in the field of Logic, by Hilary Putnam (1981), and empirically by a number of studies in different subject areas, which shared similar concerns, such as a focus on phenomena in which human understanding was required to account for meaning and reason (for example: categorising, concept framing, metaphor, polysemy etc.). Even though the postulate of determinacy has collapsed due to empirical studies, there remains a need for more empirical research to investigate the nature of indeterminate meaning. Because the study reported here focuses on metaphor in use, as interpreted by genuine readers in the natural classroom environment, it can contribute to unveiling the indeterminate nature of metaphoric meaning. It seems to us that the most important action is deep data analyses that point to what a theory of meaning, especially that of indeterminate meaning, must explain. And it is likely that even more complex theories will be needed to deal with indeterminacy as an inherent characteristic of ordinary language (Rommetveit 1985). As stated in the introduction, theoreticians in the field of semantics and pragmatics who deal with indeterminacy have not deeply discussed the specific case of metaphor. Among the authors who present a brief discussion of metaphoric indeterminacy, we could mention Martin (1992) and Pinkal (1995). Martin (1992) dedicates some thought to “semantisme flou” of metaphors, and discusses the metaphoric entailments4 as an open field of vague boundaries. Nevertheless what we have been finding in our data does not seem to refer only to metaphoric entailment but also to different target domains mapped on to the same source domain. Pinkal (1995), for his part, considers that more or less conventional metaphors are highly vague and ambiguous, and goes
4. “Le champ sur lequel se pratique l’implication métaphorique est un champ ouvert, aux limites floues.” (Martín 1992: 224) (The field on which metaphoric entailment is practiced is an open field of vague boundaries.)
Chapter 1. Opening Pandora’s box
on to generalise saying that “all kinds of semantic indefiniteness meet in the world of metaphors” (p. 23). This last statement reinforces our hypothesis, according to which metaphor displays the same kinds of indeterminacy as literal language. On the other hand, theoreticians are generally concerned with ‘precisefying’ 5 – i.e., with specifying meaning or solving indeterminacy. Even Gibbs, when dealing with multiple meanings of metaphor, tries to specify a meaning, by estimating the author’s intention (Gibbs 1987), or by investigating “how context constrains the interpretation of the indeterminate, non-propositional meanings of metaphor” (Gibbs and Gerrig 1989: 146). But he ends up conceding that “metaphors are often viewed as open-ended precisely because their ambiguity resists a single, definitive propositional meaning” (Gibbs 1987: 44). Yet, as far as metaphors are concerned, indeterminacy seems not to be resolved by context, nor even through knowledge of the author’s intention. In other words, there is, according to Black (1979: 25), “an inescapable indeterminacy”, essential to the very nature of metaphors. For this reason, we dare to suggest that it is precisely indeterminacy that makes metaphors so special. What is generally considered a problem that scholars want to solve is actually a characteristic inherent to metaphors, making it a phenomenon that has bewitched several scholars (Haskell 1987).
Methodology Qualitative methodology and Group Think-Aloud technique As pointed out by Gibbs (1994), we have witnessed a real empirical boom in metaphor studies in the last thirty years, largely as a result of experimental studies by cognitive psychologists and psycholinguists into the comprehension process of metaphor. The glut of experimental research studies is complemented by equally large shortage of qualitative studies. Among metaphor scholars who base their work on qualitative methodology, we find Steen (1994), Cameron (2002, 2003) and Zanotto and her research group (Zanotto 1992, 1995, 1998, 2001; Nardi 1999; Palma 1998; Vieira 1999). Our research on the construction processes of multiple readings of metaphors adopts a qualitative methodology, of interpretive orientation, “in which the immediate (often intuitive) meanings of actions to the actors involved are of central interest” (Erickson 1986: 120). Within this methodological frame, we have been carrying out collective case studies of an instrumental nature, as explained further below.
5. Pinkal (1995) and Poesio (1995) proposed the Precisification Principle, according to which “lexical meaning is intrinsically undetermined and can be precisified in many ways, according to the context.” (Zanotto & Moura 2002: 4)
Mara Sophia Zanotto & Dieli Vesaro Palma
The main method used to “generate” (Cf. Mason 1996) data is the Group Think-Aloud technique (Zanotto 1992, 1995), based on Brown and Lytle’s use of interactive thinking aloud as a pedagogical tool (Brown & Lytle 1988). While experimental methodology allows researchers “to tap into on-going, very fast, mostly unconscious, cognitive and linguistic processes” (Gibbs 1999a: 38), Group Think-Aloud, in contrast, is an instrument which allows investigation of conscious reflection about a text.6 Such activity allows participants to deepen their interpretation and multiple readings to take place. The Group Think-Aloud technique is a socio-cognitive7 re-interpretation of traditional Think-Aloud procedure (Ericsson & Simon 1984), based on meta-research on this technique carried out by Cavalcanti & Zanotto (1994). It is in fact a group reading practice, “in which readers, in a face-to-face interaction, share, negotiate, construct and assess different readings” (Zanotto 1997: 3). It is similar to the adapted procedure used by Cameron (2002, 2003), also in the socio-cognitive paradigm: Goal-directed Interactive Think-aloud (GITA). In concrete terms, the practice of thinking aloud in group is a social reading event (Bloome 1983, 1993), in which the teacher’s role is only to coordinate and mediate the discussion, giving up his/her role as interpretive authority and making room for students’ voices in the construction of readings. For this reason, besides being a technique of research, the interactive Think-Aloud is also a pedagogical tool. If the teacher does not take part in the activity, s/he should find out during the retrospective interview that follows if they did indeed focus on metaphor or s/he must take this opportunity to clarify doubts on thoughts that had been elicited and remained incomplete, or other aspects that were left unclear. The students are also asked to write a reflective diary after the activity is finished. Therefore, in our study, three different techniques are combined – Group ThinkAloud, reflective diary and retrospective interview – to allow for a more complete elicitation of data and for a triangulation process in data analysis. In interpretive research, triangulation refers to the crossing and confirming of data generated by one technique with data generated by other techniques, yielding “stronger evidence than if this evidence had been originated only from one source of information” (Erickson 2001: 14). However, in this chapter, due to space constraints and to the fact that the other tools have confirmed the readings constructed in the Think-Aloud task, only the data generated by this technique will be analysed. It is important to clarify that Group Think-Aloud is not an isolated data construction event. As a social reading event, it makes room for readers’ subjectivities to be 6. As Gibbs (1999a) points out, metaphor understanding is “a process that occurs in real-time along a variety of temporal dimensions, starting in the first milliseconds of unconscious processing and extending up to long-term, reflective analysis.” (p. 38) 7. When we decide to place ourselves in a socio-cognitive perspective, we are agreeing with Cameron (2002: 687) who states that, considering Vygotsky’s work, some versions of sociocultural theory “downplay the importance that Vygotsky gave to individual learning.”
Chapter 1. Opening Pandora’s box
displayed, allowing the students to reveal themselves to the group, and exhibit personal ideas, feelings and experiences. Therefore, it is important that the students know one another and feel at ease in the company of their peers and teacher. For this reason, the activity occurs in the context of a course, when students are already familiar not only with one another but also with the teacher/researcher, who can motivate them to experience this kind of activity.
Instrumental and collective case study The problem of multiple readings of metaphor arose from previous research studies carried out by Zanotto (Zanotto & Ricciardi (1984), Zanotto (1995, 1998)). The focus of these works was metaphor comprehension processes, but, in data obtained, multiple readings emerged with such strength and complexity that finding a methodological (and theoretical) framework appropriate to the problem became a challenge. As this was an investigation into multiple readings of metaphor, it was fundamental that we worked with different groups of readers so as to allow for varied reading constructions. We thus thought it appropriate to design an instrumental and collective case study (Stake 1998), following interpretive methodology. The reason for considering this an instrumental case study is that “a particular case is examined to provide insight into an issue or refinement of theory” (Stake 1998: 88). Therefore, the case will be examined (in depth) so as to shed some light on the matter of metaphor indeterminacy. Our interest in these participating groups does not relate to intrinsic aspects of the groups themselves, but in how they can contribute to clarifying a matter external to them, that is, the question of multiple readings. And because we worked with several groups (cases), the research is a collective case study, i.e., an “instrumental study extended to several cases” (Stake 1998: 89). Having defined the design, data from Zanotto & Ricciardi (1984) and Zanotto (1995, 1998) were used to form the collective case study, together with the data generated from four groups of students who took part in the research in 2003. The investigation divides into two phases, reflecting the two moments of research. The first phase involved two groups: – GA group – was composed of Zanotto and Ricciardi, who spontaneously constructed and negotiated their readings in an oral conversation, unknowingly accomplishing what today we call Group Think-Aloud or group protocol (see Zanotto & Ricciardi 1984). – GB group – was composed of 6 postgraduate students who experienced the group protocol task, in the classroom, coordinated by Zanotto, in 1992. Data were partially analysed in Zanotto (1995, 1998) In the second phase, Zanotto worked with 3 groups (G1, G2 and G3), each of 5 students, experiencing thinking aloud in group in the classroom, in 2003. After carrying out the think-aloud task without the participation of the teacher, each group was
Mara Sophia Zanotto & Dieli Vesaro Palma
interviewed and the students were also requested to write a reflective diary. After that, the three groups took part in a general discussion in which the different possibilities of reading were discussed. They thought the readings over to see if they would accept all of them, and produced criteria on which to base their acceptance or rejection of each reading. At the same time, Palma was working with another group (G4) of 7 students, coordinating them in the activity of thinking aloud in group. After the activity, a retrospective interview was carried out, and students were asked to keep a reflective diary. A final discussion was held one week later. To sum up, we have data from two groups in the first phase and four groups in the second phase, which will be triangulated in the analysis.
The text The text used for this research was the poem ‘Fraga e Sombra’ [Cliff and Dusk] by Carlos Drummond de Andrade, a well-known Brazilian poet of the twentieth century. This text was selected because it has novel, and somewhat difficult metaphors, which should lead to non-automatic reading and to the elicitation of metaphor comprehension processes. Fraga e Sombra
Cliff and Dusk
A sombra azul da tarde nos confrange. Baixa, severa, a luz crepuscular. Um sino toca, e não saber quem tange É como se este som nascesse do ar.
The crepuscular blue shadow crushes us Dim and grim, the indistinct evening twilight A bell tolls, and not knowing the one who does Sounds as though from the air it came to light
Música breve, noite longa. O alfanje Que sono e sonho ceifa devagar Mal se desenha, fino, ante a falange Das nuvens esquecidas de passar.
Short melody, long-lasting night. The sickle That sleep and dream slowly reaps Has just become thin-drawn and a flock Of untroubled clouds, high in the sky, floats
Os dois apenas, entre céu e terra, Only the two of us between the earth and the heavenly sky Sentimos o espetáculo do mundo, Did feel the huge moving spectacle of the universe Feito de mar ausente e abstrata serra. Made of absent sea and abstract ridges infinitely high E calcamos em nós, sob o profundo Instinto de existir, outra mais pura Vontade de anular a criatura.
And we crushed inside us under the profound Instinct of surviving, other, even more intense, Vivid will, that of annihilating our very essence.
ANDRADE, Carlos Drummond de. Reunião. Rio de Janeiro, Livraria José Olympio Edit, 1969, p. 177.
(Translation by Sandra Madureira)
Chapter 1. Opening Pandora’s box
The focal metaphor The metaphor, whose comprehension process will be the focus of this study, is the implicit metaphor (Steen 1999) that appears in the first line of the second quatrain: “the sickle”. According to classic rhetoric, the implicit metaphor is a “metaphor in absentia”. This kind is opposed to the “metaphor in praesentia”, or explicit metaphor (Steen 1999), whose terms A and B, are usually clearly stated. Metaphor in absentia brings only one of the concepts, B, and the reader needs to seek A. As stated by Cameron (1999: 12), it is important to escape from the “recent trend of reducing all metaphors to the form of A is B”, and so we have chosen this implicit metaphor to focus on.
Data analysis The analysis is organised in two parts, reflecting the two phases of research. Each part contains subdivisions for each of the groups analysed. We sought to describe, in depth, the construction processes of the multiple readings for the two groups in the first phase, whilst the second phase readings are used to confirm or complement the data found in the first phase. The analysis focused on extracts related to the comprehension process of the sickle metaphor, and was constructed by an inductive search of relevant theories that could explain aspects of the understanding processes found in the data.
First phase – GA and GB groups Group GA The text was read by Zanotto and Ricciardi, who individually constructed different readings, and later negotiated their understandings. During a retrospective conversation, they reached a consensus, accepting the different readings, whose processes were later discussed in Zanotto & Ricciardi (1984).8 First reading of sickle metaphor. The first reading of sickle metaphor was constructed by Zanotto, who read sickle as moon in a bottom-up process of inferences based on textual clues (in italics), as in extract 1 below: Extract 1 In the cataphoric sense, we have, on the surface, sleep and dream which, metonymically, by temporal contiguity, suggest night. This possibility is 8. Our objective in this paper was to focus on a metaphor found in an authentic text, so as to verify the role played by the context in the construction of metaphorical meaning.
Mara Sophia Zanotto & Dieli Vesaro Palma
anaphorically confirmed by Short melody, long-lasting night. The cataphor also offers us another clue: flock of clouds, which, by spatial contiguity, evokes sky (. . .). Thus, aware of the sickle blade shape, we were able to coherently organise the result of inferences and arrive at a possible referent: MOON, both in the shape of a new moon or crescent moon, and in the shape of the last quarter. Anaphoric elements such as shadow, evening, crepuscular and bell tolls – the latter metonymically evoking by cultural-religious contiguity, the hour of the Hail Mary (6 p.m.) – can now be seen as weaving a temporal circumstance, preparing the reader for the arrival of the Moon . . . (From Zanotto & Ricciardi 1984: 45).
In this reading construction, some inferences are made in terms of spatial, temporal and cultural religious contiguities, which we considered to be metonymic processing, by analogy with the distinction between metaphoric processing and processing metaphoric language – proposed by Gibbs (1999a). According to Gibbs, metaphoric processing is “a general mode of understanding that can be applied to any kind of situation or language” (p. 40). On the other hand, processing metaphoric language occurs when there is special linguistic expression or material that is metaphoric. Hence, by analogy, when we inferred the Hail Mary hour from the words “the bell tolls”, we were using a metonymic processing on a linguistic expression that is not metonymic. After the temporal (‘sleep’ and ‘dream’ which, metonymically, by temporal contiguity, suggest ‘night’) and spatial (‘flock of clouds’ which, by spatial contiguity, evokes ‘sky’) circumstances had been inferred by metonymic processing, similarity emerges between the form of a sickle and that of the moon. Consequently, the moon reading involved metonymic processing and the processing of metaphoric language. In this case, however, the metonymic processing, which closely follow contextual clues, serves the purpose of supporting the construction of the metaphorical meaning.
Second reading. The reading of time was constructed by Ricciardi also through a bottom-up process, as we can see in extract 2 below: Extract 2 Taking into account a co-textual element, we notice that to the Moon is attributed the eliminating action of reaping, a verb that synthesises twoimportant ideas; movement and destruction. The Moon is therefore destructive in its movement. (. . .) The subjective attribution of the eliminating characteristic to the Moon is achieved indirectly considering that the true eliminator is the evening time, not the moon itself. It is time that, in its trajectory, destroys night and brings dawn, chasing away sleep and killing dreams. In this sequence of inferences, we conclude that the direct referent of sickle could well be time, if we base our reading on the acting similarities
Chapter 1. Opening Pandora’s box
that both determine in their movements. Incidentally, Khronos, the Time God, brings in his hands a destructive sickle, (. . .) The author, however, presented a chain of references by metaphorically using sickle instead of moon and moon metonymically, instead of evening time, in a temporal contiguity association. (From Zanotto & Ricciardi 1984: 45–46)
In the first paragraph, there is an explanation for the metaphoric interpretation of moon as a reaper because it receives the attribute of carrying out the eliminating action of ‘reaping’. In the second paragraph the metaphorical meaning time is a reaper is constructed by analogical reasoning based on the similarity of action between ‘moon’ and ‘time’. The third paragraph brings reasoning based on the similarity between ‘sickle’ and ‘time’, whose similar trait is the “similarity of action and effect that both determine in their movement” – ‘both’ meaning, in this case, ‘sickle’ (as moon) and ‘time’. At the end of the paragraph, we state that ‘moon’ was used metonymically to replace ‘evening time’, in temporal contiguity association, thus, a metonymic processing. We can therefore confirm that the following readings were metaphorically constructed in a bottom up process: sickle as moon, sickle as time, moon is a reaper, time is a reaper, as well as the reading of moon for time, which involved metonymic processing. Readings occurred at different levels, but showing a tight connection between them – one of them, sickle as moon, serving as support for the construction of the other readings, moon for time and time is a reaper. This leads us to raise the question of whether only the first reading might have been a processing of metaphor language, whilst the others represented metonymic processing and metaphoric processing respectively. Another question relates to the problem of how we count or individuate the readings: do we have four readings or two? For the time being, we consider there to be two: sickle as moon and sickle as time, both metonymic readings connected to the metaphoric readings: moon as a reaper and time as a reaper.
Group GB First reading. The participants of GB group, coordinated by Zanotto, were: Carina, João, Sílvia, Antonio, Cristina and Emília (pseudonyms). The students’ conversation about the text followed the patterns of Group Think-Aloud since this made possible on-line constructions of meanings and moments of immediate retrospection, thus eliciting different readings and different processes of construction. The text was handed out to the students and they were given a few minutes for silent individual reading, after which time, students engaged in conversation about the text. The expectation was that the readings would emerge in the same order as in the GA group, since the moon reading seemed logically necessary to reach the reading of time. Let us see what happened.
Mara Sophia Zanotto & Dieli Vesaro Palma
After 14 initial conversational turns, João focuses on the verse in which the sickle appears, unchaining the construction of one of the readings, as shown in Extract 3:9
Extract 310
1
João life carries on between sleep and dream, right? . . . . and . . . I mean, there is no problem . . . that . . . that background that he shows is just the balance between dream and sleeping, I mean . . . there is nothing else . . . there’s transcendence, isn’t there? 5 Teacher you can say it . . . she has got something there Carina I thought of something else because . . . when he speaks of bell . . . short melody . . . long-lasting night . . . I thought that. . twilight that is short when compared to night that is long . . . I thought Cristina But music could be related to the bell . . . 10 Carina I thought the bell sound I related the bell sound with that of the twilight, at six in the evening . . when the bell tolls . . . that’s what I thought . . . which is exactly at the crepuscular time . . . and which is very short in relation to the night and then . . . he speaks of the . . . er . . . of the sickle, doesn’t he? 15 João Which is . . . Carina which I checked (in the dictionary) to know what it means . . . and it is a sabre . . . João Which is . . . Carina a cutting weapon which would really be time passing by and cutting 20 down the time that people have to dream or to sleep . . . sleep and dream . . . as the night goes past, right? . . . I thought a lot about time/it was an idea that frequently kept returning . . . João Time? Carina time .. the idea of time going past . . .whether slowly or quickly . . . 25 time during the night that goes past. . that keeps cutting . . .
9. The original data were in our native language – Brazilian Portuguese – but they will be presented only in translation into English. 10. Transcription conventions , continuing intonation contour . final intonation contour . . short pause . . . longer pause [ ] overlapping talk ??? indecipherable talk “italic” – reading of the poem
Chapter 1. Opening Pandora’s box
In lines 1 to 3, João brings the hypothesis of “life carries on between sleep and dream” or “(it) is just the balance between dream and sleeping”. At this moment, Carina takes the turn (lines 6–8), after the teacher’s approval (5),11 saying that she had thought of something different and presents her interpretation for sickle, in a kind of immediate retrospection, given that she had just finished her reading of the text. Although Cristina (9) and João (15, 18 and 23) make some attempts to interrupt Carina, her flow of thought seems not to be under threat. She begins by interpreting “short melody, long lasting night”, saying that twilight is short when compared to the night, which is long. In this turn (6–8), she seems to be carrying out a metaphoric processing, since she replaces melody by twilight, seemingly reading, in a metaphoric manner that the “twilight is a short melody”. However, Cristina calls their attention (9) to the reading that the short melody is that of the bell, i.e., she reads literally, and not metaphorically as the melody of the twilight. Carina answers her, making her metonymic processing explicit when she relates the bell tolling to the twilight through temporal contiguity. After that, Carina focuses on the sickle (14). She first constructs its literal meaning: a sabre (17), a cutting weapon (19),12 and then she constructs its metaphorical meaning, seeking for an entity in another domain: which would really be time passing by and cutting down the time that people have to dream or to sleep. She bases her reading mainly on reaps sleep and dream and constructed an analogy between sickle and time by means of the cutting action or function. Analysing Carina’s reading in accordance with Lakoff & Turner (1989: 41, 86), we could say that she activated the conceptual metaphor: time is a reaper,13 reconstructed through analogical reasoning and supported by the metonymic reading of sleep and
11. The numbers of lines that are being analysed will be indicated in parenthesis. 12. She needed to refer to the dictionary – which was in class for students to use – because in Portuguese, the author uses a less well-known word ‘alfanje’ [sickle], rather than ‘foice’ [scythe] – which is more common. In a Portuguese dictionary (Ferreira 1999), alfanje [sickle] has two meanings: (1 a) short, broad bladed sabre; (2 a) long-handled scythe used for cutting grass. In literature, sickle is more used with meaning one, as a metaphor for the moon and a symbol of death. This can be explained because sickle was in its origin a ‘Moorish sabre with a short , broad blade, convex in the shape and employed as a weapon up to the mid-XIV century”. (Grande Enciclopédia Larousse Cultural (Great Larousse Cultural Encyclopedia (1998) São Paulo: Plural Publishers). 13. In fact, this metaphor is part of a composition of metaphors, as pointed by Lakoff & Turner (1989: 74): “. . . time can be personified as a reaper because we have a basic metaphor that people are plants; that time can be a changer because of our common sense notion that the passage of time plays a causal role in bringing about events, especially changes; that time can be a thief because we have the further metaphor that life is a precious possession, and so on.”
Mara Sophia Zanotto & Dieli Vesaro Palma
dream as ‘time to sleep and dream’.14 In this case, we could say that Carina is the sole author of this reading; it was not a case of co-construction since she presented it in immediate retrospection. The fact that this group’s first reading was that of time, by top-down process ing, was contrary to Zanotto’s expectations that the first reading would be that of moon – which, being a symbol of time would inevitably lead to the reading of time, just as with Zanotto and Ricciardi, who constructed their interpretation of time through bottom-up processing. We can therefore conclude that the reading of time could be constructed by following two different paths: a metaphoric path, in group GB and GA, and a metonymic path, in group GA. This variation in processes was not expected.
Transition to the second reading In the extract below, when the teacher asks Carina to confirm her reading of time (1–2), we have the beginning of another reading process initiated by Sílvia, who, using another sentence whose subject is also the sickle (has just become thin-drawn), disagrees with Carina’s reading (3–4). Extract 4 1 Teacher The sickle that sleep and dream slowly reaps – did you interpret this as time? Sílvia Let’s see . . as time? . . . the sickle becomes thin-drawn ahead of clouds? Time. . .drawn? Thin. . thin before clouds? it can’t be so . . .
14. It is important to note that when speaking of time passing by and cutting down the time that people have to dream or to sleep, Carina constructs a causal tautology. Fauconnier & Turner (1999: 84), when explaining the conceptual integration that occurs in the image of death (a similar process to the one that occurs in the reading of ‘time’, whose anthropomorphic representation can be seen in the image of Khronos, the God of Time) state that, in one of the spaces found in the blending, there is ‘an abstract pattern of causal tautology in which an event of certain kind is caused by an abstract causal element: e.g.,: Death causes dying. Sleep causes sleeping. Lust causes lusting.” We could add that: Time cuts time. . . Therefore, in Carina’s speech there are two occurrences of time, generating a situation that is apparently illogical, but that constitutes the causal tautology. The first occurrence is that of ‘time’ as a subject for the verb ‘to cut’ and, if ‘time cuts’, we can easily infer that time is a reaper. The second occurrence of ‘time’ is that of an object of the verb ‘to cut’ – that which suffers the action of being cut down and that, by tautology, is time itself. Nonetheless, this occurrence is followed by a defining relative clause “(time) that people have to sleep and to dream”, which would constitute an indirect form, classified by rhetoric as ‘periphrastic’ (or circumlocution) to say that ‘evening time’ or ‘nights’ that, in turn, would be an indirect way of saying ‘lifetime’. In short, the periphrastic expression shows that the idea of ‘lifetime’ is encapsulated, which could lead the reader to the global inference: time reaps lives, i.e., it causes death, which is another possible reading for sickle, as we will see later.
Chapter 1. Opening Pandora’s box
5 Carina ah . . . it can be [she laughs] João no.. the sickle slowly reaps . . . sleep and dream . . . Sílvia What becomes thin drawn? Carina The sickle . . . João the sickle .. the sickle is hardly drawn thin . . . thin . . . ahead of the 10 phalanx of clouds . . Carina Is hardly drawn. . . João Thin-drawn Sílvia That’s why I abandoned the idea of time . . .
As we can see in Silvia’s speech (4), she uses the substitution of sickle by time, and that creates an incongruity (Cameron 2003: 12) whose resolution can lead to the construction of another reading. Then substitution is being used as a strategy to metacognitively monitor comprehension, both in order to verify Carina’s hypothesis for the reading of time and as a means of monitoring her search for another reading, as we will see in the next extract. Carina laughs and re-states her reading of time (5), and João repeats the words found in the text (6). Nonetheless, Sílvia returns to the issue at hand (7), by asking what becomes thin-drawn? i.e., time is an abstract concept and cannot be drawn, to which Carina replies with the literal reading: the sickle (8). João and Carina continue to think aloud (9–12), and Sílvia concludes her thought by stating that because of that (incongruity) she abandoned the idea of time (13). The group, however, leaves Sílvia’s last turn without a response, and the problem unanswered, moving to other aspects of the text.
Second reading – a riddle to be solved The group continues to discuss other aspects of the text during 145 turns, and it is only through intervention from the teacher that the group returns to the issue raised by Sílvia: Extract 5
1 Teacher You abandoned something ah . . Silvia said that . . . that idea of time . . . that ah . . . that Carina had brought up . . . you (silvia) said that you abandoned it . . . but you didn’t say why. Sílvia because he says . . . time is not drawn.. thin-drawn . . . especially . . . 5 thin . . this made me forget about this idea of time . . . Cristina The sickle that sleep and dream slowly reaps has just become thin-drawn Sílvia I’m under the impression that he wants to use an image there . . . João He uses an image with the word slowly. . . which is also about time. 10 Antonio but short time . . . Teacher And so? . . . what did you think it is? Sílvia Short melody . . . I put it like this . . .who is or what is this sickle that terminates sleep and dream and is thin-drawn ahead of clouds? . . . it
Mara Sophia Zanotto & Dieli Vesaro Palma
doesn’t seem to be the sun . . . in the morning . . . because . . . it 15 could be . . . Antonio I think time is quite strongly there Teacher let’s allow Silvia to finish . . . João I don’t think that . . . Sickle is really sickle, there is no interpretation beyond that . . . because as the bell is being tolled by someone . . . 20 whom we don’t know . . . it spreads that grievous song . . . within the twilight. . . the sickle is . . . cutting something . . . but it slowly harvests . . what? Sleep and dream . . . I mean.. it is there as a part of . . . do you understand? Of something that is thin-drawn . . . Sílvia well . . . to me this sickle is not clear . . . 25 Carina to me it is time going by . . . Sílvia yes . . . I thought of that . . . but . . . João It is an element that separates sleeping from dreaming . . . Sílvia I think that it is not . . . an element that keeps cutting sleep and dream . . it is not separating the two . . . 30 João no . . . it is not separating the two . . . it is rather, intercepting . . . it decreases the possibility of someone dreaming and . . . sleeping . . . Sílvia and who does that? Carina time . . . it goes by and you . . .??? Sílvia yes . . . but why does he say that time can be thin- drawn? 35 Carina it is thin-drawn. . . it doesn’t draw itself . . . it is thin-drawn . . . Sílvia but if it is hardly drawn or if it is thin-drawn, the word hardly is
determining something here . . .
After the teacher returns to the issue of the reading of sickle with the group (1–3), Sílvia also recovers her thought (4–5), stating that sickle cannot be time because time cannot draw itself. In the following turns (6–10), Cristina, Sílvia, João and Antonio think aloud about the problem and the idea of time is raised again. However, the teacher speaks to Sílvia (11), asking her what she thinks sickle is. Silvia then replies (12–14), formulating the problem by a new and interesting perspective: if time cannot draw itself, she says: Who is or what is this sickle that terminates sleep and dream and is thin-drawn ahead of clouds? In other words, Sílvia uses here an unexpected strategy of comprehension: she empties sickle of its reference and temporarily changes it in a deictic expression: who is or what is (the sickle) while its metaphoric reference is sought (see Zanotto 1990). She also raises the hypothesis of sickle being the sun, but refutes it immediately (13–14). She is then interrupted by Antonio who restates the idea of time (16). However, at this point, the teacher asks the group to allow Sílvia to complete her reasoning (17). João, though, does not listen and reaffirms his literal reading (18–19). Sílvia will still insist on the group, returning, several times, to the same problem (24, 26, 28, 29, 32 and 39) sometimes receiving Carina’s time reading as a reply
Chapter 1. Opening Pandora’s box
(25, 33), other times having João’s interpretative attempt, by means of the deictic strategy, in which he speaks of something (23) or of an element (27). At this moment, the thinking aloud process about the sickle is interrupted, and the group starts to think about the phalanx of clouds metaphor (flock of clouds in the translated version),15 which takes ten turns, after which Sílvia brings their attention back to the sickle:
Extract 6 1 Sílvia what confused me was this ‘thin-drawn’ . . . who is going to draw itself? It reaps and it draws itself ahead of the clouds? João ??? Carina the night is long . . . what was short was the twilight . . . the night is 5 long . . . and so . . . the time is going by very slowly . . . Teacher You are trying to solve a problem Carina we are indeed . . . (she laughs)
Silvia, therefore, puts the question of incongruity between time and thin-drawn forward again (1–2). What Silvia wants to say here is that time is abstract and cannot be drawn. Thus, it is necessary to seek a referent from another domain that can be drawn. The verb ‘to draw’ is therefore being used as a clue for another reading. It is interesting to note that Sílvia uses the deictic strategy again: who is going to draw itself? (1–2) On the other hand, Carina goes back to thinking aloud about time (4–5), and the teacher says that they are trying to solve a problem (6), to which Carina agrees by laughing (7). The group, however, goes back to the discussion of flock (phalanx) of clouds for the 13 turns that follow. Up to this point in the discussion, we can notice that the group is not taking Silvia’s proposition into account. They do not listen to Silvia, worried as they are with their own voices echoing inside their minds. Carina, for example, keeps bringing back her reading of time, while João maintains his literal reading. Besides, sometimes the group digresses and discusses other aspects of the text. Thus, until this moment, the group had been holding a monological kind of discussion, while Silvia was dialogically trying to construct meaning from her colleagues’ words, adequately acting as a mediator. They seemed to be functioning within the frame of the conduit metaphor (Reddy 1979), i.e., within the principle of one sole and pre-determined reading, by casting out 15. The translated version uses flock of clouds, because the translation is intended to be poetic, rather than being word for word. However, in order to achieve this, the original meaning was altered, since the expression in Portuguese would have corresponded to phalanx of clouds. Thus, the students were focusing on the phalanx metaphor when thinking aloud. In order for readers to better understand the students’ discussion, we have included a word for word translation of the verses in which sickle appears: “the sickle that slowly reaps sleep and dream, is barely drawn, thin, ahead of the phalanx of clouds forgotten to pass.”
Mara Sophia Zanotto & Dieli Vesaro Palma
the reading of others since this would imply accepting another possible reading. At this point, the teacher comes to Silvia’s assistance:
Extract 7
1 Teacher but I think Sílvia is right . . . something has become thin-drawn . . . I don’t mean that the interpretation of time is wrong . . . but something has become thin-drawn . . . Sílvia yes . . . 5 Carina It’s the sickle! Teacher her question . . . is valid . . . . Carina It is thin-drawn . . . this is what I thought, a phalanx full of arms . . . an army full of arms . . . a sickle? It is hardly thin-drawn . . . a sickle . . . it isn’t . . . easily told apart . . . it is only one . . . 10 Sílvia so who is this sickle? All Laugh Cristina that’s a problem, isn’t it? Teacher I think it would be interesting if you solved it . . . Sílvia now . . . it is interesting that phalanx (falange) and sickle (alfanje) are 15 written with the same characters João but that’s not a problem . . .??? Carina and sound the same too. . . interesting this Cristina yes, it is.. João I don’t think so . . . let’s see . . . the bell is an object . . . that exists . . . 20 right? The sickle is precisely an object that exists . . . I think there isn’t any interpretation beyond that . . . besides what’s there . . . do you understand? Sílvia Right . . . João It speaks of someone tolling the bell . . . there is someone cutting 25 some . . . something . . . but so finely cutting it that one can hardly notice . . . due to the presence of the phalanx of clouds . . . thin . . . why thin? Because it is all a dream . . . Sílvia Is it possible that the sun is in the shape of a sickle? It isn’t, is it? João The sun . . . at least . . . 30 Teacher You’re getting warm . . . João Are we getting warm, teacher? Cristina No . . . the night . . . the long night. . . Teacher The Sun at night, folks? That’s not possible. Emília The moon, folks! . . it’s the moon there! . . . in the shape of the 35 moon All Yeah! . . you got it! . . it’s the moon! . . . the new moon [laughing a lot] João ah . . . the sickle . . . Teacher That’s it . . . you found it! Carina That is hardly drawn . . . 40 All ahh! . . that’s it . . . (laugh and clap)
Chapter 1. Opening Pandora’s box
45
João Hey! . . . look . . . of course! . . . of course! . . . the last quarter . . . Sílvia It is the moon, everyone . . . (laugh) look . . . Antonio It could be the first or the last quarter Teacher Isn’t it an enigma? Sílvia It’s delightful, isn’t it? Teacher it is a great enigma and . . . so . . . what is the link between the
Emília 50 Carina
moon and time? What do you say? Is it the moon or the time? . . . which one do you accept? Or do you accept both? both . . . especially because with that moon . . . which is a phase of the moon that we relate to time . . . the moon goes past the sky, doesn’t it? When it finishes . . . the
night is over . . . it represents . . . it is the symbol of time . . . still, I keep my opinion . . .
The teacher goes back to the problem that had been left on hold, by saying that Sílvia was right in looking for another interpretation of time (1–3). Sílvia agrees with her (4), but Carina replies interrupting the teacher and stating that it was indeed the sickle that was being drawn (5). She carries on with her thought (7–9), trying to justify that the sickle is being drawn by opposition with the word phalanx, which is an army of weapons – the sickle being just one of them, thin-drawn. Sílvia, nonetheless, returns to the matter making everyone laugh because of her persistence: so who is this sickle? (10) Cristina’s response strengthens the idea that it is indeed a problem (12) while the teacher encourages them to solve it (13). However, in the following four turns (14–18), Sílvia, Joa~o Carina and Cristina think aloud of the Portuguese sound similarity between alfanje (sickle) and falange (phalanx of clouds, translated as flock of clouds in the poem), and João seemingly wants to say (16) that this is not a problem; that the real problem is being left aside. Returning to the problem at hand, João pursues his literal reading of sickle, reiterating that there is no other understanding apart from that (19–21). Sílvia, apparently agrees (23) with João, and he continues to think aloud of the meaning that sickle (24–27) might have, by using a deictic strategy – since he replaces sickle for someone – which seems to indicate that he is not quite sure about what or who the sickle might be. Sílvia raises the hypothesis of the sun being the sickle (28), but she herself dismisses it. João begins a thought that is left incomplete (29), and the teacher says that they are getting warm (30), i.e., they are getting closer to solving the enigma, to which João, feeling encouraged, asks for confirmation that they are really getting warm (31). Cristina backs Sílvia, by saying that the text mentions “a long-lasting night” (32). At this point, the teacher strengthens her point by saying: The Sun at night, folks? That’s not possible (33). Therefore, they could infer that sickle could not be the sun. At this moment, the solution comes from Emilia (34) – who had, silently, been tagging along with the discussion and now was showing that she had been following everything that had been said: the moon, folks! It’s the moon! In the shape of the moon.
Mara Sophia Zanotto & Dieli Vesaro Palma
The reading of sickle as ‘moon’ was therefore co-constructed by this group (especially by Sílvia, Emília and Cristina), through bottom-up processing, in which the verb “to draw” was an important clue for the readers to complete the mapping across domains (Lakoff 1993) between sickle and moon. The construction process was similar to that of a solution of an enigma16 (at the end of the activity, Sílvia made the comment that it was a delightful enigma (45), and her peers seem to think similarly, if we consider their laughing and clapping (40). After that they discuss whether the moon would have been a new moon, a first or a last quarter. This metaphor functioned as a novel metaphor for these readers.17 João and Carina accepted the reading of moon, but Carina also maintained her interpretation of time and João gave up his literal reading. In short, the group GB accepted two different readings: moon and time and rejected the literal meaning. It is important to note that the process by which Silvia and the group coconstructed the reading of sickle as moon was an on-line process, and that Silvia used several strategies: – At first, she substituted sickle by time, and realised the incongruity between time and drawing; which led her to the rejection of the time reading hypothesis. This fact shows that the substitution was a strategy to verify if this hypothesis would solve the incongruity – which leads us to the conclusion that incongruity, apart from being a necessary condition for the identification of the metaphor (Cameron 2003), is also important guide for interpretation. – Then, she replaced sickle by the deictic expressions who or what temporarily erasing its literal reference, while its metaphoric reference was sought; – After that, she replaced it by the ‘sun’ – another reading hypothesis, but she herself rejected that, and so did the group, because the poem is about the twilight, the beginning of night. We thus notice that textual coherence and relevance (Sperber & Wilson 1986/1995) are important in the evaluation of readings, and can constitute an important path to investigate the question of acceptance or rejection of the multiple readings.
16. “the interpretation of this metaphor is similar to that of an interpretive guessing game or riddle (Georges & Dundes 1963) that, incidentally, is essentially a metaphoric game: who or what is it (sickle is emptied of its reference and works like a deictic) Thin-drawn in the sky at night (clues of place and time) Which is sickle shaped ( similarity) But is not a sickle? (dissimilarity) (Zanotto 1998: 30) The answer is the moon (solution: metaphorical reference)” 17. However, in the literature, it is a metaphor that has been used quite often. Kövecses (2002: 49) gives an example of a sonnet by Shakespeare in which this metaphor occurs. Love’s not Time’s fool, though rose lips and cheeks/within his bending sickle’s compass come. (Shakespeare, Sonnet 116)
Chapter 1. Opening Pandora’s box
–
Finally, Emilia, based on the whole process that had been constructed by Silvia, came up with the solution in an authentic process of co-construction (Jacoby & Ochs, 1995) of reading.
Another important aspect is that the reading hypothesis of moon was constructed from a similar trait (the shape) and from two non-similar traits, which were the result of metonymic inferencing for the place (the sky) and the time (at night) in which the sickle is placed. Mapping per se was finalised after the students explored similarities between sickle and moon, discussing whether it was a first or last quarter moon.
Third reading - death After the construction of moon and time by the group, the professor synthesised what had been said and added a third possible reading: Extract 8 1 Teacher There is something else . . . then . . . the sickle as a moon . . . it’s a metaphor, isn’t it? There is a space of intersection between them . . . the form, right? And . . . perhaps in other aspects too . . . what occurs to me now is the form . . . and Silvia deducted . . . that it was 5 something that had been there . . . because it was thin-drawn ahead of the phalanx of clouds . . . it was finely drawn . . . in the sky. . she thought. . she thought . . . is it the sun? No, it can’t be. . so I gave you a clue . . . sun. . at night? Impossible! It’s the moon. . which is to say that you were very close to the moon . . . now . . . the moon . . . 10 crossing the night . . . or, in other words.. reaping dream and sleep. . this takes us to the idea of time . . right? Now there is a link there João . . . temporal . . . Teacher a link . . . I don’t know if metaphoric or metonymic . . . which
15 20
is the moon . . . we need to think a little further to see . . . I know that the moon . . . slowly passing through the night . . takes us to . . perhaps it is another metonymy here . . but, in any case, it
forms two levels . . and time passing by . . . leads to death . . . which was the other part that you . . justified by speaking of the sickle, because you recalled that the sickle is the symbol of death . . . also . . . one can go straight to (to that reading)
In lines 1–11, the teacher synthesises the processes of the readings of moon and time and then constructs the reading of death as a consequence of time passing by (lines 17–20), thus using the reading of time as support for the construction of this new reading. In this case, we could say that there was an inference based on a relation of cause and effect, though it is not a metonymy within the text – and this is a different case from when the reader sees sickle standing for death, based on the metonymic
Mara Sophia Zanotto & Dieli Vesaro Palma
relationship of instrument for agent. In this case, we could say that the sickle represents a metonymy in the text.
Discussion of the readings of the first phase Table 1 summarises the readings of the first phase and shows the processes, the textual clues and the authorship for the readings. It also allows us to see the complexity of the phenomenon that we initially thought might be ‘a metaphor’. In fact, what we have is a network of metonymic and metaphoric readings with semantic relations between them. Simply calling this phenomenon ‘a metaphor’ does not seem to be appropriate. What is its identity then? How can we define this metonymic and metaphorical network? These questions motivated Zanotto to seek theories and a methodology for this investigation. Indeterminacy, for example, was not the focus of her research work; but the search for theories that might explain the phenomenon, however, led her to the theoreticians who dealt with this matter. Nonetheless, this search led to the understanding that there were no specific studies on metaphorical indeterminacy, thus, the challenge remained to carry out further empirical research that might contribute to the elaboration of theories around metaphoric indeterminacy. Comparison of data from groups GA and GB showed that the reading of time was possible prior to that of the moon, due to top-down processing activated by the verb reaps, which was the most relevant clue (see Cavalcanti 1989) for Carina. Besides, we also had the reading of death, which had not appeared in group GA. What if other groups experienced the thinking-aloud in group of the same text? Would we have further surprises? It was then that we had the idea of designing a collective case study to check if other surprises might occur, or if the readings would begin to repeat. Thus, data from Zanotto & Ricciardi’s (1984) and Zanotto’s (1995, 1998) works were revisited, becoming part of the first phase, and Palma was invited to take part in the second phase.
Second phase In this phase, the data analysis focuses only on the readings validated by the groups – and for time and space constraints, it will not be as detailed as the analysis of first phase.
Group 2 The readings of the first three groups found in the second phase, G1, G2 and G3, who experienced the Group Think-Aloud in a course taught by Zanotto, were not mediated by the teacher. However, she later ran retrospective interviews with them, and requested them to write diaries. What was noticed was that, without the teacher’s mediation, students simply put forward their individual readings to the group, rather than working dialogically. For this reason, we cannot say that there was co-construction, nor
Chapter 1. Opening Pandora’s box
Table 1. Synthesis of the first phase readings Group Participants Readings Processes GA Zanotto and sickle Ricciardi as moon moon is a reaper moon for time time is a reaper GB 6 students time is a and Zanotto reaper sickle as moon sickle is death
Textual clues
Bottom up by Sleep and dream analogical Long-lasting night reasoning untroubled clouds supported by Shadow, evening, metonymic crepuscular processing Bell tolls Bottom up by reaps analogical reasoning Metonymic Sickle processing Bottom up by analogical reaps reasoning Top down reconstructed Bell, short melody, by analogical long-lasting night, reasoning evening, supported by crepuscular, metonymic sleep, dream processing Bottom up by Has just become analogical thin-drawn reasoning and metonymic processing of clues for the metaphorical interpretation Bottom up by Sickle reaps metonymical reasoning based on instrument for agent and on cause for effect relation
Author of reading Zanotto
Ricciardi Ricciardi Ricciardi Student Carina
Co-construction of Sílvia and Emília
Zanotto
Mara Sophia Zanotto & Dieli Vesaro Palma
negotiation of readings. They did not discuss the readings put forth by their peers, each one remaining with their own understanding of the text. In the interviews and in the general discussion between the groups, they were later requested to think about the readings that had been constructed. We will only consider data from group G2 because they constructed the only reading accepted as valid in the general discussion. This group was formed by five students: Cecilia, Paula, Sueli, Pedro and Juliana. They constructed three different readings for sickle: death, insomnia and the shadow of a rock in the shape of a sickle. The first was accepted by the group and the other two were rejected, assessed as unwarranted. The reading hypothesis of death was constructed by Pedro, described when he produced the following retrospection, in the group discussion:
Extract 9 Pedro Here you have a metaphor of Death . . which . . which is the . . sickle . . it’s Moura-Torta [character] . . . from children’s stories . . . the woman that came with a sickle . . . to reap lives . . . so the sequence is a metaphor of death . . . that thing of love and death . . .
Pedro presents an equivalence: the sickle . . . it’s Moura-Torta from children’s stories, and explains it: the woman that came with a sickle to reap lives. This explanation is evidence that Pedro had constructed two types of inferences: one based on contiguity: (1) the sickle is an instrument used by Moura-Torta (a personification of death); (2) and another inference based on similarity: “reaps sleep and dream” reminds him that Moura-Torta reaps lives, i.e., death is a reaper (Lakoff & Turner 1989). Therefore, we have metonymic as well as metaphoric processing in this reading. As he reports it, the metonymy is the first one to appear. The readings that were later rejected by the group were of “a shadow of a rock which constructs a sickle shape” and “insomnia”. Pedro thought that the shadow of a rock would make a thin drawing of a sickle, but in the general discussion with the other groups, this reading was left aside. The reading of death, made by Pedro, was not easily accepted and was refuted by Cecilia, who proposed the idea of insomnia.18 But this reading was also refuted in the general discussion since the students concluded that insomnia did not have any relationship of similarity, nor of contiguity, or any other relationship, with the sickle. Group 1 did not construct any reading that was accepted. Given the time when the poem was published, the students related the sickle with military power of the
18. Cecilia: I think that it is insomnia . . . that long night that never ends . . . ( ) and you don’t dream don’t dream . . . and you don’t sleep . . . for something that . . . that doesn’t . . . that deprives you . . . of thatmoment of yours, do you understand?
Chapter 1. Opening Pandora’s box
dictatorship period in Brazil, but in the general discussion, they rejected their own reading, saying that the Brazilian army does not use a sickle as weapon. Group 3 did not construct any reading for sickle.
Group G4 The group coordinated by Palma was composed of the following students: Valéria, Aurora, Silmara, Ana Carla, Eugênia, Angélica and Amarílis. During the group discussion, the teacher acted as a mediator, coordinating the interventions and asking questions, so that the group could deepen their reflections about the sickle metaphor. Two interpretations were given for sickle: moon and death. First reading – moon. Extract 10 below is preceded by a long discussion whose trigger for the construction of the meaning of moon was given by Amarílis who insisted on analysing the lines: The sickle that sleep and dream slowly reaps has just become thin-drawn while keeping up a dialogue with the mediator. After this dialogue, the teacher synthesizes what had been said and asks Amarílis what the sickle might be. The student answers by presenting the group a reading hypothesis (see below line1): Extract 10 1 Amarilis would the sickle be the sun? ?? how come? Amarilis the sun . . . in order to have shadows you need the light of the sun . . . Ana Carla but look at what he says short melody long-lasting night. . .would 5 this . . . this short melody and this long-lasting night be referring to the sickle? Amarilis no, it is referring to what was said before . . that which tolls . . . Ana Carla the sickle . . . thinly drawn ..the sickle . . . Silmara when I read the text I couldn’t do it . . . 10 Amarilis is it the moon? The moon is thin . . .
In light of the skepticism shown by a voice which is not identified (2), Amarílis seeks to justify the solution presented, by relating the hypothesis made to a term found in the poem, i.e., the ‘shadow’, thus establishing a retroactive relation (Palma 1998) between sickle and shadow (3), in the sense that she goes back to the text seeking clues to decipher the metaphor. Ana Carla follows Amarílis’ strategy and relates short melody long-lasting night to sickle (4–5). However Amarílis answers (7) that short melody longlasting night is related to a bell tolls. After that Ana Carla insists on the drawing of sickle (8) and is interrupted by Silmara (9), who does not actually finish her explanation of her line of thought because she is interrupted by Amarilis. She brings the hypothesis of sickle as the moon (10), based on the similarity between the shape of sickle and moon, and on the verb to draw, that was recovered by Ana Carla. It is important to highlight
Mara Sophia Zanotto & Dieli Vesaro Palma
that, before that, the group had discussed the curved format of the sickle, and Angélica had even drawn it on the board, in a similar form as the first quarter moon. Thus, in constructing her reading of sickle as moon, Amarílis seems to have related the drawing on the board to the idea of sickle being thinly drawn, since, by deciphering the enigma, she also says the moon is thin, and draws the moon in the air, pointing to what was represented on the board. The group carried on in their bottom-up processing, in which the verb to draw was a fundamental element in establishing a cross-domain mapping (Lakoff 1993) between sickle and moon. Thus, it was through a process of analogical reasoning that the reader constructed this meaning of sickle, which can be seen in the use of the adjective thin applied to moon, closely bound to its first quarter form. Second reading – death. The reading of sickle as death was constructed by Silmara through top-down processing. The following excerpt shows how this meaning was constructed:
Extract 11
1 Angélica A bit of our hopes, of positive expectations in life..as if it refused us the very right to dream which is something that human beings need in order to live. . .those who do not dream, do not live, right?. . . Silmara Then we would have something like the twilight which is a kind of 5 death, isn’t it? . . . and beside that there is (?? ) trying the attempt to cut, there is the image of death that we have socially . . . the card with a hood, the dark face and a sickle . . . isn’t it? So this image came to my mind now, though it was very abstract, but now it is closer to the shape . . . of actually cutting . . . expectation of dream 10 and of hope . . .
It is with Angélica’s turn, stating that those who do not dream (1–3) do not live, that Silmara attributes to the sickle the idea of death (4–7), supported by the metaphor twilight as death (4–5). By thinking of her background knowledge – the Tarot card, with elements that compose its image such as the hood, the dark face and the sickle – she constructs this reading as a social representation of death. She also highlights that, initially, this image was presented in abstract form, but now, in view of the verb to reap, it has become more concrete (7–10). By bringing death to this social archetype, Silmara activates the conceptual metaphor death is a reaper (Lakoff & Turner 1989), thus constructing this reading, through top-down processing, highlighting the allegories of death and sickle (or scythe) (Cf. Cirlot 2005: 259). In short, in G4, there were two readings of sickle: sickle as moon and death is a reaper. The first occurred through bottom-up processing – with similar meaning construction to that of GB group, in the first phase. The second reading occurred through top-down processing, as had happened to the reading time as a reaper by Carina, in group GB.
Chapter 1. Opening Pandora’s box
Discussion of the readings of the second phase In the second phase, three readings were repeated: sickle for death, sickle is moon and death is a reaper. The only reading constructed by bottom-up processing and the teacher’s mediation was sickle is moon. The other readings – including both the ones accepted by the group and the ones rejected by them – were constructed by top-down processing. It seems that the students, without the teacher’s mediation, have difficulty in carrying out bottom-up processing when reading the poetic text. In sum, the data found in this phase did not present much novelty, except for the top-down processing in the reading of death is a reaper. Table 2, below, provides an overview of the readings of the second phase Table 2. Synthesis of second phase readings Group Participants Readings Processes G1 Five students — G2 Five students death is a reaper sickle for death G3 Five students — G4+Teacher 7 students+ sickle is teacher moon death is a reaper
Textual clues
— — Top down by Sickle reaps activation of conceptual metaphor and metonymic processing — — Bottom up Sickle, thin-drawn process Sickle, reaps Top down process
Authors — Pedro
— Amarilis and Anacarla Silmara
General discussion The methodology used – Group Think Aloud technique in a design of collective case study – to focus on multiple readings of ‘the metaphor’ resulted in the generation of valuable data which, besides shedding light on some mechanisms of the comprehension process, made it possible to discuss the very identification of the figure, considering that what emerges is a network of readings interwoven by metaphoric and metonymic threads. That which one might classify simply as ‘a metaphor’, proved to be a more complex matter, a polysemous amalgam of metaphors and metonymies. In other words, confronting metaphor in use leads the researcher to face these types of phenomena which constitute a challenge for metaphors scholars to explain. In short, the sickle is like a chameleon – depending on the processing employed by the reader, it can be a different metaphor or a different metonymy. Therefore, focusing on the process, which was the intention stated in the first research question (how do readers co-construct the multiple readings of the focal metaphorical
Mara Sophia Zanotto & Dieli Vesaro Palma
expression?), has led us to notice the variety of processes involved in the construction of the multiple readings, and has made us confront an unexpected problem: how to count and individuate the readings in this amalgam of metaphoric and metonymic threads? At the end of the first phase, we thought there had been three different readings, since we were only taking into account the different target domains that had appeared in the readings: moon-time-death. However, taking into account the different construction processes, we could think of more readings, for example, sickle as moon (based on the form similarity) or moon as a reaper (based on the function). The same can be said about sickle for time (based on the metonymic relation of instrument for agent) and time as a reaper (based on the function). Therefore, the prior position of counting only three readings seems to be reductive from the point of view of meaning. Nonetheless, if we were to consider the readings organised in a network whose nodes would be moon, time, and death, would the nuances created by metonymic or metaphoric processes be the links between these nodes? This is one question that we leave open for future research. The discussion of individuation of readings makes evident, in fact, that accounting for the construction process can shed light – but also make more complex – the discussion about the product of interpretation, i.e., the multiple readings. We therefore leave another question unsettled: Do the different processes of readings construction have influence on the nature of the readings or on the criteria for individuating the readings? Apart from this, the focus on the process provided different evidence on bottom-up and top-down processing. The bottom-up processing, analysed in depth in group GB’s reading, in the first phase, showed several strategies used in the construction of the reading of moon by Silvia and the group: substitution strategy, the role of incongruity in the search for another reading, deictic strategy, and textual strategies for searching textual clues for the construction and evaluation of reading hypothesis. On the other hand, the readings carried out through top-down processing constituted evidence of the activation of conceptual metaphors by the readers: time as a reaper and death as a reaper. The activation of conceptual metaphors or of contiguity relations, having previous knowledge as a basis seems to be related to the reader’s most relevant textual clue. This leads us to the discussion about the role played by the context. In relation to the role of context, the data analysis suggests that depending on the textual clue of greater relevance (Cavalcanti 1989)19 for the reader, and depending on
19. In her study on the reader’s pragmatic interaction with the text, Cavalcanti adopts the principle of relevance as central to reading comprehension and, considering the difference in
Chapter 1. Opening Pandora’s box
his previous knowledge, the reading is processed differently. If sickle is the strongest clue, for example, it is likely that the metonymic reading based on the relation instrument for agent takes place. In this case, we have sickle for time or for death, but immediately the reader associates time or death to reaps, and then, this time or death, inferred by a metonymic process, becomes a metaphor time is a reaper or death is a reaper. Thus, metonymic and metaphoric processing come together in processing. If the strongest clue is reaps, the more probable readings are the metaphoric time is a reaper, as in Carina’s reading (group GB), or death is a reaper, as in Pedro’s reading (group G2), or even in Silmara’s case (group G4). If, however, the strongest clue is thin-drawn, the most likely reading is moon, constructed metaphorically. We can, therefore, see that the context allowed for several readings, but it also invalidated others, such as those of ‘the sun’, ‘insomnia’ and ‘shadow of a rock which constructs a sickle’. In other words, the context can broaden readings, but also limits them. As for the coincidence in the readings, the lack of variation shows that meanings constructed, even if new to some of the readers, have culturally been linked to the focused figure for millennia, so much that they are found in symbols dictionaries (Chevalier & Gheerbrandt 1982/1997; Lurker 1988/1997). Therefore, reading variations were not as numerous as expected. In fact, the ‘multiple readings’ are quite stable. According to Chevalier and Gheerbrant, in their Dictionary of Symbols, the moon in its different phases has been used by people to mark time and death: The Moon is a symbol of biologic rhythms, (. . .) of living time of which it is a measure due to its successive and regular phases. The Moon is also the first dead. During three nights, in each lunar month, it is as if dead, it disappears . . . then it re-appears and grows in brightness. In the same sense, it is believed that the dead acquire a new modality of existence. The moon is for men (sic) a symbol of this passage from life to death and from death to life; it is even considered by many peoples as the place of passage, as in the underground. (pp. 561–562)
The moon, in the first quarter phase, which symbolizes death and re-birth, is often also represented in literature by the scythe or by the sickle, which, in turn, are symbols of time and death, as discussed by Lurker (1997), in his Dictionary of Symbols: “sickle and scythe are symbols used for reaping, of time and death, and for this reason they are attributes of Saturn-Khronos and of the personification of death” (p. 275).20 This reinforces Gibb’s claim (1999b: 146) that we “should think about metaphor and its relation to thought as cognitive webs that extend beyond individual minds and are spread out into the cultural world”. points of view, she distinguishes the reader’s and the author’s relevance in what she calls reader relevance and author salience. 20. In this dictionary, one can also find the explanation that Khronos, God of Time is represented with a scythe, and that Saturn /Crono is represented with a sickle.
Mara Sophia Zanotto & Dieli Vesaro Palma
As for the second question – what type of indeterminacy characterizes the multiple readings? – we can say that answering it is very complex and challenging. This is so because, if there is already a lot of discussion about the indeterminacy of literal language, that of metaphor will provoke even more discussion. To initiate the debate, we will pose some questions because we are not ready to provide answers as yet. If we consider that polysemy involves compatible meanings that are semantically related (Moura 2001), we can then consider that this is a case of polysemy (regardless of whether it is represented by three or more readings). However, if we consider that polysemy requires meanings to be lexicalised (i.e., that they are found in language dictionaries), and that the readings that occur for sickle are not lexicalized, we would have to abandon this label. Nonetheless, we could question this requirement for lexicalisation in the case of the metaphor, since dictionaries do not generally present an appropriate treatment of metaphoric meanings. What we have at present are dictionaries of symbols, and they show us that the meanings attributed to sickle and moon have culturally been linked to them for millennia. Besides, if we draw a comparison with the literal language in which words used to name artifacts allow us to think in terms of form and function and, for this reason, are polysemic (Moura, personal communication), we could think that, just like sickle in literal language is polysemic because it allows readings as to the form and the function, the word sickle in the poem is also polysemic because it allows for metaphorical readings in terms of the form: sickle as moon, and in terms of the function: moon as a reaper, time as a reaper, death as a reaper. And we still need to add that sickle, being an artifact, also allows for metonymic readings based on the instrument for agent relation: sickle for time, sickle for death. This reasoning, based on the comparison with literal language can shed light on the question of individuation of readings, besides reinforcing our hypothesis that metaphoric language will display the same types of indeterminacy as literal language. However, the nature of this metaphoric polysemy deserves deeper discussion, together with the issue of individuation of readings and that of the interaction of metaphor and metonymy. Although it holds similarity with literal polysemy of sickle, a discussion on the specificities of metaphorical polysemy, for example, is still needed – an interesting characteristic of this polysemy is that all meanings are valid in the context, i.e., it does not become precisified. In the end, we raised more questions than answers due to the immense complexity of the issue of metaphoric indeterminacy – a challenging and bewitching subject. . .
Acknowledgments This research is supported by grant from CNPq (Grant 301053/2005–7), Brazil, awarded to Mara Sophia Zanotto. We would also like to thank Ray Gibbs, Lynne Cameron, Marilda Cavalcanti, Heronides Moura and the anonymous reviewer for their invaluable
Chapter 1. Opening Pandora’s box
comments, and the students who participated collaboratively in the data generation. We also would like to thank Sandra Madureira for having translated the poem.
References Black, Max (1979/1993). More about metaphor. In A. Ortony (Ed.), Metaphor and thought (2nd and revised edition) (19–41). New York: Cambridge University Press. Bloome, David (1983). Reading as a social process. Advance in reading/language research, 2, 165–195. Bloome, David (1993). Necessary indeterminacy and the microethnographic study of reading as a social process. Journal of Research in Reading, 16, 98–111. Brown, Carol S. & Susan L. Lytle (1988). Merging assessment and instruction: protocols in the classroom. In S.M. Glaser, L.W. Searfoss & L.M. Gentile (eds.), Reexamining reading diagnosis: New trends and procedures (94–102). Newark, NJ: International Reading Association. Cameron, Lynne (1999). Operationalising metaphor for applied linguistics. In L. Cameron & G. Low (Eds.), Researching and applying metaphor (3–28). Cambridge: Cambridge University Press. Cameron, Lynne (2002). Metaphors in the learning of science: A discourse focus. British Educational Research Journal, 28, 673–688. Cameron, Lynne (2003). Metaphor in educational discourse. London: Continuum. Cavalcanti, Marilda C. (1989). Interação leitor-texto – Aspectos de interpretação pragmática [Reader-Text Interaction – Aspects of Pragmatic Interpretation] Campinas: Editora da UNICAMP. Cavalcanti, Marilda C. & Mara S. Zanotto (1994). Introspection in applied linguistics: Metaresearch on verbal protocols. In L. Barbara & M. Scott (Eds.), Reflections on language learning (148–156). Clevedon: Multilingual Matters. Chevalier, Jean & Alain Gheerbrant (1982/1997). Dicionário de símbolos (Dictionnaire des symboles) Rio de Janeiro: José Olympio (ed. Robert Laffont & ed. Jupiter). Cirlot, Juan-Eduardo (2005). Dicionário de símbolos (Dictionary of symbols). São Paulo: Centauro. Denzin, Norman K. & Yvonna S. Lincoln (Eds.) (1998). The landscape of qualitative research. London: Sage. Ericsson, K. Anders & Herbert A. Simon (1984). Protocol analysis. Cambridge, Mass: MIT Press. Erickson, Frederick (1986). Qualitative methods in research on teaching. In M.C. Wittrock (Ed.), Handbook of research on teaching (119–161). New York: MacMillan. Erickson, Frederick (2001). Prefácio (Preface) In M.I.P. Cox & A.A. de Assis-Peterson (Eds.), Cenas de sala de aula [Classroom scenes] (9–17). Campinas: Mercado de Letras. Fauconnier, Gilles & Mark Turner (1999). Metonymy and conceptual integration. In K. Panther & G. Radden (Eds.), Metonymy in language and thought (77–90). Amsterdam/Philadelphia: John Benjamins. Ferreira, Aurélio B. de H. (1999) Novo Aurélio – O dicionário de língua Portuguesa. [New Aurélio – The dictionary of Portuguese language] Rio de Janeiro: Editora Nova Fronteira. Georges, Robert A. & Alan Dundes (1963). Toward a structural definition of the riddle. Journal of American Folklore, 76, 111–118.
Mara Sophia Zanotto & Dieli Vesaro Palma Gibbs, Raymond W. (1987). What does it mean to say that a metaphor has been understood? In R.E. Haskell (Ed.), Cognition and symbolic structures: The psychology of metaphoric transformation (31–48). Norwood, New Jersey: Ablex Publishing Corporation. Gibbs, Raymond W. (1994). The poetics of mind: Figurative thought, language and understanding. New York: Cambridge University Press. Gibbs, Raymond W. (1999a). Researching metaphor. In L. Cameron, and G. Low (Eds.), Researching and applying metaphor (29–47). New York: Cambridge University Press. Gibbs, Raymond W. (1999b). Taking metaphor out of our heads and putting it into the cultural world. In R.W. Gibbs & G. Steen (Eds.), Metaphor in cognitive linguistics (145–166). Amsterdam: John Benjamins. Jacoby, Sally & Elinor Ochs (1995). Co-construction: An introduction. Research on Language and Social Interaction, 28, 171–183. Johnson, Mark (1987). The body in the mind: The bodily basis of meaning, imagination and reason. Chicago, IL: University of Chicago Press. Kövecses, Zoltán (2002). Metaphor: A practical introduction. Oxford: Oxford University Press. Lakoff, George & Mark Johnson (1980). Metaphors we live by. Chicago: The University of Chicago Press. Lakoff, George & Mark Turner (1989). More than cool reason. A field guide to poetic metaphor. Chicago: The University of Chicago Press. Lakoff, George (1993). The contemporary theory of metaphor. In A. Ortony (Ed.), Metaphor and thought (2nd edition) (202–251). New York: Cambridge University Press. Lurker, Manfred (1997/1988). Dicionário de simbologia [Wörterbuch Der Symbolik. Stuttgart: Alfred Kröner Vorlag]. São Paulo: Martins Fontes. Mason, Jennifer (1996). Qualitative researching. London: Sage. Martin, Robert (1992). Pour une logique du sens. Paris: PUF. Moura, Heronides M.M. (1998). Indeterminação na língua e na poesia [Indeterminacy in language and in poetry]. Revista da ANPOLL, 5, 107–120. Moura, Heronides M.M. (1999). Indeterminação e contradição: Um caso de conflito de papéis sociais [Indeterminacy and contradiction: a case of social role conflicts]. Intercâmbio VIII, 399–408. Moura, Heronides M.M. (2001). A determinação de sentidos lexicais no contexto [Determination of lexical meanings in the Context]. Cadernos lingüísticos, 41, 111–125. Nardi, Maria Isabel A. (1999). A metáfora e a prática de leitura como evento Social: instrumentos do pensar a biblioteconomia do futuro [Metaphor and reading practice as a social event: Tools for thinking future library science] Pontifícia Universidade Católica de São Paulo, São Paulo, Unpublished PhD thesis. Palma, Dieli. V. (1998). A leitura do poético e as figuras de pensamento por oposição: Caminhos e descaminhos de paradigmas na modernidade. [Poetic reading and thinking figures by opposition: Paths unfolding in modern paradigms]. Pontifícia Universidade Católica de São Paulo, São Paulo, Unpublished PhD thesis. Pinkal, Manfred (1995). Logic and lexicon: The semantics of the indefinite. Netherlands: Kluwer Academic Publishers. Poesio, Massimo (1995). Semantic ambiguity and perceived ambiguity. In K. van Deemter and S. Peters (Eds.), Semantic ambiguity and underspecification. Stanford: CSLI Publications. Putnam, Hilary (1981). Reason, truth and history. Cambridge: Cambridge University Press. Rommetveit, Ragnar (1985). Language acquisition as increasing linguistic structuring of experience and symbolic behavior control. In J.V. Wertsch (Ed.), Culture, communication, and cognition: Vygotskian perspectives (183–204). Cambridge: CUP.
Chapter 1. Opening Pandora’s box Sperber, Dan & Deirdre Wilson (1995) Relevance: Communication & cognition (2nd edition). Oxford: Blackwell. Stake, Robert E. (1998). Case studies. In N.K. Denzin & Y.S. Lincoln (Eds.), Strategies of qualitative inquiry (86–109). London: Sage. Steen, Gerard (1994). Understanding metaphor in literature: An empirical approach. London: Longman. Steen, G. (1999). Metaphor and discourse: Towards a linguistic checklist for metaphor analysis. In L. Cameron & G. Low (Eds.), Researching and applying metaphor (81–104). Cambridge: Cambridge University Press. Vieira, Josalba R. (1999). Metáforas e conflitos: A leitura de poesia e a discussão em grupo na sala de aula de Inglês como Literatura estrangeira [Metaphors and conflicts: Poetry reading and group discussion in an English as foreign literature class]. UNICAMP, São Paulo, PhD thesis. Wierzbicka, Anna (1996). Semantics: Primes and universals. Oxford: Oxford University Press. Zanotto, Mara S. (1990). A natureza dêitica da metáfora alegórica [The deictic nature of allegoric metaphor]. Linguagem 7, 81–85. Zanotto, Mara S. (1992). O processo de compreensão da metáfora na formação dos Professores de língua materna [The process of understanding metaphors in the teacher education for teachers of the mother tongue]. In M.S. Zanotto and M.A. A. Celani (Eds.), Lingüística aplicada: da aplicação da lingüística à lingüística transdisciplinar [Applied linguistics:From application of linguistics to the crossdisciplinary linguistics] (233–246). São Paulo, EDUC. Zanotto, Mara S. (1995). Metáfora, cognição e ensino de leitura [Metaphor, cognition, and the teaching of reading]. D.E.L.T.A., 11, 241–254. Zanotto, Mara S. (1997). A leitura como evento social para um enfoque humanístico do ensino de línguas [Reading as a social event for a humanistic approach to language teaching]. Paper presented at XIX Congresso Mundial da FIPLV, Recife, Brazil. Zanotto, Mara S. (1998). A construção e a indeterminação do significado metafórico no evento social de leitura [The construction and indeterminacy of metaphorical meaning within the social reading event]. In V.L.M. Paiva (Ed.), Metáforas do cotidiano [Everyday metaphors] (13–38). Belo horizonte: Edit. da UFMG. Zanotto, Mara S. & Heronides M.M. Moura (2002). Indeterminacy and negotiation of meaning. In J. Verschueren, J-O. Östman, J. Blommaert & C. Bulcaen (Eds.), The handbook of pragmatics (1–13). Supplement 2000. Amsterdam, John Benjamins. Zanotto, Mara S. & Dieli V. Palma (2003). Opening Pandora’s Box: A Case Study in Development. Paper presented at RaAM 5 – Researching and Applying Metaphor. Paris, Université Paris 13. Zanotto, Mara S. & Maria Laura P. Ricciardi (1984). A construção textual do significado metafórico [The textual construction of metaphorical meaning] IX Anais do Grupo de Estudos Lingüísticos de São Paulo (42–47), Batatais.
chapter 2
Metaphor shifting in the dynamics of talk Lynne Cameron This chapter explores what happens to a metaphor after it has been first used. Metaphor shifting refers to changes and adaptations made to the metaphor as the talk or text proceeds. Shifting is analysed from the perspective of what language users do with the vehicle (or source domain) term, its connecting concepts and lexical fields, in the dynamic flux of language use. The data on metaphor shifting is drawn from two very different discourse contexts: school classrooms and conciliation talk. In each context, certain types of shifting are found which serve the particular discourse purposes or goals. This adaptive variation with context is found to connect to underlying stability in the available processes of metaphor use. Metaphor shifting thus illustrates the research issue that arises in applied linguistic approaches to metaphor, and as discussed in the Introduction to this volume, of relating generalised principles to context-bound descriptive frameworks. Empirical evidence is presented of three types of shifting: vehicle re-deployment; vehicle development, and vehicle literalisation, in which metaphor shifts into the real world. The range of types of shifting are exemplified and described, together with the rhetorical and discourse effects of their use in context. The final section brings these processes together in a general framework of metaphor shifting, and considers implications for applied metaphor research and theory-building.
Background The changing and shifting nature of metaphor in use can be found at different levels. At the broadest level of polytropy, language users might move between metaphor and various other types of figures, including metonymy. Kimmel, taking an anthropological approach to metaphor, argues for the study of metaphor in its dynamic discourse context and highlights shifting between metaphor and other figures in “interpenetrating, overlapping, or nested tropes” (2004: 291). Friedrich (1991: 23) lists five inter-related ‘macrotropes’ that language users, whether poets or ordinary people in conversation, combine and adapt: imagistic, modal, formal, contiguity-based (metonymy) and analogical (metaphor). The connections between metaphor and metonymy are explored in several chapters in two recent volumes (Barcelona (Ed.), 2003; Dirven & Pörings (Eds.),
Lynne Cameron
2003) but are more concerned with ‘interaction’ at a conceptual level (e.g., Croft 2003; Goossens 2003) rather than, as here, with how speakers shift from one figure to another in the course of on-line talk, and the impact such shifting has on the discourse. Of all the macrotropes, metaphor is the only one that introduces an ‘outsider’ into the discourse, in the form of the vehicle terms (lexical items) and the conceptual domains that speakers connect to them. Cognitive metaphor theory holds that conceptual metaphors are static, fixed mappings between Topic (or target) and Vehicle (or source) domains. When we study metaphor in the dynamic context of language use, our concern is with linguistic mappings that are adaptive and changing. These are tracked by successive uses of and changes in the Vehicle term, and related words and phrases. The domains that underlie Vehicle terms are inherently fuzzy in nature (Croft 2003). For real speakers in real discourse contexts, the domain of a lexical item is everything activated in memory by the use of that item. This encyclopaedic ‘everything’ will include images, words, meanings, concepts, personal episodic memories. There will be a certain commonality across speakers with similar cultural experiences and from our shared physical interactions with the world, but there will also be many individual differences. Metaphor shifting in literary texts is addressed by Goatly (1997) in a chapter entitled “The interplay of metaphors”, where he sets out categories and labels to describe how writers of novels, plays and poems shift their metaphors across and within domains as a text proceeds. However, the labels he chooses for his descriptive framework are not consistent, particularly in capturing a sense of dynamics; some, such as ‘repetition’ and ‘diversification’, describe actions, but other labels describe a property, e.g., ‘multivalency’. Kövesces (2003: 80) uses the term ‘scope’ of metaphor to describe the different Vehicle domains that can be linked to a Topic, but again this is not a dynamic label that can describe movement in talk or text. Ponterotto (2003) does acknowledge the dynamic, context-based nature of conversation, but starts from the premise that conceptual metaphors are present in speakers’ minds and then purports to show how they help maintain cohesion in talk. She is thus engaged in a different type of research aim from the empirically-driven work reported here. Steen (1992) offered a set of ‘metaphor processes’ to describe within-domain shifting in the utterances of research participants in think-aloud studies. This set was adapted for the studies with school students reported in Cameron (2003) and is explained in more detail when it is used later in this chapter.
Discourse contexts Metaphor shifting will be described in two types of talk, differing in discourse context, participants and purposes. One data set comprises talk between a perpetrator of violence and a victim in a series of meetings arranged at the request of the victim to try to understand the motivations that led to the violence. The participants were
Chapter 2. Metaphor shifting in the dynamics of talk
Patrick Magee, who in 1984, acting for the Irish Republican Army, planted a bomb in a UK hotel where the Prime Minister and members of her cabinet were staying, and Jo Berry, whose father Sir Anthony Berry was killed by the bomb. Video recordings of some of their meetings were provided by Jo Berry and analysed in a research project that aimed to explore the role of metaphor in the dynamics of reconciliation talk.1 The transcribed talk was analysed for the use of linguistic metaphors, following the procedures set out in Cameron (2003, chapter 3). The use and development of metaphors was examined within and across turns and speakers, focussing particularly on ‘metaphor clusters’, i.e., episodes with particularly high densities of metaphor use (Cameron 2007; Cameron & Stelma 2004). The participants came to the conciliation meetings with the discourse purposes of explaining their feelings and actions to the other person, metaphorised as exchanging stories. They entered the discourse from very different starting points, and undertook a journey in order to come to understand the very different experiences of the other person, or, taking a difference or ‘alterity’ view of talk (Cameron 2003), to reduce the gaps in knowledge and understanding between them.2 The other discourse context from which I draw examples of metaphor shifting is the primary (elementary) school classroom (reported in Cameron 2003). The participants were a class of fifteen students, aged between 9 and 11 years and all using English as first language, and their teachers from a small school in the north of England, serving a rural community. In addition to recordings of classroom talk, an adapted form of think-aloud technique was used to elicit students’ understanding of metaphor in written texts on scientific topics: the ozone layer and the heart. In the think-aloud studies, two girls from the class were given a specific goal in reading the texts, to evaluate its accessibility to a slightly younger child. In practice, the students explained text content, sentence by sentence, to each other and to the researcher. The discourse purposes of classroom talk can be summarised as the teacher helping children to master ideas or skills while maintaining control and motivation. In the think-aloud tasks the students operationalised their goal as the joint construction of an explanation of the text, helping each other when there were problems in understanding. As in the conciliation context, talk aimed to reduce alterity; in contrast with the conciliation context, alterity was concerned with external conceptual knowledge and understanding rather than the interpersonal.
1. We thank the participants for their permission to use and publish the data, and for their co-operation in the research. We acknowledge the support of the Arts and Humanities Research Board, UK in carrying out the research project. 2. Alterity derives from Bakhtin’s notion of ‘Otherness’ and is used to describe what separates speakers, to contrast with intersubjectivity as what they share.
Lynne Cameron
Vehicle re-deployment – shifting in topics Vehicle re-deployment describes the process in which, after its first use, a vehicle term is re-used in identical or similar form but with a different topic. The label is chosen to reflect the action as centred in the vehicle, rather than the topic. Conciliation talk revealed a significant type of vehicle re-deployment in the adoption by one speaker of a metaphor used originally by the other. We may call this discourse process of taking over the metaphor of another speaker an ‘appropriation’ of a metaphor, a label from socio-cultural theory (Wertsch 1998). The appropriating and sharing of metaphors across speakers appears to be a significant indicator of achievement of the discourse goals of conciliation, particularly where the vehicle expressions are emotive. The metaphor vehicle healing was first used by Jo in a poem that she brought to the meetings and read aloud to Pat. In the poem she connects the vehicle to two different topics, recovering from grief at the loss of her father and to the peace negotiations in Ireland:3
Extract 1
I feel that my heart heals, as Ireland heals.
In response to the reading of the poem, Pat picks up the metaphor vehicle and uses it to refer to a slightly different topic, helping other victims of the violence (Extract 2):
Extract 2
1 Pat 5
I am hoping it will be – er, . . .(1.0) it can act as a – a– a lesson or an example, . . . that somebody else can – er,
3. The conciliation talk was transcribed into ‘intonation units’ (Chafe 1996). The ends of intonation units are marked with the following symbols:
, continuing intonation contour . final intonation contour – a truncated (incomplete) intonation unit
Other symbols:
. . . . .(2.0) [ ] ???
short pause longer pause, time in seconds overlapping talk indecipherable talk
Chapter 2. Metaphor shifting in the dynamics of talk
10 Jo Pat
. . .(1.0) er – gain something from it. . . hmh . . . and er, . . move on in their own, . . you know, . . healing process.
A more significant shift occurs as Pat talks about how he feels meeting Jo and listening to her pain. With a lot of hesitation and pausing, suggesting some unease, he uses the healing vehicle for the first time in reference to himself (extract 3, line 14):
Extract 3
1 Pat 5 10 15 Jo Pat 20 Jo Pat 25
(to) be confronted, . . with your pain. . . . that’s a consequence that – er, . . .(3.0) you know, I suppose I deserve. . . .(2.0) you know, . . .(1.0) and – er, . . .(2.0) seems very – how do you put it, er, . . .(2.0) maybe that’s part of healing too, . . my healing. your healing. . . [yeah]. [yeah]. . . .(1.0) you know, er, . . .(2.0) it’s – er, something I have to go through. . . . hmh . . . if I’m gonna sort of – er, . . .(1.0) really retain my humanity.
To claim an entitlement to a similar process could be seen as unjustified from a perpetrator of violence, but it may also be seen as a measure of the success of their meeting that he feels able to do so. From Jo’s response in line 15, and from the use of healing in the rest of the conversation, there seems to be an acceptance that Pat can use the
Lynne Cameron
same metaphor as Jo, even though his process of working through the consequences of political decisions is very different from Jo’s recovering from grief. At the end of the meeting (extract 4), his use of the healing metaphor seems to refer to both of these processes:
Extract 4
. . . the process . . . (1.0) that we’re in. . . . (1.0) that er – it has . . . provided – er, . . . a sort of, . . an opportunity, a window to . . . reflect on the past. and that’s a part of the [healing] process.
Appropriation takes place through re-use of a metaphor vehicle with a changed topic. In the conciliation context, appropriation through vehicle re-deployment seems to offer a safe way for a speaker to attempt alignment with the other, and appropriation of metaphors is an important signal and measure of increasing alignment between participants. Less direct vehicle re-deployment can be seen in the following extract from the first analysed conversation. Here Jo describes how the idea of meeting Pat came to her after her father was killed by the bomb. Terms from the vehicle domain of movement or journey used first to metaphorise the actual process of conciliation and then redeployed to metaphorise the idea of it.
Extract 5
1 Jo 5 Pat Jo 10 Pat
. . .(1.0) [and] I – and I saw very clearly. . . .(1.0) that the – . . the end of that journey, would be, . . . sitting down and, . . . talking to the people who did it. . . . hmh . . . that just came in a moment, and then went away, and then – . . . there’s been a long long . . 16 years of [getting to this point]. [hmh hmh]
Chapter 2. Metaphor shifting in the dynamics of talk
In line 4, Jo uses the metaphor vehicle the end of that journey to refer to the topic of coming face to face with the person responsible for killing her father. The journey is a metaphor for the process of coming to terms with her grief and finding out about the motivation of the bombers. The shift occurs in lines 9 and 10, where came and went away, which are related to journey and movement, are used to talk metaphorically, not about the meeting, but about the idea of meeting. She returns to the former topic in line 12, where she rephrases the vehicle the end of the journey to getting to this point, a within-domain vehicle shifting (see next section). Vehicle re-deployment in extract 5 involves a greater change in vehicle terms than in the previous example of healing. One of the enduring problems in metaphor analysis lies in describing the ‘domains’ underlying topic and vehicle (Cameron 1999; Croft & Cruse 2004), particularly since the basic identifying criterion of metaphor is a mapping between two ‘distinct’ domains. To see the shift from journey to came / went away to getting to this point as re-deployment of a vehicle relies on seeing these lexical items as connected through an underlying conceptual domain. Cognitive metaphor theory takes this as a starting point (although the theoretical problem then becomes the limiting of domains). A discourse-based theory of metaphor in use can perhaps take a bottom-up, inductive approach to domains, in which the vehicle domain is the lexical space ‘carved out’ by the talk. In the process of analysis, came / went away may be tentatively grouped in an emerging vehicle domain with journey for later re-analysis. Vehicle re-deployment is thus an adaptive process in which the same or related vehicle term is applied to a changed topic domain. Even if the lexical form of the vehicle remains the same, its sense will shift slightly as it co-adapts to the new topic. Vehicle re-deployment is connected to the ‘multivalency’ (Goatly 1997: 255), a term connected to the science of atoms and molecules. Multivalency describes the (static) property of a metaphor vehicle to work with several different topics; vehicle re-deployment describes the dynamic and interactional process in which language users take advantage of multivalency. To understand the vehicle re-deployment in terms of the discourse context and purposes, we need to know that there was no external management of the conversation, apart from one or two requests from the film crew; the two sat together for long periods of time, moving from one topic to another as their thoughts took them. They had responsibility for managing the process of the talk towards their overall goal of reducing differences in knowledge and understanding. This contextual constraint was reflected in utterances, as here, that made meta-comments on the content or nature of the talk. When we move to the discourse of education, we again find instances of vehicle re-deployment. One particular type emerged as significant and was labelled ‘topic reference shift’ (Cameron 2003: 184). This time the shifting was accidental and connected to lack of topic knowledge on the part of the students. In the first think-aloud study about the ozone layer, the students discussed the text sentence: The atmosphere is the blanket of gases that surround the Earth. The topic of the metaphorical vehicle blanket is atmosphere, and the sentence includes other terms
Lynne Cameron
from the same or related domains: gases, the Earth. As the students try to make sense of the sentence, Ellen produced the explication in Extract 6:
Extract 6
Ellen:
there’s actually two types of gases . . there’s a bad type of gas and a good type of gas which is a . . blanket around our earth
In her unpacking and re-construction of the metaphor in the above extract, Ellen uses the vehicle term blanket with a good type of gas (referring to the ozone layer), thereby shifting the topic reference from the atmosphere to just a single gas or part of the atmosphere. This seemed to be a temporary shifting, since the protocol concludes (Extract 7 below) with an accurate metaphor construction in which the blanket vehicle is connected to gases:
Extract 7
1 Ellen: Louise: 5
cos when you’re in . . bed you’ve got a blanket sort of . . . protecting you from the cold and then there’s another type of blanket which is of gases . . surrounds the earth
However, when asked to recall the content of the text one week later, the shifted version was produced:
Extract 8
R: what is the ozone layer? Louise: . . . it’s like an invisible shield around the earth Ellen: or blanket . . as the book puts it
In Extract 8, the vehicle blanket has become linked in the students’ minds to the single layer of gas, the ozone layer, rather than to multi-layered atmosphere of the original text. (The topic shift has also happened with the metaphor shield which occurred later in the same text and that Louise recalls here). This particular instance of Topic Reference Shift does not do any educational ‘damage’, in that the metaphor works equally well for both ozone layer and atmosphere, but other occurrences result in greater inaccuracies. Shifts in topic reference through vehicle re-deployment occurred in both thinkaloud studies. Topic reference shift occurred more often with metaphorically used verbs than noun phrases. Talk around 5 of the 8 verbs used metaphorically in the ozone layer text included topic reference shifts. Verbs used non-metaphorically were much less likely to show such shifts in meaning. They were sometimes compounded by the students’ lack of topic knowledge or by the text structure, as with the mistaken pronominal reference in Extract 8. Accurate topic knowledge sometimes served to restrain the shifting, whereas absence of such knowledge might accentuate it.
Chapter 2. Metaphor shifting in the dynamics of talk
The use of linguistic metaphor vehicles in talk creates an affordance which speakers can use in various ways and for various interactional purposes. The outcomes of vehicle re-deployment manifest themselves in various ways, including metaphor appropriation and topic reference shifts, as language is used in different discourse contexts, driven by differing goals and purposes. In the educational context, where students struggled to make sense of unfamiliar ideas using incomplete knowledge of topic domains, vehicle re-deployment might happen by accident and lead to, perhaps temporarily, inaccurate interpretations of metaphors. In the conciliation context, where participants were trying to make themselves understood by the other party, more deliberate vehicle re-deployment could signal a desire for alignment through metaphor appropriation.
Vehicle development Vehicle shifting is more widely recognised in the literature than vehicle re-deployment or topic shifting. Once a vehicle is introduced into discourse it is often developed through repetition, relexicalisation and explication (Steen 1992; Cameron 2003), or extended and modified (Goatly 1997) as aspects of the vehicle are mentioned and connected, implicitly or explicitly, with the topic to build up a set of mappings between the domains. In educational talk, the development of the vehicle often forms a key part of explaining an idea through metaphor. In the following extract, several metaphor vehicles are used to help students understand how igneous rocks are formed through the action of volcanoes.
Extract 9: Volcanoes 1 T: 5 Ss: L: T: 10 Ss: T: Ss: T: 15 Ss: T:
yes . . it’s a volcano . . .(1.0) and . . . (2.0) the rocks that are formed by fire . . . just imagine rock . . .(1.0) getting so hot . . . (2.0) that it actually melts . . . (1.0) so that it becomes like . . sticky treacle ugh (whisper) treacle or even . . . like . . . runny butter ugh have you ever put . . . a little dish . . . with butter in . . . into the microwave? yes and left it for too long? yes do you know what happens? . .
Lynne Cameron
S: T: 20 25 S: T:
I did it at the weekend . . so I know what happens is molten lava like wax? yes . . . (1.0) it can be a bit like wax . . . (1.0) but do you know what happens to butter? it does . . . there are two things it does . . which are like . . volcanic . . rocks . . when they’re being ???? it bubbles it bubbles . . well done . . yes . . and it sort of keeps doing this ???? so that’s where these rocks come from . . . (1.0) (from Cameron 2003: 103)
In this extract, the metaphorical comparisons have the form of similes. Two are produced by the teacher like sticky treacle, like runny butter, and one by a student like wax.4 The first, rock . . . becomes like sticky treacle, is introduced into the talk after a process of Topic development (lines 1–4). The metaphor topic volcanic rock is talked about as rock getting so hot that it actually melts. The sticky treacle vehicle is not further expanded, but instead runny butter is introduced as a comparison for the same topic. These two can be seen as coming from separate vehicle domains (treacle vs butter) or as both coming from a domain of solid foods that can have a liquid consistency. If we choose the latter description, then the vehicle term is ‘relexicalised’ in a within-domain shift from treacle to butter. In line 17, a student offers a further vehicle relexicalisation when she asks if lava is like wax. At this point the teacher is in the middle of explaining how runny butter is like lava and she does not take up the student’s suggestion, just repeating and modifying it in her feedback as a bit like wax. In the course of the interaction, both wax and butter are repeated by speakers. Of the three vehicle terms, butter offers the most useful vehicle to be further developed, since, like rock but unlike treacle, it requires heat to change state from solid to liquid. Between lines 19 and 25, the vehicle domain of butter is further extended in terms of its action under heat and explicitly connected to the Topic volcanic rock. In the educational context, this development is labelled ‘vehicle elaboration’. In addition to elaboration, which takes the talk about vehicle to a more specific level of features, the broader category of ‘vehicle explication’ may also involve giving an example of the vehicle or expanding the vehicle without going to a more specific level.
4. Such comparisons match the criteria for metaphor in that two unlike domains are brought together and some new meaning emerges that draws from both of them. Since it is the comparison that is metaphorical, the vehicle is identified as including the tuning device like.
Chapter 2. Metaphor shifting in the dynamics of talk
In the conciliation talk, speakers used the same processes of vehicle repetition r elexicalisation, and explication. They used these for their specific discourse purposes in trying to explain and understand, not scientific concepts, but themselves and each other. In Extract 10, Pat asks Jo whether she had thought about the meeting in terms of the political or the personal, using see metaphorically to refer to her expectations:
Extract 10
1 Pat 5 Jo 10 Pat Jo
. . . did you see it as like individuals, or did you see it as a sort of a – . . .(1.0) the big . . political picture, the IRA, or, . . . the war. . . .(1.0) um you know what I mean, er, yeah, . . . you were – you were aware that there’s a – . . . it’s going to be an individual who you’d be sitting down with. hmh . . .(1.0) I saw it as both.
The vehicle term see is repeated (line 2) and used by the other speaker in its past tense form (line 14). Line 3 of the extract includes the phrase the big political picture which continues the seeing metaphor. An additional type of vehicle shifting found in the conciliation conversations was to a contrasting term in the same domain. In the next extract, Pat is talking about the impact of meeting Jo and understanding the consequences of his actions, for which he uses further vehicle terms from the domain of seeing. Within this domain, he first contrasts seeing close up (coming face-to-face) and not seeing (lose sight of), and a few seconds later, contrasts a whole picture with a glimpse:
Extract 11 1 5
you only come face-to-face with it in a situation like that. there is – . . .(1.0) it’s so easy to lose, . . . sight, . . .(1.0) of, . . you know, the enemy’s humanity. . . . it’s never the whole picture.
Lynne Cameron
10 15
. . . it’s never the whole picture. . . . that’s why, you know, on reflection you – . . sometimes you get a – like a glimpse. even in the midst of – er, a lot of struggle. . . of the other person’s humanity.
In the next example, Jo contrasts the feeling of being on a journey (on the way, line 1) with that of reaching a safe place (come home, line 9) as she talks about the process of going to Ireland to talk to people:
Extract 12 1 5
to meet so many people on the way, to – . . . to help me. . . .(1.0) I . . . first . . came to Northern Ireland in 85. . . . beginning of 85. . . .(1.0) and, . . .(2.0) when I arrived, I’d such a feeling of, . . .(2.0) I’ve come home.
Contrasting metaphors were used for emphasizing a point, as in the examples above, and also to explore alternative actions, such as when Jo discussed what it would have been like to respond to her father’s death with anger and revenge, rather than seeking conciliation. In this discourse context, the exploration of alternative responses and hypothetical scenarios is important for participants, and it seems that Vehicle development and shifting offers a useful way to do this. The possibility of vehicle shifting through elaboration was exploited at a couple of points in the conciliation talk by Pat, in order to distance himself from statements made by Jo. At these points, it seemed that she was claiming too much shared ground between them. In her poem, Jo speaks of conciliation as building a bridge. Pat comments after the poem is read aloud: in the journey. . .coming to a bridge . . with two ends
The elaboration of the bridge vehicle to emphasise two ends also forces a topic shift, away from conciliation itself to emphasize the different starting points of the two participants in the conciliation process. This move enables Pat to distance himself slightly while at the same time maintaining alignment through the use of Jo’s metaphor vehicle.
Chapter 2. Metaphor shifting in the dynamics of talk
Vehicle development over a longer time scale leads to the phenomenon of systematic metaphor, a discourse alternative to ‘conceptual metaphor’. For example, extracts 10 and 11 include lexical items connected to see and picture: lose sight of the other’s humanity until we do see each other in our true light. . we’re always going to be dealing with some reduction or a caricature get a glimpse (of the other person’s humanity) a distorted picture
Using the idea of the emergence of a vehicle domain across a discourse event, and collecting together connected metaphor vehicles as metaphors shift across the talk, we have a larger metaphor set whose vehicles describe partial, problematic or deficient ways of seeing, contrasting these with whole or true ways of seeing. This vehicle domain is connected in the metaphor set to a topic domain of not fully understanding the Other. This emergent metaphor set is a context-bound phenomenon, related to a specific discourse event and speakers, and that I call a ‘systematic metaphor’. Systematic metaphors are emergent phenomena that arise on the level of the discourse event out of decisions made in the minute-by-minute dynamics of talking-and-thinking, and are identified inductively. Systematic metaphors are not the same as generalized conceptual metaphors, abstracted across multiple discourse events and speakers. The particular systematic metaphor illustrated above might be appropriately labelled not fully understanding another person is partial or problematic seeing. Its relation to the conceptual metaphor understanding is seeing might be of variation around a larger stability, or of a specific instance of a more generalized form. Table 1 summarises the types of within-domain vehicle shifting, or vehicle development, found in talk across the two contexts. Table 1. Types of Vehicle development Vehicle development
Description
Vehicle Repetition Vehicle Relexicalisation Vehicle Explication Vehicle Contrast
the term is repeated in identical or transformed form a near synonym or equivalent is used expansion, elaboration or exemplification of the term an antonymic or contrasting term is used
Shifting from one vehicle domain to another also occurs. Conciliation, for example, is spoken of as a journey, and as building a bridge. These separate semantic domains are sometimes brought together in a combined scenario, as in Pat’s coming to a bridge.
Lynne Cameron
Shifting into the real world: Vehicle literalisation The final type of metaphor shifting to be reported is from metaphor to the real world or literal, sometimes via metonymy. Extract 10, line 6, included an example of this shifting: 1 Jo 5
. . .(1.0) [and] I – and I saw very clearly. . . .(1.0) that the – . . the end of that journey, would be, . . sitting down and, . . . talking to the people who did it.
sitting down is a ‘bridge term’ (Kittay 1987: 166): a lexical item which can apply in both vehicle and topic domains. People getting to the end of a journey might sit down for a rest and people involved in conciliation might sit down when they meet each other. When a bridge term is used, the vehicle domain is brought into the topic domain, and the metaphor is shifted into the literal (i.e., non-metaphorical) world. In the shift, the metaphor of the conciliation process as a journey collapses into the literal scenario of Jo sitting down with Pat and talking. In a radio interview, Pat literalises the scenario even more vividly and specifically by describing the place where they sat:
Extract 13
I’m sitting there beside the woman whose father I have killed and at that time I was sitting in this wee kitchen talking to this woman for the first time whose father’s dead
In the conciliation context, the movement from metaphor to literal reflects the course of the conciliation process; metaphor is used in talk about the process leading up to meeting but the meeting is described more literally. However, there is still use of a trope here, in that sitting down is a metonymy, standing for meeting. The phrases sitting and talking to / sitting down with are not innocent but, when all uses are considered across the conversations, appear heavy with import, some of it carried over from the connection to the end of the journey but most of it deriving from real world social and physical experience. Sitting down with someone (a) is an intentional action, (b) unavoidably requires an acknowledgement of the person you sit with, who is a specific individual, not a stereotype or category, and (c) renders the participants open and vulnerable to each other. As such, sitting down with is a very potent metonymy to describe
Chapter 2. Metaphor shifting in the dynamics of talk
a conciliation meeting. Similar metaphor to literal shifting, often with metonymy, also occurred with the domains connected to walking, spaces, places. A somewhat parallel process has been observed in literature, where the literalisation of a vehicle results in it taking on in its literal use some of the symbolism that it held in the metaphorical scenario (Goatly 1997: 278) In the educational talk, the same process was labelled ‘vehicle contextualisation’. Here metaphor vehicle terms were shifted into the real world of the participants, but not into the Topic domain (Cameron 2003: 107) Extracts 6 and 9 include examples. In Extract 6, the girls talking about the metaphor the atmosphere is the blanket of gases connect the Vehicle term blanket to their own lives and experience: cos when you’re in . . . bed you’ve got a blanket sort of . . . protecting you from the cold
Here the Vehicle blanket is literalised and contextualised by connecting to being warm in bed. Protecting is a bridge term, since it can apply to both blankets and the atmosphere, and it was in fact the property that the girls focussed on in their discussion of the atmosphere and ozone layer. As with topic reference shift, the affordance of literalisation may operate to help understanding or to hinder it. In the ozone layer text, the writer follows the sentence about the blanket of gases with the sentence “It is made up of several layers” where the pronoun It refers to the atmosphere. The word layers acts as a bridge term in the dynamics of the talk, linking the domains of atmosphere and blanket for the students, even though it was intended just as information about the Topic. Later in the think-aloud, Louise comments: he’s talking about it (the atmosphere) . . . a layer like a blanket
The word layers becomes linked to the vehicle domain of blanket, and, on a bed, each blanket forms a single layer. When vehicle and topic are re-combined, there are two interpretations, both different from the author’s intended meaning: the atmosphere is a single layer, or the blanket refers to just the ozone layer, as in the topic reference shift. The affordances of the text combined with the children’s own experiences lead to indeterminacy of meaning (see Zanotto & Palma, this volume). In Extract 9, the teacher contextualises the vehicle runny butter for the students when she moves the talk to her experience of overheating butter in a microwave oven. There is no bridge term used here: T: Ss: T: Ss:
have you ever put . . a little dish . . with butter in . . into the microwave? yes and left it for too long? yes
Lynne Cameron
Vehicle contextualisation is a shift into the real world which seems designed to help make sense of the metaphor by connecting it to the students’ life experience. In these cases, there is no symbolism or import to be transferred.
Discussion and conclusions Metaphor shifting has been elaborated in this paper as a phenomenon of metaphor in use. When people use metaphor in talk, they seldom produce a neatly constructed figure in a single utterance, other than when a metaphorical idiom is used at the end of a turn (Drew & Holt 1988). Spontaneous talk is dynamic and approximate, with metaphors introduced, adjusted and developed over utterances and turns. As speakers pursue their discourse aims and topics through talking-and-thinking, connections and reminders are sparked in their minds which may divert them sideways. It seems that linguistic metaphor, by bringing a vehicle term with its lexical field and semantic domain into the talking-and-thinking, acts to loosen the referential ties with the topic domain, creating affordances for the various types of shifting reported in this chapter. The introduction of Vehicle terms into the text seemed to create a kind of centrifugal cognitive force that opens up potentially endless links to other concepts . . . (Cameron 2003: 191)
Metaphor shifting occurs as speakers move through a social, linguistic and cognitive ‘landscape’, adjusting and adapting contingently. When we transcribe and analyse their talk, it is as if we are allowed to view that landscape from a distance and thereby see it as a whole in a way that the speakers themselves could not. What appears to us as a path across the landscape, to them is the avoidance of immediate obstacles, split second decisions as to direction, and the negotiation of sudden difficulties. The types of metaphor shifting identified here show us some of the possibilities available to speakers and their outcomes in particular discourse contexts. They may be summarised as in Table 2. The central two columns of Table 2 list the shifting processes that are available to metaphor users. In a particular discourse context, these possibilities interact with speakers and the discourse environment to become affordances that may be helpful in the achievement of discourse purposes, or may take the talk in inappropriate directions. The approach to metaphor taken here has been dynamical and discourse-based (Cameron 2007; Cameron & Deignan 2006). Metaphor shifting has been described from evidence of linguistic metaphor use in real discourse contexts. It has not been
Chapter 2. Metaphor shifting in the dynamics of talk
Table 2. Types of metaphor shifting Metaphor shifting
Vehicle
Topic
Vehicle re-deployment The same or semantically- changes connected lexical item is re-used with a different Topic. Vehicle development Vehicle term is remains the same ● repeated ● relexicalised ● explicated – exemplified moves to – elaborated connected – expanded Topics ● contrasted Vehicle literalization Vehicle term (bridge) merges with is used in reference Vehicle; to Topic can become metonymic
Discourse outcomes Emergent Vehicle domain. Topic Reference Shift. Metaphor appropriation. Explanations through metaphor. Extended metaphors. Challenges to metaphor. Systematic metaphor Vehicle contextualisation. Symbolisation of Topic.
necessary to take a strong cognitive approach that would assume the pre-existence of conceptual metaphors in individual minds. If conceptual metaphors are taken to be mappings that have stabilised across speech communities and cultural groups, they can be seen as one influence on metaphor shifting in a discourse context. In the metaphor of ‘the landscape of talk’ used earlier, conceptual metaphors may offer established way-routed paths that speakers are more likely to take because they are more ‘visible’. They are not, however, the only possibility open to speakers. The multiple possibilities opened up by the use of a metaphor Vehicle term are both derived from and constrained by speakers’ experience of the world, their socio-cultural contexts, and their discourse purposes.
References Barcelona, Antonio (Ed.) (2003). Metaphor and metonymy at the crossroads. Berlin: Mouton de Gruyter. Cameron, Lynne (1999). Operationalising metaphor for applied linguistic research. In L. Cameron & G. Low (Eds.), Researching and applying metaphor (3–28). Cambridge: Cambridge University Press. Cameron, Lynne (2003). Metaphor in educational discourse. London: Continuum. Cameron, Lynne (2007). Patterns of metaphor use in reconciliation talk. Discourse and Society, 18, 197–222.
Lynne Cameron Cameron, Lynne, & Juup Stelma (2004). Metaphor clusters in discourse. Journal of Applied Linguistics, 2, 107–136. Cameron, Lynne, & Alice Deignan (2006). The emergence of metaphor in discourse. Applied Linguistics, 27, 671–690. Chafe, Wallace (1996). Discourse, consciousness and time. Chicago: University of Chicago Press. Croft, William (2003). The role of domains in the interpretation of metaphors and metonymies. In R. Dirven & R. Pörings (Eds.), Metaphor and metonymy in comparison and contrast (161–206). Berlin: Mouton de Gruyter. Croft, William, & Cruse, Alan. (2004). Cognitive Linguistics. Cambridge: Cambridge University Press. Dirven, René, & Ralf Pörings (Eds.) Metaphor and metonymy in comparison and contrast. Berlin: Mouton de Gruyter. Drew, Paul, & Elizabeth Holt (1988). Complainable matters: The use of idiomatic expressions in making complaints. Social Problems, 35, 398–417. Friedrich, Paul (1991). Polytropy. In James Fernandez (Ed.), Beyond metaphor: The theory of tropes in anthropology (17–55). Stanford, CA: Stanford University Press. Goatly, Andrew (1997). The language of metaphors. London: Routledge. Goossens, Louis (2003). Metaphtonymy: The interaction of metaphor and metonymy in expressions for linguistic action. In René Dirven & Ralf Pörings (Eds.), Metaphor and metonymy in comparison and contrast (349–378). Berlin: Mouton de Gruyter. Kimmel, Michael (2004). Metaphor variation in cultural context: Perspectives from anthropology. European Journal of English Studies, 8, 275–294. Kittay, Eva Fodor (1987). Metaphor. Oxford: Clarendon Press. Kövesces, Zoltan (2003). The scope of metaphor. In Antonio Barcelona (Ed.), Metaphor and metonymy at the crossraods (79–92). Berlin: Mouton de Gruyter. Ponterotto, Diana (2003). The cohesive role of cognitive metaphor in discourse and conversation. In Antonio Barcelona (Ed.), Metaphor and metonymy at the crossroads (283–298). Berlin: Mouton de Gruyter. Steen, Gerard (1992). Metaphor in literary reception. Doctoral dissertation. Vrije Universiteit, Amsterdam. Wertsch, J. (1998). Mind as Action. New York: Oxford University Press.
chapter 3
Adding sound to the picture Motivating the lexical composition of metaphorical idioms in English, Dutch and Spanish Frank Boers & Hélène Stengers In this chapter we investigate the lexical composition of a core category of conventional metaphorical expressions, i.e., figurative idioms. More precisely, we shall be looking at idioms in English, Dutch and Spanish that can be traced back to the following general source domains: • • • • • • • • • • • • • • • • •
agriculture & gardening (e.g., nip something in the bud); buildings & construction (e.g., get in on the ground floor); clothes & adornment (e.g., try something on for size); commerce & accounting (e.g., wipe the slate clean); entertainment & public performance (e.g., play to the gallery); fauna & flora (e.g., put out feelers); food & cooking (e.g., on the back burner); games & sports (e.g., keep your eye on the ball); handicraft & manufacturing (e.g., break the mould); health & medicine (e.g., keep your finger on the pulse); jurisdiction & punishment (e.g., read the riot act); mechanisms & machinery (e.g., fire on all cylinders); religion & superstition (e.g., fall from grace); science & research (e.g., the acid test); vehicles & transport (e.g., miss the boat); war & aggression (e.g., break ranks); weather & natural phenomena (e.g., under a cloud).
As the focus of this book is metaphor, we shall not include similes, i.e., explicit analogies, as part of our investigation. We shall only take into account standardised, (semi-) fixed, multiword figurative expressions that instantiate implicit analogies. Before cognitive-semantic theory, the form-meaning relation of idioms was predominantly treated as arbitrary, in the sense that their figurative meaning could not be inferred by simply adding up the semantics of their lexical constituents. In other words,
Frank Boers & Hélène Stengers
idioms were considered as non-compositional (e.g., Fernando & Flavell 1981). Since the 1980s, however, cognitive semanticist (Johnson 1987; Lakoff 1987; Gibbs 1994) have argued that the meaning of many idioms is not arbitrary but in fact motivated, i.e., explainable (albeit in retrospect). The genesis and semantics of various classes of figurative idioms have been shown to be motivated by general conceptual metaphors (e.g., Lakoff & Johnson 1980) and/or conceptual metonymies (e.g., Kövecses 1990). The meaning of the idioms that we shall be looking at in this chapter is motivated by the original, literal usage of the expressions. More precisely, these figurative idioms project the experiential ‘logic’ or value-judgements associated with a typical scenario in their source domain onto more abstract domains. For example, hitting below the belt, being on the ropes, and throwing in the towel project the experiential logic and value-judgements associated with typical scenarios of boxing matches. Jumping the gun projects a typical scenario of a running contest. Pulling something out of a hat projects a typical scenario of a magician’s performance. And so on. In contexts of second language acquisition or foreign language teaching, presenting idioms as semantically motivated is obviously an attractive alternative to the traditional approach in which idioms are presented as arbitrary, because motivated meaning carries the potential of insightful learning instead of blind memorisation. Several controlled experiments have already revealed the benefits of cognitive semantic approaches to teaching conventional metaphorical expressions (such as figurative idioms) with regard to comprehension (Boers 2000a) as well as retention (Boers 2000b; Boers 2001; Boers et al. 2004a; Kövecses & Szabó 1996), and even with regard to register awareness (Boers et al. 2007). The purpose of the present chapter, however, is not to motivate why given idioms mean what they mean. Instead, its purpose is to investigate whether it might be possible to motivate the precise lexical composition of idioms, i.e., why certain word combinations rather than others are the standard ‘choice’. At a conceptual/semantic level it is to be expected that idiomatic language will be nourished by lexical fields that are connected with experiential source domains that have been relatively salient in the linguistic community. In other words, the heavy presence of certain lexical fields in a particular community’s repertoire of idioms may reflect that community’s preoccupations (albeit of the past). For example, it is probably not surprising that the repertoires of idioms of communities with a rich ‘seafaring’ past (such as English and Dutch) contain a considerable number of idioms that are derived from the source domain of boats and sailing (e.g., in the doldrums; on an even keel; a steady hand on the tiller) and that thus contain words belonging to the associated lexical field. Some experiential domains may be quite culture-specific and consequently their derived idioms may be quite language-specific. Idioms containing words borrowed from the domains of baseball and cricket, for instance, are obviously more common in English than are idioms containing keywords borrowed from the domain of bullfighting. Some quantitative evidence of such cross-linguistic variation in source-domain salience will be presented below, in section two. Language learners apparently find idioms that are derived from
Chapter 3. Adding sound to the picture
‘unfamiliar’ source domains especially hard to interpret and remember (Boers et al. 2004b). While the relative salience of a given source domain may help explain the occurrence of its associated lexical field in a considerable fraction of a language’s idiom repertoire, it often falls short of explaining the precise choice of lexis. Given the phenomenon of (near-)synonymy, the question remains why certain source-domain content words have made it into standardised idioms rather than their (near-)synonyms (when available). For example, if the idiom Cut and run is derived from the nautical domain where it literally meant cutting the anchor cable so that one could escape from sudden danger (Speake 1999), then we may wonder why cut and flee or cut and sail away have not become standardised phrases. Likewise, we may wonder why jump the gun has become the standardised phrase rather than precede the gun or start before the gun, why we say leave someone high and dry rather than leave someone up and dry, why time will tell rather than time will say, why it takes two to tango rather than it takes two to waltz, and so on. Conceptual metaphor theory may help language learners appreciate the relatively systematic nature of source-domain to target-domain mappings at a conceptual level, but learners would benefit even more if they could see some rhyme and reason for the precise choice of lexis in the instantiations of conceptual metaphors at the linguistic level. We propose that in a number of conventional expressions a kind of phonological motivation is at play, which could explain why certain word combinations rather than semantically close-equivalent ones have had special appeal to become ‘institutionalised’. More specifically, we propose that alliteration and assonance can play important roles in the lexical composition of idioms, a role which to our knowledge has not received much attention in conceptual-metaphor research. Quantitative evidence of the prevalence of these sound patterns will be presented below, in section three. Recognition of the role of alliteration and assonance in the lexical make-up of idioms can indeed be beneficial in SLA or FLT contexts. Experiments have shown, for example, that alliteration significantly facilitates learners’ recollection of multiword expressions (such as strong collocations and idioms), and that this mnemonic effect can be enhanced by awareness-raising (Boers & Lindstromberg 2005). As we shall see below, however, the proposed types of phonological motivation may not be equally prevalent across languages. At least two typological continua may make a given language more or less prone to, for example, alliteration (e.g., above board). The first typological continuum that comes to mind is related to word stress and the prosodic distinction between stress-timed and syllable-timed languages (e.g., Abercrombie 1967; Catford 1986). Alliteration is probably most pronounced if it precedes a stressed vowel (or accentuated foot). Languages that are generally believed to lean towards stress-timed rhythm, such as English and Dutch, may therefore be more prone to alliteration than languages that are generally believed to lean towards syllable-timed rhythm, such as French and Spanish. Furthermore, the prototypical type of alliteration, i.e., the repetition of word-initial consonants (e.g., from pillar to
Frank Boers & Hélène Stengers
post), for instance, may be especially common in languages where word stress is commonly placed at the beginning of words, such as English, but less so in languages that give prominence to the final syllable of words, such as French. As a word of caution, it needs to be mentioned that the distinction (and certainly the dichotomy) between stress-timed and syllable-timed languages has increasingly come under fire from experimental phoneticians, some of whom claim that some languages (e.g., Spanish) do not correspond to either type (e.g., Dauer 1983) and some of whom claim that the typology itself may be a fallacy (e.g., Pamies Bertràn 1999). At any rate, our data (see below) reveal that Dutch (a so-called stress-timed language along with English) is much less prone to alliteration than English, which suggest that the prosodic typology cannot be the whole story. A second typological continuum that may help explain cross-linguistic variation in the popularity of (word-initial) alliteration is the distinction between fixed-word-order languages with poor inflection and loose-word-order languages with rich inflection (e.g., Comrie 1981). Inflection (which is richer, for example, in Spanish than in English) tends to make words longer and thus creates more distance between word-initial consonants that are repeated, which may weaken the alliterative effect. Alliteration is probably most powerful when it involves monosyllabic words (e.g., to a tee; through thick and thin). The relative fixedness of word order (which often appears to compensate for the absence of inflection) may also enhance the ‘productivity’ of alliteration. If word order is fixed, then this may increase the likelihood of phonologically appealing word combinations being reiterated and subsequently becoming standardised. The English alliterating idiom beat about the bush, for example, will still cohere in a complex subordinating clause, such as in My wife complained that I shouldn’t have beaten about the bush again last night. By contrast, if word order changes along with clause types, such as in German and in Dutch where in subordinating clauses the verb is moved to final position, the alliterating components of an expression may not cohere anymore. The alliterating components in the Dutch idiom “iemand bij de neus nemen” (Lead someone by the nose), for example, will be separated in “Mijn vrouw klaagde dat ik me niet bij de neus had mogen laten nemen” (My wife complained that I shouldn’t have let myself be led by the nose). Furthermore, a grammatical necessity of adding prefixes under certain syntactic patterning may turn word-initial alliteration into word-internal alliteration and thus weaken its effect. For example, the past participle in Dutch takes the prefix “ge-”, which may weaken the alliteration again, as in “Ik vroeg me af of ik bij de neus was genomen” (I wondered if I had been led by the nose). If the effect of alliteration is outspoken only in the ‘canonical’ form of idioms but weakened in many of their actual occurrences, then it is less likely to be a motivating force behind their precise lexical selection. In view of these cross-linguistic, typological variables, we hypothesise that of the three languages whose idiom repertoires we shall be looking at (i.e., English, Dutch and Spanish) English is likely to be richest in prototypical alliteration, since it has the least inflection and a comparatively fixed word order.
Chapter 3. Adding sound to the picture
Part of the picture The purpose of this section is twofold: (i) to describe the databank that our study is based on, and (ii) to demonstrate that at a conceptual/semantic level the lexical composition of metaphorical idioms in a language is fuelled by the lexical fields of source domains that have been especially salient in the culture behind that language. We chose English, Dutch and Spanish as objects of investigation because (i) these were languages we were both familiar with and (ii) comparable well-established idiom dictionaries for each of them were available to us. These were The Oxford Dictionary of Idioms (Speake 1999), The Collins Cobuild Dictionary of Idioms (Sinclair & Moon 2000), the Van Dale Idioomwoordenboek (de Groot 1999) and the Diccionario Espasa dichos y frases hechas (Jiménez 1997). The advantage in using these particular dictionaries was the fact that they each provide background information about the origins of many idioms, which often helped us identify their source domains. The dictionaries also appeared comparable in terms of the scope opted for in our study (i.e., standardised metaphorical expressions), although the Spanish dictionary seemed to include more similes than the English and Dutch dictionaries (which was an additional reason for excluding similes from our study). The three dictionaries were screened for metaphorical idioms that could be traced back to the source domains listed at the beginning of the chapter (games & sports, warfare & aggression, etc.). The dissection of concrete reality along these particular categories is not at all meant as an ‘epistemological’ statement. It is the pragmatic outcome of an earlier pedagogical project in which our students were given the task to categorise idioms along source domains with a view to helping them remember the expressions (see Boers et al. 2004a). The hand counts in the dictionaries generated a databank of 1,497 English idioms, 1,368 Dutch idioms and 1,116 Spanish idioms. As experiential domains cannot always be clearly delineated, various idioms had to be listed more than once, under different source-domain headings. For example, some of the idioms derived from hunting (e.g., a sitting duck) and fishing (e.g., swallow something hook, line, and sinker) could be included under ‘sports’ as well as under ‘fauna’, and also under ‘food’. This categorisation under more than one source domain was required for about six per cent of the idioms (in the three languages). Figure 1 presents an overview of the relative contribution of the different source domains to the idiom repertoires of English, Dutch and Spanish. As expected, the distribution of the three idiom repertoires over the range of source domains appears to reflect some cross-cultural variation. Perhaps the most striking example is the greater proportion of Spanish idioms that are derived from the source domain of religion and superstition, which makes up over 17% of the Spanish idioms in our databank. Various marked contrasts appearing from graph one can be accounted for by the distribution of idioms at the level of more specific source domains than the general
na
u fa
n
sio
es gr
ns tra
rt po
i ok co
ng
ce
an
rm rfo
n
tio
ti rs
t
en
m ish
g
rin
u ct
Dutch Spanish
g
tin
n ou
English
t
en
m rn
ry
e in ch
a
en om
g in
n de
e
n ici ed
h n io rc ct ea s u r & tr n a re e a o c m ga & ns uf ra ag he & m un up ac ad e & & o n o p s p d e & c p a & l fl s o & r & s& nc lth m re s& e& ra ic fo cle n s& ea bl rc tu wa tu & he on sm cie o g l hi e i i t a i h u s e u n t ft g n p i v m a li an ic clo ld ric re ch icr t& r& om ui isd ag e d c e r n b n e h m ju at m ha in we tr a te en source domains
ts or
sp
Figure 1. Source domains of English, Dutch and Spanish idioms.
& es
0
2
4
6
8
10
12
14
16
18
m ga
percentage of repertoire
20
Frank Boers & Hélène Stengers
Chapter 3. Adding sound to the picture
ones selected for the primary categorisation. For example, the great proportion of English idioms that are derived from games and sports is partly due to (i) a high number of English idioms derived from ball games (45 idioms as compared to 18 in Dutch and only 8 in Spanish), 21 of which are baseball and cricket expressions (e.g., go in to bat for someone; knock someone for six), and (ii) a high number of card-games and gambling idioms (e.g., turn up trumps; break the bank) (56 idioms as compared to 41 in Spanish and 24 in Dutch). At the level of such more specific source domains, English also appears richest in ‘hunting’ and ‘horse-racing’ idioms. By contrast, a considerable fraction of Spanish games-and-sports idioms consists of 42 bullfighting expressions (while the Dutch and English bullfight-idiom collection seems confined to a red rag to a bull, take the bull by the horns and see red). The observation that English appears to have the most idioms derived from the general source domain of fauna and flora is partly due to its comparatively high number of ‘dog’ expressions (33 as compared to 18 in Spanish and 16 in Dutch). The fact that Spanish appears to have the fewest idioms derived from vehicles and transport is partly due to its comparatively low number of ‘boats-and-sailing’ expressions (34 as compared to 77 in English and 96 in Dutch). However, the connection between culture and metaphorical language is obviously very intricate (e.g., Boers 2003; Deignan 2003), and any conclusions with regard to culture based on the idiom repertoires of languages should therefore remain very tentative. It is important to remember, for example, that figure 1 represents a distribution over different source domains as calculated on the basis of hand counts in a lexicon, irrespective of any differences in frequencies of occurrence among the idioms. As a result, the relative contributions by the different source domains to the idiomatic lexicons available to language users does not necessarily correspond to what segments of that available lexicon are actually used on a regular basis. After all, a given source domain may have generated a high number of idioms, but a large proportion of these may actually be low-frequency items (or their popularity may have decreased over time). Conversely, a source domain that has generated relatively few idioms may still be quite salient in language use, if those few idioms happen to be high-frequency items. To illustrate this possibility, figure 2 shows the distribution over the different source domains of 564 English idioms of our databank that are signalled in The Collins Cobuild Dictionary of Idioms (Sinclair & Moon 2000) as the most frequent ones (as calculated on the basis of their occurrence in The Bank of English). Figure 2 leads us to suspect, for example, that the source domain of fauna and flora may be much less reflected in present-day English idiomatic language than suggested by figure 1. A corpus search confirms this suspicion. Looking up all the idioms derived from the four ‘most dominant’ source domains in the (British part of) The Bank of English (http://www.cobuild.collins.co.uk) results in: • •
approximately 4,750 hits for the domain of games and sports; approximately 4,150 hits for the source domain of vehicles and transport;
es
&
rts
a
un
fa
n io
s es gr
t
or
p ns
ng
i ok co
n
io
t sti
rm
g
in
t un
t
en
hm
g
rin
u ct
Frequent idioms
source domains
t
en
nm
All idioms
ce
an
ry
e in ch
a
en m no
ng ni
e rd ga
e
n ici
ed m
h n io rc ct ea s u r s & re a or co str fa e tra ni fo pe & ra ag & m ac ad er & & on nu ph su pu d e & c p a & & l flo s c & h o & & r & s n lt m re & ra ic fo cle es ce n ea bl wa tu & cie on gs ltu hi er io th ism a h i u s e u n t ft g o n n p i v c m a li cl ld ric ha & di m cr re t& ui ag ec di co er ris b n en h u m j at m ha in we tr a te en
o sp
Figure 2. Source domains of English idioms: ‘all’ vs. ‘frequent’ idioms.
g
0
5
10
15
20
am
percentage of repertoire
25
Frank Boers & Hélène Stengers
Chapter 3. Adding sound to the picture
• •
approximately 3,300 hits for the source domain of war and aggression; approximately 2,250 hits for the source domain of fauna and flora.
According to its presence in the corpus, the set of idioms derived from fauna and flora (which appeared to be one of the most productive source domains measured in terms of the variety of expressions available in the idiom repertoire as a whole) is now relegated to fourth position. By contrast, the domain of vehicles and transport appears comparatively more ‘productive’ than was suggested by figure 1. This is not so surprising, considering earlier quantitative pilot studies (e.g., Boers & Demecheleer 1997) that showed a very high frequency of JOURNEY metaphors in English (often instantiated by ‘boats and sailing’ idioms). At the same time, the frequency data confirm the relative contributions made by other source domains suggested by figure 1. For example, the very high ‘productivity’ of sports and games as a source of figurative idioms in English is clearly confirmed by the corpus data. We believe frequency data like these may inform materials writers who wish to present idioms to learners in a ‘source-thematic’ fashion, with a view to facilitating insightful learning (e.g., Lazar 2003; Wright 1999). For example, although a theme of ‘animal’ expressions may at first sight appear to be a useful subject for pedagogic exploration, precisely because of its lexical wealth, learners would probably benefit more if priority were given to a thematic exploration of ‘games’ expressions or ‘travel’ expressions, as these tend to be much more common in actual usage. It would be interesting to also compare the distribution of the Spanish and Dutch idioms over the different source domains with their relative frequencies of occurrence in present-day language use. Unfortunately, neither Spanish or Dutch idiom dictionaries provide information on frequency bands. Conducting a corpus search remains a possibility, but considering the fact that both Spanish and Dutch show a much richer inflection (including vowel changes) and a much looser word order than English, this promises to be an intricate and time-consuming exercise.
Adding the sound While culturally salient source domains can motivate why certain lexical fields are drawn from for the lexical make-up of idioms, they cannot motivate why sometimes one word rather than a semantic (near-)equivalent within the same lexical field is ‘selected’ for the composition of a given idiom. We demonstrate in this section that phonological patterning is at play in such selection in a fair number of metaphorical idioms. More specifically, we shall be looking at the sound patterns of alliteration and assonance. In reality, however, these should not be viewed as independent phenomena, since they tend to work in tandem with others to create prosodic effect. For example, alliteration may not be very overt in hook, line and sinker and lock, stock and barrel, but it nonetheless supports the rhythmic pattern of these three-part expressions.
Frank Boers & Hélène Stengers
We screened our databank of English, Spanish and Dutch idioms for the following patterns: • • • •
prototypical alliteration, i.e., repetition of word-initial consonants (e.g., a cash cow); non-prototypical alliteration, i.e., repetition of consonants including a non-wordinitial one (e.g., an iron fist in a velvet glove; off the cuff); prototypical assonance, i.e., rhyme (e.g., be left high and dry); non-prototypical assonance, i.e., incomplete rhyme (e.g., jump the gun).
Some idioms (1.5% in English, 0.8% in Spanish and 0.3% in Dutch) show a combination of alliteration and assonance. This is usually the case when a content word is repeated (e.g., neck and neck), but also in a couple of other expressions (e.g., through thick and thin). These were counted twice, both as instances of alliteration and assonance. As the aim of the exercise was to find motivation for lexical selection, we took only content words into account (i.e., no articles). Assonance that was determined by grammatical rules rather than the ‘choice’ of words, such as gender agreement in Spanish adjective-noun combinations (e.g., “mosquita muerta”) was not taken into account either (as other adjectives would take the same ending, anyhow). The screening was carried out by both authors and if either of them felt hesitant about classifying an idiom as phonologically motivated, then a ‘conservative’ approach was adopted and the idiom was not classified under the suggested sound pattern. Since word-initial alliteration and rhyme are easily recognisable, ‘disagreement’ between the two authors only occurred with regard to the categories of non-prototypical alliteration (i.e., non-word-initial alliteration) and non-prototypical assonance (i.e., incomplete rhyme). Figure 3 shows the relative prevalence of the examined sound patterns in the idiom repertoires of English, Dutch and Spanish.
20 18
percentage of repertoire
16 14 12
assonance alliteration total
10 8 6 4
total
2 0
alliteration English
Dutch
assonance Spanish
Figure 3. Alliteration and assonance in English, Dutch and Spanish idioms.
Chapter 3. Adding sound to the picture
According to our hand counts, 19.71% of the English idioms in our databank show either alliteration and/or assonance that may have played a part in the ‘choice’ of at least one of the content words in those items (and that could thus help learners remember these idioms, too). Overall, the sound patterns under examination turn out less frequent in Spanish (15.78% of the repertoire) and in Dutch (13.38% of the repertoire), but they nonetheless provide considerable ‘coverage’. Chi square shows the difference between English and Spanish to be significant at p < .02 (χ2 = 6.44), that between English and Dutch at p < .001 (χ2 = 16.30). Alliteration appears to be especially popular in English, where no fewer than 17.84% of the idioms show alliteration (15.03% word-initial), as compared to 12.10% in Spanish (11.02% word-initial) and 12.06% in Dutch (10.09% word-initial – but see below). Chi square shows these differences to be significant at p < .001 (χ2 = 15.75 and 17.99, respectively). Assonance turns out to be much less common, as it was found to occur in 3.41% of the English repertoire (1.87% rhyme), 4.48% of the Spanish repertoire (3.67% rhyme), and only 1.61% of the Dutch repertoire (1.24% rhyme). The lower frequency of assonance in the Dutch repertoire in comparison with English is significant at p < .01 (χ2 = 8.64). In comparison with Spanish it is significant at p < .001 (χ2 = 17.06). These comparative findings corroborate the hypothesis put forward in the introduction, which was that languages with little inflection (and many monosyllabic words), and with a relatively fixed word order might be most prone to taking prototypical alliteration into account when forming multiword expressions. A more detailed analysis of the subsets of idioms that show prototypical alliteration provides more supporting evidence of this typological link. Firstly, it was argued above that the impact of alliteration is likely to be greatest when monosyllabic words are involved, as the distance between the repeated consonants is then minimal. In the English subset of prototypical alliteration, 52.8% of the cases involve two monosyllabic words. This compares to 7.89% in Dutch and only 2.25% in Spanish (which happens to have the highest degree of inflection of the three). Secondly, it was argued that SVO-to-SOV transformation (which is the case in Dutch) could reduce the likelihood of alliterating phrases cohering all the time. In addition, adding a prefix to the past participle (like “ge-” in Dutch) could reduce the likelihood of prototypical alliteration involving verb forms in actual syntactic patterning. Consequently, the majority of Dutch idioms in which alliteration occurs with a finite verb have variants in actual usage in which the alliteration in weakened. In fact, almost half of the Dutch idioms that were counted in graph three under prototypical alliteration do not always show this word-initial alliteration in linguistic practice. Only 18.42% of the alliterating verb-object combinations maintain word-initial alliteration at all times, either because they are invariable, set phrases or because the alliterating consonant happens to be /G/, as in “de geest geven” (give up the ghost) (in which case the past participle preserves the alliteration, as in “hij heeft de geest gegeven”). By comparison, over 42% of both the English and Spanish subsets of prototypical alliteration involves finite verb forms.
Frank Boers & Hélène Stengers
The finding that the lexical composition of up to twenty per cent of English metaphorical idioms can be motivated phonologically must be welcome news to EFL learners facing the enormous challenge of mastering English phraseology. Still, one may wonder if the use of alliterating or rhyming idioms by native speakers might not be confined to purposes of amusement or poetic effect. In other words, one may wonder if the reported high percentage may not be skewed by the presence in our databank of alliterating or rhyming idioms that are actually restricted to very low frequency bands. However, if we zoom in on the 564 idioms in our bank that are signalled in The Collins Cobuild Dictionary of Idioms (Sinclair & Moon 2000) as the most frequently used ones (and therefore – according to the editors – as the ones that should be given priority in EFL), the pattern that emerges is quite compatible with the overall findings. As shown in figure 4, we find that over 23% of the set of ‘frequent’ idioms show a form of phonological motivation, mostly alliteration again (14.18% word-initial). This suggests that phonological motivation is not only widespread in the idiomatic lexicon as a whole that happens to be available to speakers, but that those speakers do certainly not shy away from ‘choosing’ phonologically motivated expressions for actual use.
25 percentage of repertoir
e
assonance alliteration total
20 15 10 total
5
alliteration
0 All idioms
assonance Frequent idioms
Figure 4. Alliteration and assonance in English idioms: ‘all’ vs. ‘frequent’ idioms.
Conclusions In this chapter we have argued in favour of an additional type of motivation for metaphorical idioms, i.e., phonological motivation, that goes beyond the conceptual/semantic level proposed by cognitive semanticists. The advantage of this addition is that it helps motivate the precise lexical composition of a fair number of idioms, i.e., why the
Chapter 3. Adding sound to the picture
combinations of exactly those words rather than potential (near-)synonyms make up these standardised phrases. From an SLA/FLT perspective, the advantage of this additional type of motivation lies in the before-mentioned mnemonic benefits of having learners notice the phonological patterning. We have focused on two phonological patterns, i.e., alliteration and assonance, of which especially the former turns out to be quite ‘productive’ in the languages we have examined. We feel we have been rather ‘conservative’ in our estimates (a ‘when in doubt, leave it out’ – approach), and yet our hand counts in the combined databank yield between 13% and 20% of idioms where phonological motivation is likely to have played a part in lexical selection. From an affective point of view, this may offer some encouragement to learners facing the challenge of mastering (part of) L2 phraseology, especially when the L2 happens to be English. Sceptics may argue, of course, that this still leaves about 80% of figurative idioms without phonological motivation. Some of these, however, consist of just one ‘content’ word (e.g., The works; A toss-up), in which case the question of selecting a word among (close-) synonyms to create an alliterating combination does not even arise. The comparative findings of the study hint at effects of cross-cultural variation at the level of lexical-field selection (the relative ‘productivity’ of source domains) and at cross-linguistic differences at the level of precise word selection (the relative ‘productivity’ of given phonological patterns, possibly explainable by typological reasons related to matters of word stress, inflection and word order).
Perspectives for further research Of necessity, ours was just an exploratory study that should be interpreted as an invitation for further research. After all, we have focused on only two sound patterns, in only one segment of phraseology, in only three languages, and by taking only a synchronic vantage point. The following lines of investigation seem especially appealing. •
•
•
We have found corroboration of the proposed typological hypotheses for the relative frequency of alliteration by looking at the idiom repertoires of only three languages. Obviously, more languages would need to be examined before generalised claims can be made. We have presented evidence of phonological motivation in just one segment of phraseology, i.e., metaphorical idioms. It may be worth investigating whether similar motives are at play in the lexical selection in other standardised multiword expressions, such as non-figurative idioms (e.g., spic and span), similes (e.g., dead as a dodo; drunk as a skunk), strong collocations (e.g., wage war; road rage; live and learn), etc. Ours was a comparative study of phonological patterning in a segment of standardised, conventional phraseology. It would be interesting to know whether the
Frank Boers & Hélène Stengers
•
•
same contrastive trends might be observed in the literatures of the three languages, especially in connection with the metric preferences in their poetry, drama and songs. We have implicitly considered alliteration and assonance as potential motives for lexical selection for reasons to do with ‘auditory effect’. However, the same (and similar) phenomena could also be looked at from the perspective of articulation, especially ‘articulatory economy’. Perhaps ease of articulation might be an additional driving force behind the standardisation of some multiword expressions rather than their potential (near-)synonymous counterparts. For example, one of the motives for saying It’s raining cats and dogs rather than It’s raining dogs and cats might be the ease of linking consonants that happen to be phonetically close rather than distant. The question remains whether such patterns are indeed widespread in phraseology. Ours was essentially a synchronic study of lexical selection. To lend extra strength to the hypothesis that phonological patterning is indeed a motivating factor in the way standardised phrases are formed, our study should be complemented by a diachronic study of lexical adaptation. For example, many alliterating idioms have non-alliterating variants (e.g., home and hosed vs. home and dry; come up trumps vs. turn up trumps; hard/hot on your heels vs. close on your heels; sow the seeds vs. plant the seeds; go to ground vs. go to earth; beat your breast vs. beat your chest; fill the bill vs. fit the bill). If diachronic research were to show that, when alliterating forms compete with non-alliterating ones for the status of standardised, institutionalised idiom, the alliterating ones often win out, then this finding would corroborate our phonological-motivation thesis.
Acknowledgements We would like to thank the following graduates at the Erasmus College of Brussels for allowing us to use some of the data they collected for their MA dissertations: Erika Blanckaert, Kathy Elsen and Annemie Vandiest.
References Abercrombie, David (1967). Elements of general phonetics. Edinburgh: Edinburgh University Press. Boers, Frank (2000a). Enhancing metaphoric awareness in specialised reading. English for Specific Purposes, 19, 137–147. Boers, F. (2000b). Metaphor awareness and vocabulary retention. Applied Linguistics, 21, 553–571. Boers, Frank (2001). Remembering figurative idioms by hypothesising about their origins. Prospect, 16, 35–43.
Chapter 3. Adding sound to the picture Boers, Frank (2003). Applied linguistics perspectives on cross-cultural variation in conceptual metaphor. Metaphor and Symbol, 18, 231–238. Boers, Frank & Murielle Demecheleer (1997). A few metaphorical models in (western) economic discourse. In W.A. Liebert, G. Redeker, & L. Waugh (Eds.), Discourse and perspective in cognitive linguistics (115–129). Amsterdam and Philadelphia: John Benjamins. Boers, Frank, Murielle Demecheleer & June Eyckmans (2004a). Etymological elaboration as a strategy for learning idioms. In P. Bogaards & B. Laufer (Eds.), Vocabulary in a second language: Selection, acquisition, and testing (53–78). Amsterdam and Philadelphia: John Benjamins. Boers, Frank, Murielle Demecheleer & June Eyckmans (2004b). Cross-cultural variation as a variable in comprehending and remembering figurative idioms. European Journal of English Studies, 8, 375–388. Boers, Frank & Seth Lindstromberg (2005). Finding ways to make phrase-learning feasible: the mnemonic effect of alliteration. System, 33, 225–238. Boers, Frank, Hélène Stengers & June Eyckmans (2007). Presenting figurative idioms with a touch of etymology: more than mere mnemonics? Language Teaching Research, 11, 43–62. Catford, John C. (1986). Prosody: A practical introduction. Oxford: Clarendon. Comrie, Bernard (1981). Language universals and linguistic typology. Oxford: Routledge. Dauer, R. (1983). Stress timing and syllable timing reanalysed. Journal of Phonetics, 11, 51–62. de Groot, Hans (Ed.) (1999) Van Dale idioomwoordenboek. Utrecht/Antwerpen: Van Dale Lexicografie/Reader’s Digest. Deignan, Alice (2003). Metaphorical expressions and culture: an indirect link. Metaphor and Symbol, 18, 255–272. Fernando, Chitra & Roger Flavell (1981). On idiom: Critical views and perspectives [Exeter Linguistics Studies, 5]. Exeter: University of Exeter. Gibbs, Raymond (1994). The poetics of mind: Figurative thought, language, and understanding. Cambridge: Cambridge University Press. Jiménez, Alberto B. (1997). Diccionario Espasa : dichos y frases hechas. Madrid: Espasa Calpe. Johnson, Mark (1987). The body in the mind: The bodily basis of meaning, imagination, and reason. Chicago and London: University of Chicago Press. Kövecses, Z. (1990). Emotion concepts. New York: Springer. Kövecses, Zoltan & Péter Szabó (1996). Idioms: a view from cognitive semantics. Applied Linguistics, 17, 326–55. Lakoff, George (1987). Women, fire, and dangerous things: What categories reveal about the mind. Chicago and London: University of Chicago Press. Lakoff, George & Mark Johnson (1980). Metaphors we live by. Chicago and London: University of Chicago Press. Lazar, Gillian (2003). Meanings and metaphors: Activities to practise figurative language. Cambridge: Cambridge University Press. Pamies Bertràn, Antonio (1999). Prosodic typology: on the dichotomy between stress-timed and syllable-timed languages. Language Design, 2, 103–130. Sinclair, John & Rosamund Moon (Eds.) (2000). Collins cobuild dictionary of idioms. Glasgow: HarperCollins Publishers. Speake, Jennifer (Ed.) (1999). The Oxford dictionary of idioms. Oxford: Oxford University Press. Wright, Jon (1999). Idioms organizer. Hove: Language Teaching Publications.
Frank Boers & Hélène Stengers
Appendix: some more examples of the investigated kinds of phonological motivation 1. Prototypical alliteration (repetition of word-initial consonants) look to your laurels; run someone ragged; a new lease of life; settle a score; set the scene; a busy bee; a feeding frenzy; bitten by the bug; bright-eyed and bushy-tailed; cut the cackle; have a bee in your bonnet; have bats in your belfry; make a mountain out of a molehill; neither fish nor fowl; play possum; rule the roost; the fur is flying; the rat race; turn tail; turn turtle; have other fish to fry; peas from the same pod; at sixes and sevens; below the belt; break the bank; cut corners; dice with death; get someone’s goat; pull your punches; pip someone at the post; too close to call; turn the tables; watch grass grow; a carbon copy; cut from the same cloth; shape up or ship out; the tools of the trade; cut the cord; bay for blood; read the riot act; prime the pump; pull the plug; a flight of fancy; belt and braces; done and dusted; fan the flames; fighting fit; go great guns; hide nor hair; leave someone in the lurch; meet your match; pack a punch; part and parcel; pinch pennies; practise what you preach; run riot; the bare bones; the cream of the crop; the gift of the gab; the worse for wear; without rhyme or reason; short shrift; read the runes; a close call; a leading light; broad in the beam; the coast is clear; carry the can; a war of words; at daggers drawn; a bolt from the blue.
2. Non-prototypical alliteration above board; in fits and starts; keep your pecker up; rain on your parade; froth at the mouth.
3. Prototypical assonance (rhyme) an eager beaver; a fat cat; horses for courses; pie in the sky; the bee’s knees; steer clear.
4. Non-prototypical assonance raise Cain; flip your lid; cook the books; cakes and ale; pop your clogs; pick nits.
chapter 4
Metaphor and positioning in academic book reviews Graham Low Why study book reviews Book reviews represent something of a paradox. On the one hand, they are published regularly in many of the world’s most prestigious academic journals, they can at times exert a powerful influence on a whole discipline, and they provide a channel of communication for both experienced and new academics. On the other hand, they are relatively unremarked. They are rarely cited in books or journals (Wiley 1993), rarely recognised in research rating exercises, frequently not subjected to peer review, and rarely analysed by discourse analysts (Motta-Roth 1998: 30–1; Römer 2005). Moreover, they are not always easy to read. Guo (1989) noted for example that the problems posed by metaphor, sarcasm and the indirect ways in which reviewers at times express their opinions mean that second language readers can have serious problems reading ‘between the lines’. So, while knowing how to read and write a book review is an important part of any academic’s professional toolkit, little is yet known about the tools. Motta-Roth’s (1998) analysis of 60 texts showed that academic book reviews could validly be treated as a genre characterised by four main moves: ‘Introducing the book,’ ‘Outlining the book’, ‘Highlighting parts of the book’ and lastly ‘Providing a closing evaluation of the book’. There were some differences between reviews in the three areas she studied (Linguistics, Economics and Chemistry), but these were primarily differences of content and focus. Thus, for example, chemists emphasised recency of publication and the inclusion of tables and appendices, economists emphasised mathematical modelling and linguists tended to emphasise appeal to particular groups of readers. The Motta-Roth analysis is valuable in that it suggests a coherent structure and flow of ideas underlying reviews, but it does not explain why second language learners might have problems reading them. To understand Guo’s difficulties, we need to look more at how evaluations are carried out across the text as a whole and how reviewers suggest opinions. This difference of emphasis will require a different type of analysis. One way forward is to look at reviews in terms of Positioning Theory. Positioning Theory is based on the idea that social relations are primarily constructed and maintained in terms of conversations and that the participants, or ‘social actors’, repeatedly
Graham Low
position and reposition themselves in a fluid way within and across interactions, in terms of attributes like power and believability, in order to achieve long-term and short-term goals (van Langenhove & Harré 1999a). As book reviews are by definition evaluative and texts about texts, they would seem well suited to this sort of approach. Moreover, the social constructivist approach to academic research emphasises the need for researchers to ‘sell’ their work in a ‘market’ and thus to make use of public relations and marketing techniques, in order to generate a reputation within the relevant academic community. As no-one can hope to keep abreast of everything published, reviews constitute an essential part of the marketing activity, by ‘pushing’ or ‘advertising’ the book (van Langenhove & Harré 1999b). However, reviewers need to keep or construct a reputation just as much as the original writers do, so Positioning Theory encourages the analyst to look for ways in which reviewers present themselves with respect to the readers, just as much as they react to the author and the content of the book. Claiming a position in a text means that you claim a number of rights, duties and obligations and that you recognise that readers will have certain expectations of your subsequent behaviour (van Langenhove & Harré 1999a). We may usefully expand this slightly by allowing such claims to represent what LePage & Tabouret-Keller (1985) called ‘acts of identity’ whereby individuals by the way they talk (or write) attempt to position themselves inside or outside a whole range of social and intellectual groups. Positioning can be achieved in various ways, depending whether the positioning is relative to an idea or another actor. Four commonly cited ways, however, are by redescribing something in a very selective way, by evaluating, by giving biographical details and by using indexicals like ‘I’ and ‘we’. Several of the above ideas are close to concepts developed within Conversation Analysis, such as alignment, appeals to the reader, breakdowns and repairs and interactional sequences. The primary difference between Positioning Theory and Conversation Analysis is that Positioning Theory recognises long-term strategic planning, rather than just the ‘here and now’ and does not require all inferences to be based on the actual reactions of a listener (Törrönen 2001). Positioning Theory is thus more suited to a purely textual analysis of written reviews, though the four Conversation Analysis concepts above can usefully be retained. Descriptions of Positioning Theory tend to emphasise the formal separation of initial positionings from the reactions and repositionings which build on them: van Langenhove and Harré talk of ‘first’, ‘second’ and ‘third order’ positionings. Such numbering is not always terribly informative in practice, however (and will not be used here), but the general insight that positioning reflects a layering of events and activities is an important one and provides a valuable way of describing the context underlying both the reviews and the analysis of them. 1. First, the writer writes the book and positions him/herself with respect to the material, to other research and to other researchers.
Chapter 4. Metaphor and positioning in academic book reviews
2. Second, the reviewer reads the book and reacts to some or all of the reviewer’s claims. 3. Third, the reviewer writes the review, selecting which reactions to include for an assumed readership and in the process claiming positions backwards with respect to the book and forwards with respect to the readers. 4. Lastly, readers read the review, react to the reviewer, the book/author and generate reactions of their own. The key point is that actors have differential access to these layers; the readers may or may not know of the author, may or may not have read the book and may or may not know of the reviewer. The same applies to the researcher; in this case, I know none of the authors, have read none of the books reviewed, though I do know of, and have read work by, the reviewer in Review 17. A related point is that the actors (or their work) are variably included in the layers; for example, the reviewer may or may not have been mentioned or attacked in the book and the reviewer may or may not choose to mention this. Positioning is thus likely to be complex in book reviews and to be a locus for clashes between roles such as being an expert, selling the book and helping the reader. It can be seen that Positioning Theory is somewhat broader than metadiscourse in the sense of “reader-writer interactions” (Hyland & Tse 2004: 159), or even “writers project[ing] themselves into their work to signal their communicative intentions” (Hyland 1998: 437). Applying Positioning Theory also avoids the need for the sort of discrete countable formal or functional categories which both Hyland (1998) and Le (2004) found problematic, due in part at least to the ambiguity or vagueness of pronouns like ‘we’.
The present study The present study focuses on one of Guo’s problematic areas, metaphor. Metaphor is known to be closely associated with evaluation in academic texts generally (Littlemore 2001) and also to play a major role in attempts by writers to mediate between academic research and readers who have not, or could not, read it, as in science books for children, newspaper articles about science, or reports in ‘popular’ science journals like The New Scientist (Cameron & Low 2004; Low 1997, 1999a, 2005). In this latter role particularly, metaphor has been observed to operate in collaboration with a wide range of other rhetorical devices such as extreme-case formulation,1 direct questions, alliteration,
1. The term ‘extreme case formulation’ is borrowed from Conversation Analysis. It will be used
here to include not just hyperbole and exaggeration, but cases of genuine reporting of large size or great intensity. The grouping of figurative and ‘literal’ is justified by the finding in Low (2005) that writers use both for similar rhetorical purposes.
Graham Low
humour, use of animacy metaphors, reference to familiar, everyday objects, reference to the reader and even reference to the reader’s likely reactions. Moreover, studies of oral educational discourse by Cameron (2003) and Corts & Meyer (2002) and, more relevantly to the present context, my examination of the four academic book reviews in Low (1997), found that a small number of metaphors were repeated, developed, or used systematically across stretches of text and that these ‘recurrent’ metaphors reflected positioning on the part of the reviewer.2 Indeed, the review in Nature of an edited book of brain research papers, opened with a highly complex paragraph where the reviewer positioned himself repeatedly and saliently via two intertwining recurrent metaphors (of a unfolding map and a nourishing buffet) and a complex structural system of (boundary) frames within frames. There was thus good reason to believe that, if more reviews were analysed, metaphor would be found to be used with some regularity for positioning purposes, both (a) on its own, in the form of recurrent metaphor (functioning as what Ponterotto (2003) called ‘overarching metaphor’) and (b) in conjunction with other figurative devices. To this end, a small corpus of twenty academic book reviews was constructed, half from science journals, the other half from social science journals. The three primary research questions were: RQ1: How much textual evidence is there of the academic reviewers positioning themselves (or other actors)? RQ2: How far do the reviewers employ metaphor when doing so? and RQ3: If they do use metaphor to aid positioning, what sort of metaphor is involved and how is it used?
A secondary, more reflective, question was: RQ4: Is Positioning Theory useful as a framwork for evaluating metaphor in written academic discourse?
Although Motta-Roth (1998) compared reviews in three subject areas, the focus in this paper is more general and the data from the science and social science reviews are pooled. Even so, in order to give an indication of the source of the phenomena discussed, references to science reviews will be given in bold type (e.g., R13).
Method Nine reviews were taken in July 2004 from current Social Science Citation Index journals and nine from Science Citation Index journals (see Appendix). One review came
2. The adjective ‘recurrent’ is preferred to Cameron’s (2003) ‘systematic’, as no assumptions are made about the form or function of the various exponents of the metaphor. There is also no requirement for it to be conceptually developed, as is the case with Lakoffian ‘extended metaphor’.
Chapter 4. Metaphor and positioning in academic book reviews
from a 2003 issue, simply because the 2004 one had no reviews. As Nature regularly reviews science and social science books, one of each was selected. The result was ten reviews in each subject area, totalling 18,891 words. The reviews were first read, then checked for basic quantitative data like word length, number of rhetorical sections, number of direct questions, number of references to first, second and third persons and the number of humorous items. Extremecase terms, including superlatives, were also identified (see note 1). The texts were then coded for metaphorically-used words, in the sense of Steen (2004), and Pragglejaz (2007). Delexical verbs like ‘give’ and ‘make’ were treated as non-metaphoric unless they were involved in personification (as in ‘This book gives an overview’). Specialist technical terms were checked with subject experts, though the coding decision was ultimately made by myself. Although the individual words in a simile or an A is B metaphor (like ‘Time is money’) are not technically metaphorically used, they were coded as metaphoric if the two terms of the comparison were perceived as discrepant; in the event, there were only two such examples, neither of them similes, in the entire dataset. Where a problem still existed, the Oxford Advanced Learners Dictionary was used as an aid. The percentage of metaphorically-used words, and percentage of animacy versus non-animacy-based metaphors was then calculated. The fourth stage was to identify text involving positioning. The personal references, extreme-case terms and markers of ‘pseudo interaction’ to use Wales’s (1980) term (especially direct questions) were used as an initial guide to points where positioning might be occurring. The sequences of acts of positioning emerged with repeated readings. Identification of the type(s) of positioning involved could then be made. Lastly, positioning sequences were examined to see whether they contained metaphoric expressions; as one may assume that much of the positioning in a review is both ‘intentional’ and ‘deliberate’ (in van Langenhove and Harré’s terms), so the focus was primarily on what Cameron (2003) called ‘deliberate’ metaphors. Nevertheless, words in a positioning sequence which were based on purely conventional metaphor, or even words taking on metaphoric resonance near a metaphoric expression, were checked to see if they contributed to the positioning.
Results Baseline data The main characteristics of the reviews are given in Table 1. It can be seen that the science journals tended to publish very few reviews, while the social science ones had more; indeed, the maximum (R02) was 27. The science reviews in the sample were also significantly shorter at the 5% level than the social science ones (U = 17, p = 0.013). Only three reviews (R04, R05 and R18) were formatted as academic articles, in the sense that they had a set of text-final references; all three were in social science journals.
Graham Low
Table 1. Baseline data for the 20 sample reviews Reviews N per issue
Word length Metaphor (average) density (%)
Total Social science Science
967.7 1164.2 749.3
20 10 10
4.4 7.1 1.3
10.5 11.7 9.2
Animacy density (%) 1.2 1.3 1.3
Three other reviews included full references in the body of the text and again two of these were in social science journals (R19, R20) and the third (R17) was a review of a social science book. The reviews varied in their structure; most adopted a tripartite structure of Introduction, Contents and Concluding evaluation, though no clear discipline-related pattern was found. Eight of the reviews contained no recommendation, including six where the reviewer was generally positive about the book. The indices for metaphor use showed that the greatest density (15%) occurred in a social science review and the least (6.4%) in a science review, and average difference between the groups was again significant at (U = 19, p = 0.019). Animacy expressions occurred about every hundred words and almost all involved the book being treated as a person or living organism. The proportion of animacy to non-animacy-based metaphor did not, however, differ significantly across the two groups (U = 49.5, p = 0.971), nor did the proportion of reviews containing ‘I/me’ or ‘we’ references (60% for both groups). Fourteen of the reviews were written by men and five by women (one case was unclear). The aim here is not to explore gender-related writing patterns, but the metaphoric density of the women’s texts is consistently in the lower half of the density figures and only one woman (R20) included humour (25% of the women versus 36% of the men). Interestingly the pun in R20 is related to positioning and will be discussed below. In short, although the focus here is on the 20 texts as a group, there are nevertheless several disciplinary differences between them, which would be worth exploring with a bigger corpus and from which we might perhaps infer that scientists value and/ or use reviews rather differently from social scientists.
Recurrent metaphors Apart from personification metaphors (“The book states”), there are only two examples in the corpus of recurrent metaphor which has more than two occurrences; both are from Nature (R13, R14) and both link the reviewer and the reader in a journey through the book. R13 is a review of research into the promise of extending life. Exponents of WRITING IS A PERFORMANCE, elaborated as boxing, acting and oral story telling, intertwine
Chapter 4. Metaphor and positioning in academic book reviews
with exponents of WRITING IS A JOURNEY. The reviewer begins by establishing a fight scenario in line 4, with a neutral reporting noun (“account”):
(1) . . . behind its blow-by-blow account of intrigue and scientific skulduggery, the book . . . 3
The image is made specific in the middle of the second section, where a boxing ring is implied:
(2) In the opposite corner, we encounter the enigmatic Michael West, whose career . . .
The “account” previously also became a play or melodrama at the start of the second section:
(3) Hall sets the stage by describing a visit to the home of Leonard Hayflick
Then it becomes a saga and a tale in the fourth section: “The stem cell saga, as told by Hall”, indeed “grippingly told”, and by the conclusion, “The tale is well told”. The telling is also presented as a journey in the third section and the two metaphors intertwine:
(4) The twists and turns [FIGHT or JOURNEY] of the discovery of telomerase . . . are well told [SAGA]. Along the way [JOURNEY], we learn of the squabbles [FIGHT?] too.
In sum, the metaphors develop, intertwine and play off each other from the start to the end of the text. It can be seen from the above that the review becomes explicitly interpersonal as the metaphors unfold; the reviewer as previous reader of the book aligns with the current reader of the review, to create a hypothetical, and chronologically implausible, joint team moving through the tale, as the author tells it (“we encounter”, “we learn”). Review R14, about the role of inositol rings in cell biology, introduces the further complexity of a recurrent metaphor interlocked with a similar recurrent metaphor. Both metaphors are contained within the second section, listing the contents of the book. The first is the guide/team; as in R13, the reviewer and the reader form a joint (interactive) team on a journey through the book:
(5) Sergio Grinstein and colleagues take us through the process of phagocytosis and introduce the reader to PtdIns(4,5)P2
(6) The chapter provides a comprehensive guide to the dynamic nature of the process
(7) We are given a tour of proteins
3. Underlining is used to mark metaphorically used words or lexical units. The reviews themselves are referenced as R01 – R20.
Graham Low
(8) The final chapter takes us into the nucleus
At step one, however, phagocytosis itself is described variously as a journey, a battle and birth:
(9) Cells of the immune system remove invading micro-organisms
[(9) is followed by (5) above – hence the interlocking]
(10) an accumulation of this lipid allows the recruitment of
There is then a series of process terms before the birth and journey metaphor: Subsequently. . . disappears . . . converted into . . . degraded . . . removal of PtdIns(4,5)P2 . . . leading to the dismantling of the actin network (11) which then attracts a new set of proteins to take the nascent phagosome to the next stage in its journey.
In both cases, the recurrent metaphors are used by the reviewers to deliberately reposition themselves with respect to the readers in terms of solidarity, in order to create a pseudo-joint activity. Few assumptions are made about the readers beyond a willingness to read jointly and the ability to keep up intellectually with the argument. However, although inclusive “we”s are used repeatedly, there is little evidence of either reviewer abandoning the (tacit) claim to authority and control; indeed the choice of terms like “blow-by-blow” and “opposite corner” in R13 suggest a deliberate positioning of ironic, if emotionally positive, detachment and distancing above the author.
Self-positioning in other contexts The reviewers in R13 and R14 clearly use recurrent metaphor to help them position themselves with respect to the reader(s), but the questions remain of whether selfpositioning is widespread in the other reviews and whether metaphor is used to support it. The answer to the first question is that all twenty reviews show some textual evidence of self-positioning by the reviewers. In five cases this is fairly minimal (R01, R02, R19, R09, R10) with just two or three examples; evaluations like “This is a great book” (R01) or “a quantum leap” (R19), or “one of the strengths of the book. . .” (R02), a comment like “the last chapter, which is typically planned as a review” (R09), or the suggestion that adding some critical discussion might help readers (R01), flag that the reviewer has expertise in the field, is well-read and, in the last case, is less likely to be confused than the readers. Nevertheless, in all cases reviewers make some effort to position themselves above the readers in terms of academic expertise, and thus status. We might also expect reviewers to position themselves below the author, where the book or author is admired, but above him or her where the book is found wanting. Although there are two examples of humility in the corpus (R20 and particularly
Chapter 4. Metaphor and positioning in academic book reviews
R19 – discussed in more detail below), the situation is not straightforward and the reviewer still tries to be ‘on top’. R08 provides a good example of how reviewers tend to react even in the face of a book that is “superb”, “original”, “a remarkable achievement” and metaphorically “a major step along the road towards accurate quantification”. Despite the explicit admiration, the reviewer takes care to lay claim repeatedly to at least five related roles or characteristics (relevant words are bolded): A member of the relevant research community: (12) Which Dudarev was perhaps the first to bring into our field from earlier Russian work
Having expertise: (13) Certainly the most thorough, unified and comprehensive treatment . . . to appear for many years. (14) All the relevant theoretical background is given here.
Experienced: (15) The chapter on RHEED summarizes the major contribution that this group of authors has made to that field over many years
Well-read: (16) The best treatment of their angular dependence to be found in any textbook (17) It is . . . unusually clearly written and complete.
Up-to-date: (18) In biology, too, there are now renewed efforts to go beyond the simple Beer’s law . . .
In general, claims to authority in the corpus show a frequent use of extremecase formulation, which in turn implies a more general claim to positioning oneself at the head of the field, by implication close to, or even above, the authors. From a metaphor point of view, the claims to authority rarely go beyond conventional expressions like “bring our field” (R08) or “strengths” (R02), except where an academic advance is flagged (“a step along the road” (R08), “a quantum leap” (R19)). From a Positioning Theory point of view, we may also note that the claims are in most cases implied rather than explicit, meaning that it is frequently unclear whether the positioning is tacit or deliberate. R08 is somewhat different, in that the combination of claiming five characteristics, the frequent extreme-case formulation and the repetition of the claims throughout the text strongly suggests deliberate positioning.
Graham Low
The reviewers at times go further than simply claiming authority and in at least three cases (R04, R18 and R20) directly compete with the authors. R18 and R20 are particularly interesting in that they represent rather different approaches to competition. In R20, the reviewer begins by building up the book and the author, using extreme case terminology: (19) I recommend this book very highly (20) A specialist in American religion (21) . . . travelled to 14 carefully chosen and very diverse places (22) . . . paying particular attention to (23) These pages are a tour de force in themselves
Within this list there is a short joke, in the form of a play on words, presumably by the reviewer: the author travels to “two multi-religious “mindfields” (Israel and India)”. Unlike many of the jokes in the corpus, it is not targeted against the author or the reader, but would appear to serve to show that the reviewer is up to date in the field and feels secure enough to share a joke; I interpret this, at least on one level, as a gesture of solidarity with the author. After a brief description of part two of the book, the reviewer devotes a whole paragraph to describing and evaluating her own work. She begins by claiming part of the field as the equal of the author: (24) I have attempted the same exercise myself. Europe: the Exceptional Case. Parameters of Faith in the Modern World (publisher details), in many respects does for Europe what [author] does for the USA.
There is then a short description of what the reviewer (not the author) did. The claim to equality is, however, toned down in three ways: (a) the description is followed by an apology (“I have to confess”), (b) there are two overt statements about things the author did better than the reviewer, and (c), more relevantly for present purposes, the description itself is couched in third person terms, using personification metaphor: (25) It looks at Europe from the outside rather than within and asks not only what Europe is in terms of its religious life, but (importantly) what Europe is not.
However, the toning down (or humility) effect of the personification metaphor would seem to be somewhat reduced by the claim implied by the adverb “importantly”. The reviewer therefore ultimately draws back from the claim to equality, but retains the claim to be a European expert. This claim presumably warrants the selective focus on UK-related errors, with its disparaging “such” and emphasis on clearly demarcating areas of expertise: (26) Just occasionally, this embeddedness of such ideas in the mind of an American scholar leads to a misreading of the comparative cases. The comments on England, for example, . . ..
Chapter 4. Metaphor and positioning in academic book reviews
The reader remains uncertain whether the foregrounding of the reviewer’s expertise at the expense of the author is part of the earlier claim for equality, a way of repairing a problem with evaluating non-European data, or conceivably both. This is in fact a good example of where ‘communicative intention’ is unclear. The reviewer in R18 is in a slightly different position; she would appear to not to be currently in direct competition with the author, but rather to imply that she would like to be. She admires the book, but is not included in it. She also admits to learning from it, and aligns herself with the reader, though in a more self-abasing way than the ‘joint reading team’ reviewers in R13 and R14 (above): (27) Children, the editors tell us, . . . (28) Nevat-Gall’s chapter shows us how . . . (29) Similarly, Fusilo, Lieberati, and Ponte-Corvo’s chapter shows us that. . . (30) Kasuya’s report informs us that . . .
By the end of the review, the reviewer has effectively positioned herself below the authors, and away from the elite in-group of researchers. The two-paragraph conclusion is accordingly used to repair the situation. First, an apparently neutral comment is made that the book “provides considerable food for thought for further research on these topics”. The metaphor would appear to serve a generalising, decoupling and topic-changing, or at least refocusing, function, of the sort described by Drew & Holt (1998) for conversation. This is followed by five direct questions suggesting unknowns. Second, the questions are followed by a sort of answer, involving a self-reference and a fairly conventional metaphor, which acts as the climax to the paragraph: (31) Although the work of Blum-Kulka (2003), Cook-Gumperz (1995), Goodwin (1990), Ervin-Tripp and Kuntay (1997), and Kyratzis (1999) has begun to tackle these problems, more contexts of unstructured, spontaneous peer talk among children need to be considered, where these are applicable. . .
The reviewer references herself in a diplomatic yet clear-cut way, as active, up-andcoming and a member of the elite group (along with the authors) researching the topic. The self-reference, in other words, appears to be more than simply referencing a third party in support of an argument (an ‘evidential’ within Hyland’s ‘textual metadiscourse’ or Hyland & Tse’s ‘interactive resources’). The link with the authors is then firmed up in the short final paragraph of the review, where the reviewer, it seems, hopes to publish in the same series. The reader has to infer this, as the claim is again diplomatically couched, exploiting the ambiguity of the third person “one”: (32) The topic . . . is essential for research in language development . . . One can hope that this book will serve as the carefully conducted initial set of analyses of a long series to take on this important problem.
Graham Low
It is axiomatic in Positioning Theory that acts of positioning have both a moral (or institutional) and a personal component (van Langenhove & Harré 1999: 20). In both the above challenges to the author, the reviewers employ personal arguments, such as the explicit biographical detail in R20 and the reference to it in R18. However, at no point does either reviewer depart from a professional academic role, so the point where the moral becomes the personal is unclear; indeed the two would not seem to be necessarily mutually exclusive. Metaphor occurs in both R20 and R18, but its role in furthering the challenges is subtle. The use of “embeddedness” in R20 to emphasise the claim of restricted expertise and of personification to tone the challenge down are fairly transparent, but the use of the conventional “tackle” and “take on”, along with “serve” and “long”, in R18 to emphasise the challenge and validate the reviewer’s future effort is somewhat more debatable. The use of metaphor is somewhat clearer where the author overtly admits to being in a difficult status position with respect to the author and feels the need to repair the situation. There are two examples of this in the corpus, both from science reviews (R07 and R14). I will use the first to illustrate the point. The R07 reviewer begins his review by directly admitting ignorance and fallibility: (33) Neonatologists are not always directly involved in the intensive care of neonates as surgical patients. In my own case this has led to a slightly blinkered approach. I am very familiar with perinatal stabilisation of problems such as gastroschisis, with the intensive care of infants with diaphragmatic hernias, and with the referral of infants with less acute problems. However, perioperative management, particularly of uncomplicated cases, and the mysteries of operative techniques have been beyond my reach. A book, with neonatologists within its scope, ideally with strong emphasis on presentation, embryology, and associations as well as describing surgical options, would plug a significant gap in my knowledge.
However, it may be noted that the lowering of the reviewer’s status is only relative and is accompanied by hedging (“slightly blinkered”), a firm explicit statement of expertise using extreme case terminology (“I am very familiar”), a joke (“mysteries”) and a string of three unlinked metaphor groups (“blinkered”, “beyond my reach” and “plug . . . gap”). The effect is to create a conversational tone and a matey relationship with the reader, but the positioning (i.e., the recovery or repair) is transparent, as is the supporting role of the metaphors. The rhetorical position adopted by the reviewer, of being a genuine reader, allows him to admit – at times using metaphor to do so (see second example below) – to feeling humbled before parts of the text: (34) Fascinating drawings . . . left me wondering that such a complicated operation could be so simply described. (35) The five sides dedicated to respiratory management emphasised to me the potential rewards to be reaped from closer integration . . .
Chapter 4. Metaphor and positioning in academic book reviews
Unfortunately, the role of humble reader proves ultimately too hard to maintain, and the emotion and the metaphor are used at the end to restore expertise: (36) I’m glad to say that this book is the one to plug the gaps in my knowledge
Despite the repeated, and presumably quite genuine, extolling of collaboration, the reviewer’s approach to the actual readers is not quite so humble. The recommendation in the final sentence to “fellow pediatricians” and the more distanced “I would encourage surgeons and neonatologists” is given a slight edge by the previous paragraph’s comment that: (37) We have found the book valuable in furthering our understanding of the problems we see on a day to day basis.
The “we” refers solely to the reviewer and his team, not to the reader. The reviewer may be a reader, but he is not a reader like the colleagues to whom the book is recommended; he is someone who has already benefited and learned, and is thus somewhat above the other (still relatively ignorant) readers. The claim remains implicit rather than explicit, however, so it is again unclear whether tacit or deliberate positioning is involved, though (as with R08) the repetition, this time of metaphoric expressions, might suggest it was deliberate. The reviewer consistently foregrounds the personal throughout R07, using biographic detail about himself and (by implication) the readers, but as with R20 and R18, he does so without departing from the institutional role of informed academic. The boundary between the moral and the personal thus remains blurred.
Positioning the reader and third parties Many of the self-positionings described above inevitably serve to position the reader at the same time. Where the reviewer aligns with the reader via a journey metaphor to create a joint reading team, for example (as in R11, R13, R14), this necessarily positions the readers as relatively informed and willing to be led. Again, when the reviewer in R08 discusses his own ignorance, he is positioning many of his colleagues as similarly lacking, However, in all cases in the corpus of alignment with the reader the reviewer strives to maintain the upper hand. This may simply involve a statement implying that the reviewer knows better than the reader, (38) . . . readers will need to look elsewhere for information on . . .. (R12)
or it can involve actively ordering the reader around:
(39) Readers will find the chapter on woody plant DNA marker analysis to be useful (R12) (40) Markers have many useful roles . . . But we should not . . . lose sight of the central concern, which is expressed variation. (R11)
It can even, as in R08 or R14, involve putting the reader down (in the latter text via a humorous allusion): (41) The field is moving rapidly but this volume will be a useful addition for many researchers who have only recently discovered that their favorite protein is regulated by PtdIns. (R14)
Graham Low
At other times the reader appears to be called up by the reviewer to help further the argument, or even to help attack the author (R03, R06, R15, R16). In these instances, the reviewers are happy to foreground themselves when it comes to offering praise, but tend to let other actors do the criticism. I shall use R03 as the example. The reviewer in R03 begins, rather like the reviewer in R20 (above), by praising the author and using a short joke to align with him and demonstrate membership of the field (“Chapman draws on . . . the human-environment approach for which geographers remain (in)famous to analyse. . ..”). “Geographers” may or may not include the readers of the review! The alignment is reinforced by the use of the metaphor “a prism” to flag the light and truth that Chapman seeks. A range of possible readers is then pulled out of the hat to make objections and locate the errors in Chapman’s arguments. The reviewer backgrounds himself by politely suggesting hedges or using pronouns (like “one”) with extremely vague reference: (42) Specialist South Asianists would be justified in challenging . . .. (43) Can one justifiably translate the Manusmriti as the ‘Laws of Manu’ . . . ? (44) The reader is left . . . without much understanding of the . . . dynamics that have shaped its rich and complex philosophy.
That these agents are simply proxies for the reviewer is made clear from the use of the metaphoric “shaped” and “rich”, which can only be interpreted as the reviewer’s own opinion and read as making a claim to expertise about Indian culture. In the next but one sentence, the reviewer also tries to repair the author’s credibility by using another metaphor, “lopsidedness” (and possibly “reflection”): (45) This lopsidedness, as Chapman explains, is a reflection of his own “Britishness”
The criticism situation then itself becomes complex, as the reviewer criticises academics who want “a more ‘critical’ analysis”, while sympathising with them, but avoids serious criticism of the much praised author. The result is the rhetorical equivalent of trying to have your cake and eat it. (46) Chapman may also be criticised by theorists of political geography / geopolitics . . . To be fair, Chapman does acknowledge the situatedness and contextuality . . . but steadfastly avoids getting bogged down in any deconstructivist analysis, presumably to maintain the smooth flow of his narrative.
Again, the metaphoric “avoid getting bogged down” and “maintain smooth flow” quietly restore the author’s status, as well as reflecting the reviewer’s expertise. Interestingly, “bogged down” echoes “the quagmire” in the Introduction, where the reviewer
Chapter 4. Metaphor and positioning in academic book reviews
sets the scene by stressing a universal problem, and thereby initiating the backgrounding of his personal accountability: (47) South Asia has been subject to a growing plethora of research, leaving students and academics alike in the quagmire of data overload.
It may be wondered why an enthusiastic reviewer needs to keep his head down rhetorically, particularly when several of the other reviewers actually shift to a personal “I” when it comes to flagging omissions and errors (e.g., R04, R11, R19). The answer may lie in the comment about Chapman’s work near the end, that: (48) This point has still not been fully recognised or valued in geography and social science study generally.
The reviewer would thus appear to be concerned to maintain his own reputation, by positioning himself as someone at the academic cutting edge, yet as someone who is nevertheless part of the mainstream. There are several examples of the reviewer using a joke with the reader to attack the author, and in one case, to attack the person the author was writing about as well (R15), but perhaps the ultimate use of the reader plus third parties to attack the author occurs in the two really negative reviews – one from science (R16), the other from social science (R06). The techniques are different, but the impact in both cases is that of a tidal wave. R16 is a reaction against the claim that “our planet is significant, perhaps being uniquely situated to foster both complex life and scientific discovery”. The review has four parts: (1) an introduction outlining Carl Sagan’s (desirable) position that we are ignorant, and stating the essence of the authors’ contrasting position, (2) a middle section giving details of the authors’ claims, (3) a section refuting the claims and (4) a brief conclusion summarising Sagan and the refutations. The Sagan quote at the start (part 1) contains three “our”s and a “we” implying all humanity – to include the reviewer, the readers, the authors and everyone else. The section focusing on the claims in the book (part 2) has no personal references at all. The objection section (part 3) has four “we”s and three “our”s, and the brief conclusion (part 4) has two “we”s and an “our”. Quite who the “we”s include is vague in the extreme, though it is clear that “our life” and “our science” are contrasted with life on other planets. The reviewer’s aim appears to be to make the objections in the name of humanity itself, and as this includes the authors, the repetition of the “we” takes on an ironic tone. This interpretation is given partial support by the sarcastic use of a quote by the authors, “the wonder of this book”, at the end of part 1, and by the ironic use of (positive) extreme case terminology in part 2. A further irony is that readers are of little interest as readers (at least of the book), since the reviewer does not want to encourage people to read the book; thus despite the tidal wave of attacking “we”s, the readers are actually addressed hypothetically and in the third person: (49) Potential readers of [book] would do well to first familiarize themselves with the biases that can result from this kind of selective sampling.
Graham Low
Though parts 1 and 4 do involve some metaphor mixed in with the positioning “we”s, the expressions “lonely speck in the great enveloping cosmic dark” and later “our blue dot” serve primarily to frame the start and end of the text. Beyond this there is minimal functional use of metaphor – just the back-handed compliment “the authors attempted to avoid cherry picking . . .”; the R16 reviewer positions himself more by the ironic use of extreme-case formulation. The other review, R06, is from a Philosophy journal, so one might expect a slightly different rhetorical approach, though in fact the reviewer is strenuously arguing in favour of ‘hard’ quantitative predictive models in the social sciences. Topically, the review has three (possibly four) parts, but interactively and functionally it has, like the book being reviewed, two main parts. Part one (the first four paragraphs) describes what the author alleges/claims, and the reviewer makes his primary objection in an impersonal, unemotional, non-interactive way: (50) The problem with this view is not . . . Rather the problem with this book is found at the beginning, in the diagnosis of the current condition of social science. (51) Here one needs to examine the author’s argument that . . .
From paragraphs 5 to 10, however, the style shifts radically and the reader is faced with thirteen direct questions plus two “Surely . . ..” statements attacking the author and his views. The situation is similar with respect to pronouns; the first four paragraphs have just two “one”s and a “we” that includes everyone but the author. But now paragraph 6 alone, with five questions, has two “we”s, an “our” and a “one”, and paragraph 10, the conclusion with the last three questions, has three “we”s and two “our”s. The “we”s appear to refer to everyone, or all social scientists, and to include the reader: (52) How can one possibly succeed in improving our society without making some prior attempt to gather knowledge concerning why humans behave as we do? (Question 5)
One may reasonably ask who the questions involve and who they might be addressed to. Questions like “Does the author forget . . ..?” (Question 1) and “What would the author say . . .?” (Question 3) seem as though they are addressed by a barrister to a court or jury about an author in the dock. This interpretation is reinforced by the multiple use of negative questions “Wouldn’t . . . ?” and the “Surely. . .” statements. The reader is thus positioned as either judge or jury. Several questions marshal third parties, both people and disciplines: “What would the author say to quantitative political scientists . . .?” (Question 3), “Are economists deluding themselves . . . ?” (Question 6), “Is it any wonder that so few social scientists take philosophical debate very seriously?” (Question 10), or “. . .. Wouldn’t much of modern meteorology and even much of quantum physics fail to measure up?” (Question 8). In the conclusion, the reviewer foregrounds himself a little and offers an answer (“No”) to Question 12 and Question13, but the others remain as questions.
Chapter 4. Metaphor and positioning in academic book reviews
There are several salient metaphors in this second part of the review. There are occasional one-off examples (“medicine . . . evolved out of its trappings” (Question 1)), but essentially they form two clusters. The first is in paragraph 8, immediately ahead of the general Question 10. (53) [the book] pretends single-handedly to resolve a debate that is still raging in the philosophy of social science. Worse is the arid assumption that with a little Foucault and Habermas, one can show that whole disciplines in the social sciences melt away as pure methodological bravado. Is it any wonder that so few social scientists take philosophical debate very seriously? (Question 10)
With its two increasingly negative sections (“pretends” → “Worse”), ending on a short three-word phrasal unit employing words that are polysyllabic, academic and metaphoric (“pure methodological bravado”), the section moves steadily towards a climax (Low 1999b). The combination of personification, fire, drought, dissolving and purity metaphors (plus the contemptuous metonymy “a little Foucault and Habermas”) serve to build up intensity prior to the attack question and thus to justify the question when it appears. The second metaphor cluster is, unsurprisingly, in the conclusion, where (54) We should not expect a revolution at all in a field until it has self-consciously embraced not just the ambitions and trappings of science, but also the scientific attitude . . .
And (55) Though it is easy to despair, there are signs that such a vision is just beginning to take hold and bear fruit in the social sciences.
The function of the metaphoric expressions here is not so much to support the attack, but, as with R16, to frame the text as a whole. Even so, although the FRUIT TREE metaphor (“take hold and bear fruit in the social sciences”) makes a significant contribution to the final climax, it still emphasises the reviewer’s own opinion and his up-todate expertise and thus contributes to the reviewer’s self-positioning.
Conclusion I shall begin by summarising the answers to the four main research questions. The twenty reviews in the corpus differ markedly in the amount of positioning by the reviewer – that is to say, positioning which can be identified without knowledge about (activity) Layers 1 and 2: the book reviewed, the author, and the reviewer’s choice of what to include and exclude. Reviewers consistently claim a position of authority, asserting or implying that they are well-read, up to date, experienced, a member of the relevant research community and/or having relevant expertise.
Graham Low
Where such status is problematic, repair procedures are invoked to rectify the situation. All the reviewers position themselves with respect to the author, and most position themselves with respect to the reader(s). Where criticism or challenges are involved, however, the reviewers may well marshal third parties, as well as the author and the readers, to share the arguing. Even in cases where the book is praised, however, the reviewers generally attempt to position themselves above the authors as well as the readers. Positioning is achieved by a range of linguistic devices, but particularly extremecase formulation, emotional lexis, humour, direct questions, indexicals and metaphor. This situation, where metaphor is but one of a range of rhetorical tools, is similar to that found in popular or less-formal science explanations (Low 1997; Cameron & Low 2004) and closely resembles the situation in oral conversation noted by Carter & McCarthy (2004) in their discussion of creativity. Although jokes (in ten reviews) and references to the actors involved (“we/our” in twelve reviews and “I/my” in seven) might be expected to result in a highly interactive, even conversational, style, in fact all the reviewers maintain a degree of distance between themselves and the readers. Only one text (R13) addresses them directly (with just two personal “you”s). References to “we” and “us” are frequently vague or ambiguous, as with Le’s (2004) broadsheet editorials. Koller (2003) noted that some of her business texts seemed to be talking to third parties, rather than the reader and the same is found here; many comments about omissions and errors, and even the forensic questioning in R13, appear to expect a response from the author, or third parties, rather than the readers. Metaphors contribute to positionings in various ways; they flag reviewer expertise and status, they highlight points where reviewer status is problematic, they help repair lapses of status, they mark changes of focus, they evaluate the author, and they intensify attacks (or help make them more credible), though conversely, they can also tone them down. Having said that, the contribution of metaphor to many of the claims to authority by the reviewer is purely via conventional metaphors, unless an academic advance is suggested. In short, the use of metaphor to aid positioning is considerably less common and less structured across the corpus than was predicted at the start. While Corts and Meyers (2002) found recurrent metaphors recurring in waves (or clusters) throughout their oral sermon data, and Low (1997) found examples in two of the four academic book reviews, the situation does not apply here. Beyond the use of personification metaphors (“The book says”), there are only two examples of recurrent metaphor (R13, R14) – though both play a part in the reviewers’ self-positionings. The result is that metaphoric expressions and underlying metaphors, in this corpus at least, tend to be one-off items, even when they work together in a cluster. The limited use of metaphor to aid positioning has to be a deliberate rhetorical choice, as the general incidence of metaphorical words in the reviews is not particularly
Chapter 4. Metaphor and positioning in academic book reviews
low, ranging from 6% (R16) to 15% (R02), with an average of 10.5%. The answer may simply be that all reviewers felt constrained by the need to retain a degree of formality in academic written texts, and that formality implied limiting the use of deliberate metaphor. Alternatively, the result could simply be the result of having a fairly small sample. Either way, the lesson for prospective review writers is that, where positioning is involved, control over extreme-case formulation, emotional lexis and humour is as important as, if not more important than, (functional) control over metaphor. Finally, it is possible to give an answer to the secondary research question, on the usefulness of Positioning Theory to studying metaphor in written discourse. The use of Positioning Theory plus a small number of functional concepts from Conversation Analysis proved a fruitful way of describing the basic context of both the reviews and the analysis itself, as well as providing a practical initial means of locating examples of positioning and of describing how these flow through the texts. This made it possible to explore the complexities and the different ways in which the authors positioned any or all of the possible social actors, in a way that would have been difficult using Conversation Analysis alone, or using a metadiscourse analysis which relied on a clear distinction (a) between addressees and (b) between the message and the writer’s attitude to it (between ‘textual’ and ‘interpersonal’ metadiscourse, or between ‘interactional’ and ‘interactive’ in Hyland & Tse’s (2004) terminology). The only ‘problems’ encountered were the difficulty at times of distinguishing between tacit and intentional positioning and between personal and moral/institutional modes. The former is hardly a ‘real’ problem, as writers frequently disguise their intentions and readers just as frequently ask themselves if something was intended. The conflation of personal and moral, however, may be something particularly associated with positionings in academic texts. There have to date been few studies employing Positioning Theory as a means of examining metaphor in discourse, beyond Berman’s (1999) study of national identity in Indonesia, but the method would seem to have distinct advantages and be well worth further investment.
References Berman, Laine (1999). Positioning in the formation of a ‘national identity’. In L. van Langenhove & R. Harré (Eds.), Positioning theory (138–159). Oxford: Blackwell. Cameron, Lynne (2003). Metaphor in educational discourse. London: Continuum. Cameron, Lynne & Graham Low (2004). Figurative variation in episodes of educational talk and text. European Journal of English Studies, 8, 355–373. Carter, Ronald & Michael McCarthy (2004). Talking, creating: Interactional language, creativity and context. Applied Linguistics, 25, 62–88. Corts, Daniel P. & Kristina Meyers (2002). Conceptual clusters in figurative language production. Journal of Psycholinguistic Research, 31, 391–408.
Graham Low Drew, Paul & Elizabeth Holt (1998). Figures in speech: The role of figurative expressions in topic transition in conversation. Language in Society, 27, 495–523. Guo, Hongyi (1989). A tentative study of academic book reviews: Stylistic features and rhetorical functions. MA dissertation, University of York. Hyland, Ken (1998). Persuasion and context: The pragmatics of academic metadiscourse. Journal of Pragmatics, 30, 437–455. Hyland, Ken & Polly Tse (2004). Metadiscourse and academic writing: A reappraisal. Applied Linguistics, 25, 137–155. Koller, Veronika (2003). Metaphor clusters, metaphor chains: Analysing the multifunctionality of metaphor in text. Metaphorik, 05/2003, 115–134. Le, Elizabeth (2004). Active participation within written argumentation: metadiscourse and editorialist’s authority. Journal of Pragmatics, 36, 687–714. LePage, Robert & Andrée Tabouret-Keller (1985). Acts of identity: Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press. Littlemore, Jeannette (2001). The use of metaphor in university lectures and the problems it causes for overseas students. Teaching in Higher Education, 6, 333–349. Low, Graham (1997). A celebration of squid sandwiches: Figurative language and the management of (non-core) academic text. University of York: Project report, pp. 46. Low, Graham (1999a). Teaching learners to use metaphor. In L. Kasanga (Ed.), Languages and the new democracy: participation, rights and responsibilities (39–63). Sovenga SA: University of the North Press. Low, Graham (1999b). Teaching foregrounding skills in academic text. In L. Kasanga (Ed.), Language(s) and the New Democracy: Participation, Rights and Responsibilities (103–123). Sovenga, SA: University of the North Press. Low, Graham (2005). Explaining evolution: The use of animacy in an example of semi-formal science writing. Language and Literature, 14, 129–148. Motta-Roth, Désirée (1998). Discourse analysis and academic book reviews: A study of text and disciplinary cultures. In J. Fortanet, S. Posteguillo, J.C. Palmer & J.E. Coll (Eds.), Genre studies in English for academic purposes (29–58). Castelló, Spain: Universitat Jaume I, Servei de Publicacions. Ponterotto, Diane (2003). The cohesive role of cognitive metaphor in discourse and conversation. In A. Barcelona (Ed.) Metaphor and Metonymy at the Crossroads. A cognitive perspective (283–298). Berlin: Mouton de Gruyter. Pragglejaz (2007). MIP: A method for identifying metaphorically-used words in discourse. Metaphor & Symbol, 22, 1–39. Römer, Ute (2005). ‘This seems somewhat counterintuitive, though. . .’ – Negative evaluation in linguistic book reviews by male and female authors. In E. Tognini-Bonelli & Del Lungo Camiciotti (Eds.), Strategies in academic discourse (97–115). Amsterdam: Benjamins. Steen, Gerard (2004). What counts as a metaphorically used word? The Pragglejaz experience. In S. Coulson & B. Lewandowska (Eds.), The literal-nonliteral distinction (299–322). Berlin: Peter Lang. Törrönen, Jukka (2001). The concept of subject position in empirical social research. Journal for the Theory of Social Behaviour, 31, 313–329. van Langenhove, Luk & Rom Harré. (1999a). Introducing positioning theory. In L. van Langenhove & R. Harré (Eds.), Positioning theory (13–41). Oxford: Blackwell. van Langenhove, Luk & Rom Harré (1999b). Positioning and the writing of science. In L. van Langenhove & R. Harré (Eds.), Positioning theory (102–115). Oxford: Blackwell.
Chapter 4. Metaphor and positioning in academic book reviews Wales, Katy (1980). Exophora re-examined: The uses of the personal pronoun ‘we’ in presentday English. UEA Papers in Linguistics, 12, 21–44. Wiley, Mark (1993). How to read a book: Reflections on the ethics of book reviewing. Journal of Advanced Composition, 13, 477–492.
Appendix: The corpus of academic book reviews Social Science (SSCI) R01 Busher, Hugh (2004). Review of Doing a doctorate in educational ethnography. British Educational Research Journal, 30, 463. R02 Kucera, David (2003). Review of, Labour market and social protection reforms in international perspective. International Labor Review, 142, 373–375. R03 Menski, Martin (2004). Review of The geopolitics of South Asia. From early empires to the nuclear age. Political Geography, 23, 493–495. R04 Braunstein, Yale, M. (2004). Review of Cost proxy models and telecommunications policy: A new empirical approach to regulation, Cambridge, MA: MIT Press. Information Economics and Policy, 16, 311–313. R05 Sloboda, Brian (2004). Review of Applied time series modelling and forecasting. International Journal of Forecasting, 20, 137–139. R06 McIntyre, Lee (2004). Review of Making social science matter: Why social inquiry fails and how it can succeed again. Philosophy of Science, 71, 418–421. R18 Kyratzis, Amy (2004). Review of Talking to adults: The contribution of multiparty discourse to language acquisition. Journal of Pragmatics, 36, 807–811. R19 Ashkanasy, Neal M. (2004). Review of The future of competition: Co-creating unique value with customers. The Academy of Management Executive, 18, 155–157. R20 Davie, Grace (2004). Review of Crossing the gods. The British Journal of Sociology, 55, 2, 301.
Science (SCI) R07 Oddie, Sam J. (2004). Review of Newborn surgery, 2nd ed. Archives of Disease in Childhood Fetal and Neonatal Edition, 89, F373. R08 Spence, John (2004). Review of High energy electron diffraction and microscopy. Acta Crystallographica Section A, 60, 355. R09 Sarto, Cecilia & Cristina Valsecchi (2004). Review of Molecular analysis of cancer. Proteomics, 4, 2203. R10 Mills, Terence, C. (2004). Review of A companion to theoretical econometrics. Journal of Time Series Analysis, 25, 315–316. R11 Pemberton, Josephine (2004). Review of Introduction to conservation genetics. Genetical Research, 83, 221–223. R12 Kane, Michael E. (2004). Review of Micropropagation of woody trees and fruits. Journal of Plant Physiology, 161, 887. R13 Kirkwood, Tom B.L. (2004). Review of Merchants of immortality: Chasing the dream of human life extension. Nature Medicine, 10, 575.
Graham Low R14 Cockcroft, Shamshad (2004). Review of Phosphoinositides in subcellular targeting and enzyme activation. Nature Cell Biology, 6, 481. R15 Kruger, Lawrence (2004). Review of Soul made flesh: The discovery of the brain – and how it changed the world. Nature Neuroscience, 7, 685. R16 Vakoch, Douglas A. (2004). Review of G. Gonzalez & J.W. Richards, The privileged planet: How our place in the cosmos is designed for discovery. Nature, 429, 808–809.
SCI journal, but social science book and social science reviewer R17 Sweetser, Eve (2004). Review of Hearing gesture: How our hands help us think. Nature, 429, 606.
part ii
Examining metaphor in corpora
chapter 5
Brothers in arms Contradictory metaphors in contemporary marketing discourse Veronika Koller Over the past 15 years, marketing as a social practice has undergone a fundamental shift. This shift has been driven by increased competition in a globalised market and concomitant shorter product life cycles. Along with product development and marketing, consumption has been accelerated, too, leading to decreased customer loyalty and thus putting pressure on sellers to gain and retain what loyalty there still is. Furthermore, e-commerce has brought about a “de-intermediation process” (Parvatiyar & Sheth 2000: 13) bringing buyer and seller closer together again. As a consequence, the buzzword these days is “Relationship Marketing”. Although the term actually made its debut in the early 1980s already (Berry 1983), it was only a decade later that the new approach caught on in theoretical writings (Peterson 1995; Reichheld 1996; Shani & Chalansani 1992; Sheth & Parvatiyar 2000). Relationship Marketing is perhaps best characterised by contrasting it with the more time-honoured concept of Transaction Marketing, which dates back to the 1960s (Kotler 1967), as summarised in Table 1: Table 1. Comparison Transaction and Relationship Marketing Transaction Marketing
Relationship Marketing
product customer acquisition short-term tactics profit maximization external competition as rivals product
customer customer retention long-term strategy mutual benefit internal competition as partners brand
Of further importance is the transition from tangible products to intangible brands, the importance of which is reflected in the fact that already more than 60 per cent of the value of fast-moving consumer goods (such as cosmetics or soft drinks) is accounted for by their brand value (Grauel 2000: 18). Brands as intangible entities represent the cognitive-affective concepts consumers hold about a particular
Veronika Koller
product or service.1 Consequently, it has become the task of marketing to “establish a strong and stable set of associations in the minds and memories of consumers” (Moore 2003: 335), i.e., to communicate and reinforce particular concepts through, for instance, advertising. Ideally, the concepts communicated will tie in seamlessly with th