Metaphor and Discourse Edited by
Andreas Musolff and Jörg Zinken
Metaphor and Discourse
Also by Andreas Musolff METAPHOR AND POLITICAL DISCOURSE MIRROR IMAGES OF EUROPE ATTITUDES TOWARD EUROPE (co-editor) DISCOURSES OF INTERCULTURAL IDENTITY (co-editor)
Metaphor and Discourse Edited by
Andreas Musolff University of Durham
Jörg Zinken University of Portsmouth
© Andreas Musolff and Jörg Zinken 2009 Chapters © their individual authors 2009 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2009 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin's Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–53730–9 hardback ISBN-10: 0–230–53730–8 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Metaphor and discourse / edited by Andreas Musolff and Jörg Zinken. p. cm. Includes bibliographical references and index. ISBN 978–0–230–53730–9 1. Metaphor. 2. Discourse analysis. I. Musolff, Andreas. II. Zinken, Jörg. P301.5.M48M462 2008 401′.41—dc22 10 9 8 7 6 5 4 3 2 1 18 17 16 15 14 13 12 11 10 09 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne
2008030687
Contents Acknowledgements
vii
Notes on Contributors 1
viii
A Discourse-Centred Perspective on Metaphorical Meaning and Understanding Jörg Zinken and Andreas Musolff
1
Part I Metaphor in Discourse: Theoretical and Methodological Perspectives 2
3
Metaphor, Culture, and Discourse: The Pressure of Coherence Zoltán Kövecses
11
Three Kinds of Metaphor in Discourse: A Linguistic Taxonomy Gerard Steen
25
4
Reading Sonnet 30: Discourse, Metaphor and Blending Paul Chilton
40
5
Collecting Political Meaning from the Count of Metaphor Christ’l De Landtsheer
59
6
Metaphor and Context: A Perspective from Artificial Intelligence John Barnden
79
Part II Metaphors in Contemporary Public Discourses: Case Studies 7
Metaphor and Political Communication Jonathan Charteris-Black
8
Missions and Empires: Religious and Political Metaphors in Corporate Discourse Veronika Koller v
97
116
vi
9
Contents
How Business Press Headlines Get Their Message Across: A Different Perspective on Metaphor Michael White and Honesto Herrera
10 MRSA – Portrait of a Superbug: A Media Drama in Three Acts Brigitte Nerlich and Nelya Koteyko
Part III 11
135
153
Metaphor Evolution in Discourse History
Shifting Identities: Metaphors of Discourse Evolution Roslyn M. Frank
173
12 ‘Neither a borrower nor a lender be’: Linguistic Mercantilism in Renaissance France David Cowling
190
13 Interpretations of the Body Politic and of Natural Bodies in Late Sixteenth-Century France Kathryn Banks
205
14 Bodies Politic and Bodies Cosmic: The Roman Stoic Theory of the ‘Two Cities’ Jeffery Zavadil
219
15 Metaphor in the History of Ideas and Discourses: How Can We Interpret a Medieval Version of the Body−State Analogy? Andreas Musolff
233
Commentary 16 Studying Metaphor in Discourse: Some Lessons, Challenges and New Data Raymond W. Gibbs, Jr. and Julia E. Lonergan
251
Index
263
Acknowledgements The motivation for this book first emerged from discussions at workshops on metaphor in discourse and conceptual history at various conferences, such as the ‘International Cognitive Linguistics Conference’, ‘Researching and Applying Metaphor’, ‘Language, Culture and Mind’, which were held in Britain, France, Spain and Poland between 2004 and 2007. We are grateful to the respective organizers for making possible the development of a continuous interdisciplinary discussion among cognitive linguists, critical discourse analysts, media researchers, conceptual historians and literary scholars. The present selection of chapters is intended to indicate the main avenues in this expanding field of research and to advance the ongoing theoretical and methodological debates. Of particular significance has been the support and guidance from Roslyn M. Frank, Zoltán Kövecses and Ray Gibbs, who not only agreed to contribute to this volume but also gave invaluable advice during its development. Jill Lake, of Palgrave Macmillan, encouraged the project to get off the ground and become a book. Abedal-mutaleb Al-Zuweiri , University of Durham, provided the index and helped with the final proof-reading.
vii
Notes on Contributors Kathryn Banks is Lecturer in French at Durham University. She graduated from the University of Cambridge, and was a Kennedy Scholar at Harvard and a Lecturer at King’s College London. Her research interests include generic and historical specificities of metaphor, topics addressed in her book, Cosmos and Image in the Renaissance (2008). John Barnden is the current Chair of SSAISB, the main academic Artificial Intelligence society in the UK. He spent many years in the USA, in computer science departments at Indiana University and New Mexico State University. In 1997 he returned to the UK to take up his present position. He has published widely on metaphor, and is a board member of Researching and Applying Metaphor, an international academic association. Jonathan Charteris-Black is Professor of Linguistics at the University of the West of England, and formerly a Senior Lecturer at the University of Surrey. He has published widely in the areas of figurative language, political rhetoric, and corpus linguistics. He is author of Corpus Approaches to Critical Metaphor Analysis (2004), Politicians and Rhetoric: the persuasive power of metaphor (2006) and The Communication of Leadership: the design of leadership style (2006). Paul Chilton is a Professor of Linguistics at the University of Lancaster and has held posts at Nottingham, Warwick, Stanford, Aston and the University of the West of England. His publications include work on metaphor in discourse, cognitive semantics, political discourse and Renaissance literature. His current research is on spatial representation and spatial modelling of discourse semantics. David Cowling is Professor of French and Head of the School of Modern Languages and Cultures at the University of Durham. He has published widely on late medieval and early modern French literature and culture, applying insights gained from cognitive metaphor theory to the analysis of allegorical and polemical texts. He is the author of Building the Text: Architecture as Metaphor in Late Medieval and Early Modern France (1998) and the editor of Les Douze Dames de Rhétorique. viii
Notes on Contributors ix
Christ’l De Landtsheer is a Professor of Communication Sciences at the University of Antwerp. Her research is on linguistic, psychological and technological aspects of political communication. Among her books are Metaphorical World Politics (2004, with Francis Beer), Beyond Public Speech and Symbols: Explorations in the Rhetoric of Politicians and the Media (2000, with Ofer Feldman), and Politically Speaking. A Worldwide Examination of Language Used in the Public Sphere (1998, with Ofer Feldman). Roslyn M. Frank is Professor Emeritus at the University of Iowa. She has published extensively in the field of cognitive linguistics, as well as in ethnoscience, with a research focus on Basque language and culture. She has co-edited Language and Ideology: Metaphor and Meaning (2001), Cognitive Models of Thought and Language, Vol. 2 (2003), and Body, Language and Thought (2008). Raymond W. Gibbs (Jnr) is Professor in the Department of Psychology, University of California, Santa-Cruz, USA. His research interests are in the fields of experimental psycholinguistics and cognitive science. He is especially interested in bodily experience and linguistic meaning. His publications include Embodiment and Cognitive Science (2005), Intentions in the Experience of Meaning (1999), Metaphor in Cognitive Linguistics (1999, co-edited with Gerard Steen), Figurative Language and Thought (co-author, 1998) and The Poetics of Mind: Figurative Thought, Language (1994). Honesto Herrera is Associate Professor of English for Business and Economics at Facultad de Ciencias Económicas y Empreariales, Universidad Complutense de Madrid. He has focused his research on how to take cognitive linguistics into the classroom. His main research interests are cognitive linguistics and research methodology. Currently, he is working on a corpus of linguistic realizations of conceptual metaphors in business and economics media discourse. He is author and coauthor of several articles and book chapters in these areas of research. Veronika Koller has a PhD in English Linguistics from Vienna University. She is Lecturer in English Language at Lancaster University, with her research focusing on critical discourse analysis and cognitive semantics (especially metaphor). Her current work addresses the conceptual structure and discursive representation of corporate brands. Her publications include Metaphor and Gender in Business Magazines (2004) and Communication in the Public Sphere (2008, co-edited with Ruth Wodak). Nelya Koteyko is a Research Fellow at the Institute for Science and Society at the University of Nottingham. Her research interests include media representations of health and illness and the relationship
x
Notes on Contributors
between corpus and discourse studies. She has recently published on the application of corpus linguistics to issues in discourse analysis and language description. Zoltán Kövecses is Professor of Linguistics in the Department of American Studies at Eötvös Loránd University, Budapest. He serves on the advisory board of several scholarly journals, including Cognitive Linguistics, and he is one of the associate editors of Metaphor and Symbol. His most recent books include Language, Mind, and Culture. A Practical Introduction (2006), Metaphor in Culture. Universality and Variation (2005), Metaphor. A Practical Introduction (2002), and Metaphor and Emotion (2000). He has taught and lectured widely at several American and European universities. He is currently working on the language and conceptualization of emotions, cross-cultural variation in metaphor, metaphor and metonymy in discourse, and the issue of the relationship between language, mind and culture from a cognitive linguistic perspective. Julia E. Lonergan is a graduate of the Department of Psychology, University of California – Santa Cruz. Her research focus is in metaphor, conceptual metaphor theory, cognitive linguistics, and language and culture. Andreas Musolff is Professor of German at the University of Durham and has published on German and English public discourse and metaphor theory. His books include Metaphor and Political Discourse (2007) and Mirror Images of Europe (2000); he has also co-edited volumes on Attitudes Toward Europe (2001). Brigitte Nerlich is Professor of Science, language and Society at the University of Nottingham and has published widely on the history of linguistics, historical and cognitive linguistics and the use of figurative language in public and scientific discourses. Gerard Steen teaches in the Department of Language and Communication at VU University Amsterdam. His most important work concerns metaphor in language, cognition, and communication, about which he has published special issues of Language and Literature, Journal of Pragmatics, and Style. His most recent book is Finding Metaphor in Grammar and Usage (2007). Michael White is Professor of English for Business and Economics at Escuela Universitaria de Estudios Empresariales, Universidad Complutense de Madrid. His main research interest is figurative language, especially its
Notes on Contributors xi
use in business and economics media discourse. He is author and coauthor of several articles and book chapters in this field. Jeffery Zavadil studied political theory at Penn State University and Arizona State University, where he taught from 2004 to 2008. He now lives and works in Washington DC, where he is completing his first book, a history of the metaphor of the body politic from classical antiquity to the European Middle Ages. Jörg Zinken is Senior Lecturer in the Department of Psychology at the University of Portsmouth (UK). He is co-editor of The Cognitive Linguistics Reader (2007). His research interests include the discursive grounding of metaphorical meaning, the use of metaphor in expressive writing, the use of metaphor in science and other real-world debates and the history of metaphors which frame debates about the significance of language diversity for understanding human psychology.
This page intentionally left blank
1 A Discourse-Centred Perspective on Metaphorical Meaning and Understanding Jörg Zinken and Andreas Musolff
Introduction In the current climate, it is taken for granted that metaphor is important and ubiquitous in language. Metaphor is no longer discussed as a ‘violation’ of normal verbal meaning (e.g., Levin, 1977), but rather as one form of normal verbal meaning. But of course if metaphor were all that ‘normal’, it would not stimulate the interest that it does. This interest is not only academic: it is not only philosophers, linguists, and psychologists who show a considerable interest in metaphor: ‘real’ people going about their everyday business of discussing events, possibilities, and problems, seem to find metaphors striking as well. As the chapters in this book illustrate, entire discourses circle around the negotiation of a metaphorical understanding. Metaphors in discourse summarize a possible stance, and such summary attracts further debate. Discourse studies make an empirical contribution to the study of metaphor: depending on the setting that is under investigation, such studies contribute to our understanding of the social realities constructed in the areas of politics, economics, science, law, doctor–patient conversation, and other areas of life. Such research can be conceptualized as an ‘applied’ endeavour that supplements (tests, provides a basis for) the aims of ‘basic’ (linguistic, philosophical) research to understand ‘metaphor understanding’. However, we argue in this chapter that a discourse perspective also calls into question assumptions made on the basis of theoretical commitments. We want to bring both of these possible contributions – of empirical analysis and distinctive theoretical perspective – together in this introduction. 1
2
Jörg Zinken and Andreas Musolff
Many philosophical accounts of how metaphor is understood stress the importance of context (e.g., Guttenplan, 2006; Leezenberg, 2001; Stern, 2000). Still, theorizing about metaphor does not often make use of empirical research in which metaphor is studied in context. One of the aims of this book, and this chapter, is to show that it should: approaching metaphors as actions that are embedded in larger discursive activities has important implications for our understanding of ‘metaphor understanding’.
1 Findings from discourse analytic studies of metaphor Let us start with an example. One of the persistent debates in the study of metaphorical communication concerns the question of how general the information is that is predicated of the topic. Relevance-theoretic approaches (Carston, 2002) answer that it is rather general. Say that the sentence my job is a jail were ever used outside the activity of writing a scientific text about metaphor. Let’s further assume that the speaker is not actually manager of a jail, but works in a fish and chip shop, the addressee knows this, and the speaker knows that the addressee knows this. What will the addressee make of the fact that the speaker said that his job was a jail? According to Carston, the addressee ‘works out’ the speaker’s communicative intentions by (a) constructing a new category ‘labelled’ jail, which includes not so much actual jails, but rather what is common to all unpleasant, confining situations, and (b) including the speaker’s job in this new, ad-hoc category. The ad-hoc category keeps what is common to jails and the speaker’s job, but loses anything that is true of jails, but not true of the speaker’s job. In other words, the word jail becomes merely a placeholder for a quite abstract conceptual category – unpleasant, confining situations – which the hearer supposedly works out when understanding the metaphor. Presumably, the speaker might just as well have said that his job was a stuck lift/elevator– another unpleasant, confining situation to be in. Another approach to metaphor which claims that the ideas involved in metaphor understanding are very general is ‘conceptual metaphor theory’ (Lakoff, 1993). On this view, the fact that polysemies can be sorted into thematic clusters indicates that these very polysemies are ‘licensed’ by associations holding between general ‘conceptual domains’ – i.e., by ‘conceptual metaphors’.1 According to this approach, thinking about a relationship as a car that is spinning its wheels, as a derailed train, or as a wrecked ship is the same thing – only the ‘general’ ideas of vehicles,
Metaphorical Meaning & Understanding
3
motion, and impediment to motion, somehow detached from the ‘particulars’ of a specific vehicle, matter. However, results of discourse analyses of metaphor do not support the assumption that the forms used in discourse are themselves irrelevant. For example, in one study of public discourse on political transformation, kettle-metaphors were used in the context of political pressures, whereas pot-metaphors were used in the context of subdivisions of political territory; ship-metaphors were used in the context of complex systems, whereas boat-metaphors were used in the context of collaboration, etc. (Zinken, 2007). Such results cast doubt on the validity of accounts which assume that metaphorical understanding involves only generic-level, abstract knowledge, representations abstracted from the particulars of situations. If ships are metaphorically meaningful in a different way than boats, then the particular ‘things’ that are employed in metaphorical communication (ships, jails, etc.) should surely play a role in accounts of how people make sense of metaphor? There are two ways to act upon such doubts thrown up by research on naturally occurring communication. One can embrace a distinction between ‘mere surface’ phenomena, such as the particular words people use in communication, and the ‘real business’ of ‘underlying’ structures, the ‘hidden’ realm of people’s representations and abilities that ‘allow’ them to do or refrain from doing certain things. This Platonic ‘solution’ is popular with theorists in the ‘conceptual metaphor’ approach, as it means that they do not need to take discourse data too seriously. It is, unsurprisingly, unpopular with discourse analysts, for the same reason. The second way of acting upon this doubt is to take discourse data seriously. Let us spell out some of the consequences that follow for our thinking about metaphorical understanding.
2
What follows from taking discourse data seriously?
The psycholinguistic literature on metaphor understanding has been focussed on how quickly participants can indicate an understanding of the metaphorical utterance. Metaphor understanding has occurred when the participant gives the relevant feedback, and that is the end of it (e.g., McGlone and Manfredi, 2001). From a discourse analysis perspective, metaphor understanding is not the end, but rather the beginning; entire fiercely fought debates centre on what follows from metaphor understanding: What would or should it mean for Britain to be ‘at the heart of Europe’ (Musolff, 2004)? Is language an ‘instinct’ (Pinker, 1994; Tomasello, 1995)? And when a particular interpretation has been
4
Jörg Zinken and Andreas Musolff
negotiated in a discourse, it generates further activities (Schoen, 1979): ethical debates about the implications of ‘deciphering the book of life’ act upon, rather than question, the assumption that establishing a person’s DNA-sequence tells you as much about that person as reading a book tells you about the contents of that book. Metaphors seem to play a vital role in keeping discussions going, at least in the discourse genres of the English-speaking world, such as science, education, and public discourse. The bottom line is: metaphor understanding in the real world might not be a matter of milliseconds passing until an individual study participant responds. It might rather be a matter of dialogue, of engagement in debate. At this point, it may be worth repeating that there is of course a commonly embraced response to the tension between psycholinguistic and discourse-analytic approaches to ‘metaphor understanding’ that we are pointing out here: Psycholinguists study what metaphor is, discourse analysts study how metaphors are used. The results from one field are of limited relevance to the work of the other. This is one way of cutting the pie. However, for the sake of argument, let’s take seriously the ‘usage-based’ credo that things are what they are because of their use. In this case, we need to discuss what we want ‘metaphor understanding’ to mean.
3
Perspectives on metaphor understanding
To the hearer, most metaphors seem ‘transparent’ (Guttenplan, 2006), and in a minimal sense, metaphor understanding seems to be rather effortless. Let’s assume we are engaged in a conversation about the meaning of life. As a seasoned metaphor scholar, you might be tempted to suggest that life is a journey, but on this occasion you want to be more original and choose a different evergreen as your metaphor vehicle. So you say: Life is a jail, or maybe Life is a wolf, or Life is the sun. All of these possible utterances immediately strike me as meaningful in the context of a conversation about the meaning of life. They are meaningful in the sense that I immediately appreciate them as contributions to our conversation, rather than as random noise or a string of irrelevant words, and in the sense that I feel I could say something in return. Based on my life-long practice of participating in communication, I appreciate their intelligibility. This ‘minimal’ experience of understanding might best be thought of as a sensation rather than as the result of ‘cold’ information-processing. This intuition is in line with proposals regarding the embodied nature of human understanding (Gibbs, 2005; Indurkhya, 1999).
Metaphorical Meaning & Understanding
5
In the psycholinguistic literature, however, metaphor understanding is usually envisaged as involving a ‘full’ interpretation: For example, consider that film was a sermon. For people who are not familiar with the film in question, there can be no a priori representation of the concept that film that includes properties such as preachy or moralistic. Yet these are exactly the sorts of properties that come to mind upon reading the statement, even when the film is not familiar to the reader. (McGlone and Manfredi, 2001: 1210) In this study by McGlone and Manfredi participants were required to press a key on the computer keyboard once they had achieved a ‘full understanding’ of each metaphor in the experiment (McGlone and Manfredi, 2001: 1212), which was assumed to look something like the interpretation given by the authors in the above quote. Experimental designs involving reaction times are commonly used in psycholinguistics to study how people arrive at such ‘full understandings’ of metaphor. Evidently, it is a different understanding of understanding that is implicit in such studies. Here, understanding is a telic project to be fulfilled entirely by the individual: it means identifying the right attributes (preachy, moralistic) and attributing them to the topic. Once this has been done, the goal of metaphor comprehension has been reached. But doesn’t thinking of a ‘preachy’ and ‘moralistic’ quality of sermons in the context of a film already require an understanding of the metaphor (Black, 1993 [1979])? Arguably, the process of coming to such a well articulated understanding involves embedding it in a meaningful narrative. Recent work shows that the ‘same’ metaphor can be understood quite differently depending on the narrative it becomes embedded in (Hellsten, 2000; Musolff, 2004) – that film might be a sermon because it is preachy and moralistic, but this film is a sermon because it is overlong and boring. For somebody else, the same film is a sermon because it grabs the viewer and gives her a renewed sense of meaning. Metaphors invite narratives – and it is the construction of the (metaphorical) narrative in a discourse community that gives the topic event meaning (Bruner, 1991). From this perspective, there is no ‘full’ understanding of a metaphor – rather, attempts at understanding are discontinued once these attempts generate more boredom than insight. The study of text and talk in context therefore throws up the question whether we actually ‘understand’ every word we hear in the strong, information-processing sense of ‘working out’2 all the entailments and implications. This view, in which understanding is an individual
6
Jörg Zinken and Andreas Musolff
project of theorizing about somebody else’s meaning, necessitated by a supposed ‘gap’ between my mind and yours, is strongly entrenched in Western academic thinking about meaning and understanding. But arguably, the effortlessness of most communication as experienced by participants is due not to the immense power of our mental ‘computers’, but because much of the time no such gap exists when we are engaged in communication (Leudar and Costall, 2004). If we were working so hard all of the time just to make sense of each other, surely a three-day weekend would be in order! In any case, the ‘felt’ understanding of a metaphor such as life is a jail is something very different from having an explicit list of predications that this metaphor can be taken to make about the meaning of life! In that sense, it would, as the information-processing perspective suggests, require some work for me to say what it could mean to say that life is a jail, or, indeed, a journey.3 But such an explicit interpretation might be not so much a precondition of understanding metaphors, but rather the result of debating what it would mean to view life as a jail, i.e. what would follow from adopting this perspective. An understanding in this ‘fuller’ sense might require the work of formulating and discussing (with others or oneself) possible interpretations.4 From the perspective of studying metaphorical text and talk in context, the hard work of ‘working out’ an interpretation seems to be distributed across participants in communication. Such communication can transcend immediate situational contexts and build up over time to discursive traditions. How long such diachronic continuity of metaphor can be assumed to last is still largely unexplored. The ‘strict father’ model of the family for the state in contemporary political discourse (Lakoff, 1996) has been related to centuries-old idioms, but should it be linked to Aristotle’s (384–322 BC) comparison of state and family in the Politics (Aristotle, 1995)? How can we ascertain which discursive traditions are still of relevance for the understanding of metaphors of such ‘long duration’? We have hinted at four ways to understand understanding: ‘embodied’ understanding as a sensation of meaningfulness, ‘dialogic’ understanding as ongoing participation in a conversation, ‘informationprocessing’ understanding by individuals working out attributes and implications, and ‘historically situated’ understanding that includes awareness of discourse traditions ‘revived’ in topical uses. Discourseanalytic approaches study metaphorical meaning and understanding as it is embedded in discursive activity. This perspective on metaphor can contribute to the theoretical inquiry into fundamental questions, such as our understanding of metaphor understanding.
Metaphorical Meaning & Understanding
7
Notes 1. Although the conclusions regarding the generality of ideas involved in metaphor understanding are reached on different grounds in the ‘conceptual metaphor’ and the ‘relevance theory’ approach, this shared conclusion has encouraged researchers to explore possible complementarities between the two approaches (Gibbs and Tendahl, 2006). 2. The idea that linguistic meaning requires ‘working out’ is prominent in the Gricean understanding of understanding ( see Clark, 1996, p. 141). 3. The variety of interpretations given by participants in Glucksberg’s (1999) study suggests that the meaning of a relatively conventional metaphor such as life is a journey is not at all obvious. 4. The point that explicit understanding is a dialogic process that depends on formulation for an ‘Other’ (even if that is oneself) should be intuitive to anyone who has started writing an essay with vague ideas and in the process ‘discovered’ much clearer ideas. The point has been developed systematically by W. v. Humboldt (Humboldt, 1963 [1830–35], vol. 3, pp. 428–9, 437).
References Aristotle. (1995). Politics. ed. by E. Barker and R. F. Stalley. (Oxford: Oxford University Press). Black, M. (1993 [1979]). ‘More about metaphor’. In A. Ortony (ed.), Metaphor and Thought (pp. 19–43) (Cambridge: Cambridge University Press). Bruner, J. (1991). ‘The narrative construction of reality’, Critical Inquiry, 18, 1: 1–21. Carston, R. (2002). ‘Metaphor, ad hoc concepts and word meaning–more questions than answers’. UCL Working Papers in Linguistics, 14. Clark, H. H. (1996). Using Language (Cambridge: Cambridge University Press). Gibbs, R. W. Jr. (2005). Embodiment and Cognitive Science (Cambridge: Cambridge University Press). Gibbs, R. W. Jr. and Tendahl, M. (2006).’Cognitive effort and effects in metaphor comprehension: relevance theory and psycholinguistics’, Mind & Language, 21, 3: 379–403. Glucksberg, S. (1999). ‘When love is not a journey: what metaphors mean’, Journal of Pragmatics, 31: 1541–58. Guttenplan, S. (2006). ‘The transparency of metaphor’, Mind & Language, 21, 3: 333–59. Hellsten, I. (2000). ‘Dolly: scientific breakthrough or Frankenstein’s monster? Journalistic and scientific metaphors of cloning’, Metaphor and Symbol, 15, 4: 213–221. Humboldt, W. v. (1963 [1830–35]). ‘Ueber die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts’. In Werke in fünf Bänden, ed. by A. Flitner and K. Giel (vol. 3, pp. 368–756) (Darmstadt: Wissenschaftliche Buchgesellschaft). Indurkhya, B. (1999). ‘Creativity of metaphor in perceptual symbol systems’, Behavioral and Brain Sciences, 22, 4: 621–2. Lakoff, G. (1993). ‘The contemporary theory of metaphor’. In A. Ortony (ed.), Metaphor and Thought (2nd edn) (Cambridge: Cambridge University Press).
8
Jörg Zinken and Andreas Musolff
Lakoff, G. (1996). Moral Politics: What Conservatives Know that Liberals Don’t (Chicago: Chicago University Press). Leezenberg, M. (2001). Contexts of Metaphor (Amsterdam: Elsevier). Leudar, I. and Costall, A. (2004). ‘On the persistence of the ‘Problem of Other Minds’ in psychology: Chomsky, Grice, and theory of mind’ Theory & Psychology, 14, 5: 601–21. Levin, S. (1977). The Semantics of Metaphor (Baltimore: Johns Hopkins University Press). McGlone, M. S. and Manfredi, D. (2001). ‘Topic–vehicle interaction in metaphor comprehension’, Memory & Cognition, 29, 8: 1209–19. Musolff, A. (2004). ‘Metaphor and conceptual evolution’, metaphorik.de, 7: 55–75. Pinker, S. (1994). The Language Instinct (New York: HarperCollins). Schoen, D. A. (1979). ‘Generative metaphors: a perspective on problem solving in social policy’. In A. Ortony (ed.), Metaphor and Thought (pp. 254–83) (Cambridge: Cambridge University Press). Stern, J. (2000). Metaphor in Context (Cambridge, MA: MIT Press). Tomasello, M. (1995). ‘Language is not an instinct’, Cognitive Development, 10: 131–56. Zinken, J. (2007). ‘Discourse metaphors: the link between figurative language and habitual analogies’, Cognitive Linguistics, 18, 3.
Part I Metaphor in Discourse: Theoretical and Methodological Perspectives
This page intentionally left blank
2 Metaphor, Culture, and Discourse: The Pressure of Coherence Zoltán Kövecses
Introduction The main question I wish to address in this chapter is the following: how can we account for metaphor variation in naturally occurring discourse using the cognitive linguistic theory of metaphor variation, as worked out in Kövecses (2005)? More specifically, what are the dimensions along which metaphors vary in natural discourse? Which aspects of metaphor are involved in this variation? And most importantly, what are the causes that produce such variation? In this chapter I propose a new notion in understanding the causes of metaphor variation in discourse: ‘the pressure of coherence’. The notion provides us with an explanation of metaphor variation in discourse that has not been available previously. As Conceptual Metaphor Theory emphasizes (e.g. Grady 1997; Lakoff and Johnson 1999; Kövecses, 2002), many conceptual metaphors are universal at a high level of abstraction. I have argued elsewhere that conceptual metaphors vary along certain dimensions in relation to several aspects, or components, of these metaphors due to a number of different causes (Kövecses, 2005). This variation is especially noticeable at a ‘culture-close’ specific level of generalization. Equally obviously, conceptual metaphors vary across cultures. Somewhat less obviously, conceptual metaphors also vary within cultures. First, I will take up the issue of how certain dimensions of metaphor variation participate in the selection of the most fitting metaphors in discourse. For the purposes of the present chapter, one of these will be examined: the style (sub)dimension. Second, I will ask how one aspect, or component, of conceptual metaphors is involved in metaphor variation in discourse. The particular component I have selected for analysis is 11
12
Zoltán Kövecses
‘cultural models’, which is one of the many aspects, or components, of conceptual metaphors. Third, I will address the issue of what the causes of metaphor variation are that contribute to the selection of particular metaphors over others in naturally occurring discourse. To handle this issue, I will propose a new theoretical notion: the principle of the pressure of coherence.
1 Dimensions of metaphor variation in naturally occurring discourse Conceptual metaphors vary along two major dimensions: intercultural (cross-cultural) and intracultural (within-culture) (see Kövecses, 2005). I conceive of culture here as a set of shared understandings. The production and comprehension of discourse assumes that speakers share such understandings of the world. The within-culture dimension of metaphor variation plays an important role in the study of metaphor variation in naturally occurring discourse, such as language used in the media. We can distinguish several subdimensions within the larger heading of within-culture dimension. Such subdimensions include the social, ethnic, regional, style, subcultural, diachronic, and individual dimensions (see Kövecses, 2005). Let us now see some examples for such variation. What I call the style subdimension here is especially relevant to the study of metaphor variation in the media. By ‘style’ I mean linguistic choice and variation according to the communicative setting, subject matter, medium, audience, etc. Let us take subject matter or topic as our first illustration of the point. Jean Aitchison (1987) made an interesting observation that bears on the issue. She noted that in newspaper articles and headlines about (American) football games, the names of the teams may select particular metaphors for defeat and victory. She found such examples as follows in American newspapers: ‘Cougars drown Beavers’, ‘Cowboys corral Buffaloes’, ‘Air Force torpedoes the Navy’, ‘Clemson cooks Rice’ (Aitchison, 1987: 143). Metaphors used in these sentences are selected on the basis of the names of football teams. Since beavers live in water, defeat can be metaphorically viewed as drowning; since cowboys corral cattle, the opponent can be corralled; since navy ships can be torpedoed, the opponent can be torpedoed, too; and since rice can be cooked, the same process can be used to describe the defeat of the opponent. The metaphors in the above sentences indicate that the target domain of defeat can be variously expressed as drowning, corralling, etc., the choice depending on the meaning of the sentences’
Metaphor, Culture & Discourse 13
constituents – namely, in the present case, what we could regard as the subject matter, or topic. Another subdimension along which metaphors vary in the discourse of the media is the one that has to do with the unique features of an individual. It is a fairly common observation that the metaphor usage of key cultural figures, such as presidents and media stars, as well as that of writers and poets, can differ markedly from one person to another and from the general population. A good illustration of this point comes from Time magazine that lists some of the metaphors that anchorman Dan Rather of CBS used in his 2001 election coverage (see Kövecses, 2002, 2005). Here are some examples of his metaphors from Time magazine (Time, November 20, 2001): (1) The presidential campaign is ‘... still hotter than a Laredo parking lot’. (2) Bush ‘has run through Dixie like a big wheel through a cotton field’. ‘... will be madder than a rained-on rooster ...’ ‘... is sweeping through the South like a tornado through a trailer park’. We can assume that, among the star journalists and anchormen, these metaphors are fairly specific to Dan Rather’s metaphorical repertoire. The images that are used seem to reflect Rather’s southern upbringing. I do not know whether other reporters who also come from the south use or would use the exact same metaphors to (jokingly) describe the election campaign. The fact that the humorous metaphors were noted and commented on by the magazine shows that there is something unique or peculiar about them and that they are likely to characterize a particular person.
2 How components of conceptual metaphor are involved in variation The various aspects, or components, of conceptual metaphors are involved differentially in metaphor variation. Kövecses (2005) identifies the following eleven components of conceptual metaphors: 1. Source domain; 2. Target domain;
14
Zoltán Kövecses
3. Experiential basis; 4. Neural structures corresponding to (1) and (2); 5. Relationships between the source and the target; 6. Metaphorical linguistic expressions; 7. Mappings; 8. Entailments; 9. Blends; 10. Nonlinguistic realizations; 11. Cultural models. Item (5), the relationship between the source and the target, may require clarification. A particular source domain may have several targets and a particular target domain may have several sources. I have called the former situation the scope of the source and the latter the range of the target (Kövecses, 2005). Of these components, the universal experiential basis and the corresponding neural structures produce universality in metaphorical conceptualization. By contrast, other elements, such as source domains, mappings, entailments, linguistic expressions, and so on, are susceptible to variation both cross-culturally and within a culture. Cultural models are interesting in this regard because one can argue that cultural models either result from conceptual metaphors or produce such metaphors. This is the issue of whether conceptual metaphors simply reflect cultural models (as proposed, for example, by Quinn, 1991) or constitute cultural models (as suggested by Lakoff and Kövecses, 1987; Johnson, 1987). What then is the relationship between cultural models and conceptual metaphors, and importantly, what does the study of media discourse tell us about the issue? To anticipate my conclusion, I’d like to suggest that both views can be maintained at the same time; in other words, conceptual metaphors can be said both to constitute and reflect cultural models. I arrived at this idea in response to some studies of media language by Musolff (2001, 2004). The more detailed answer to our question is as follows. A part of our conceptual system consists of abstract concepts that are metaphorically defined. The definition of abstract concepts by means of metaphor takes place automatically and unconsciously. This is the case when emotions are viewed as forceful entities inside us, when we think of abstract complex systems as growing (= developing), when we define our goals as ‘goals’ (to be reached), and when we believe that marriage is some kind of a union. We take these metaphorical ‘definitions’ as givens that are literal. There are many concepts like these that are defined or constituted by conceptual
Metaphor, Culture & Discourse 15
metaphors. And they are so constituted unconsciously and without any cognitive effort. I believe that this kind of definition of abstract concepts takes place at what I call the ‘supra-individual’ level of conceptualization (see Kövecses, 2002: chapter 17). It is the supra-individual level in the sense that it consists of a static and highly conventionalized system of mappings between physical source and abstract target domains. Because of the automatic and unconscious nature of the mappings, we tend to think of these abstract concepts as literal and believe, as Quinn does (1991), that the literal cultural models of the concepts ‘select’ the appropriate metaphors. The point is that although we take these models to be literal (and, consequently, that conceptual metaphors are based on them), they are in fact constituted by metaphor to begin with (and thus the cultural models are based on the conceptual metaphors instead). However, having said this, I also believe that Quinn makes a valid point. When we actually use these metaphorically constituted concepts in discourse, it is often the case that we choose metaphorical expressions that are not constitutive of our understanding of the target concept in question but that are based on an already existing metaphorical understanding of a model of a target domain. In other words, I would agree that the way discourse understanding and production works often creates situations in which metaphorical expressions arise from a prior understanding of the target as a (metaphorically constituted but literally taken) cultural model. As an illustration of this situation, let us consider Chilton and Lakoff’s (1995) work on the application of the BUILDING metaphor to the political domain; in particular, Gorbachev’s metaphor of the COMMON EUROPEAN HOUSE, that is, EUROPE (A POLITICAL STRUCTURE) IS A COMMON HOUSE . There exists the general metaphor ABSTRACT COMPLEX SYSTEMS ARE BUILDINGS (Kövecses, 2000a, 2000b, 2002). This metaphor has several mappings that can be given as submetaphors within the general metaphor; specifically: THE CREATION OF ABSTRACT STRUCTURE IS BUILDING ABSTRACT STRUCTURE IS PHYSICAL STRUCTURE (OF THE BUILDING) ABSTRACT LASTINGNESS IS THE STABILITY OF THE PHYSICAL STRUCTURE (TO STAND)
According to the standard Cognitive Linguistic view of metaphor, the source domain of BUILDING and the target domain of, in this case, POLITICAL STRUCTURE is characterized by these mappings (see, e.g., Kövecses, 1995, 2000b, 2002; Grady, 1997a, b). My claim, in line with the argument above, would be that the abstract target concept of POLITICAL STRUCTURE
16
Zoltán Kövecses
is constituted by these mappings. That is to say, the notion of political structure (as in the discussion of the unification of European countries into a single political entity) is in part defined by the metaphors below: POLITICAL STRUCTURES ARE BUILDINGS ABSTRACT COMPLEX SYSTEMS ARE BUILDINGS
And, indeed, we find numerous examples that are based on these mappings in the discourse on the integration of Europe in the 1990s, as analysed by Musolff (2001: 222): (3) ‘We want a Europe that’s not just an elevated free trade area, but the construction of a house of Europe as laid down in the Maastricht treaty’ (The Guardian, July 6, 1994) (4) ‘The common currency is the weight-bearing pillar of the European house.’ (The Guardian, June 3, 1997) The first example is based on the submetaphor THE CREATION OF ABSTRACT STRUCTURE IS BUILDING, while the second is based on both ABSTRACT STRUCTURE IS PHYSICAL STRUCTURE (OF THE BUILDING) (pillar) and ABSTRACT LASTINGNESS IS THE STABILITY OF THE PHYSICAL STRUCTURE (TO STAND) (weightbearing). These examples show that political structure is thought about in terms of the BUILDING metaphor, and, more importantly, that certain aspects of this abstract entity (and of many additional ones), such as construction, structure, and strength, are inevitably constituted by metaphor. (Notice the unavoidably metaphorical character of the words construction, structure, strength in relation to political structure.) But in the course of the debate about the unification of Europe at the time many expressions other than those that fit and are based on these submetaphors were used in the press. Musolff (2001) provides a huge number of metaphorical expressions that were not supposed to be used (according to the ‘standard’ cognitive view of metaphor). There was talk about the roof, the occupants, the apartments, and even caretakers and fire-escapes. If the BUILDING metaphor is limited to the previously mentioned predetermined aspects of the target domain, then speakers should not talk about any of these things in connection with political structure. But they do. Here are Musolff’s (2001) examples: (5) ‘We are delighted that Germany’s unification takes place under the European roof.’ (Documentation by the Federal press and information office, Bonn)
Metaphor, Culture & Discourse 17
(6) ‘At the moment, the German occupants of the first floor apartment in the “European house” seem to think that foreigners from outside the continent should be content with living in the rubbish bin.’ ( Die Zeit, January 10., 1992) (7) ‘What does he [Chancellor Kohl] need this house for, after so many years as Chancellor? – Well, it’s obvious, he wants to become the caretaker.’ (Die Zeit, May 16, 1997) (8) ‘[the European house is] a building without fire-escapes: no escape if it goes wrong’ (The Guardian, May 2, 1998) (9) ‘[it is a] burning building with no exits’ (The Times, May 20, 1998) Given these examples of metaphor usage, it seems that metaphors can do more than just automatically and unconsciously constitute certain aspects of target domains in a static conceptual system (i.e., at the supraindividual level). Once we have a source domain that conventionally constitutes a target, we can use any component of this source that fits elements of the target. Notice that there is a reversal here. In a dynamic discourse situation the activated target domain (such as political structure) in the discourse can indeed select components of the source (such as building) that fit a particular target idea or purpose. For example, if one has a negative view of the unification of Europe and has problems with, say, the difficulty of leaving the Union in case it does not work out for a particular country, then the speaker can talk about a ‘building without fireescapes’ – a part of the source that is obviously outside the conventionally used aspects of the source but that fits the target nevertheless. In other words, the examples above demonstrate that in real discourse target domains can, indeed, select the metaphors, albeit in a limited fashion. The selection of metaphors is limited because they come from a source that is already constitutive of the target. However, the rich target domain knowledge may select metaphors that are not conventionally used for the automatic and unconscious understanding of this target.
3 Causes of variation in metaphor Causes of metaphor variation can be factored into two general types: differential experience and the differential application of certain cognitive processes. Of the two, I will demonstrate the relevance of the former – differential experience – to the study of metaphors in naturally occurring discourse in the media.
18 Zoltán Kövecses
There are several distinct causes that produce differential experience and hence different metaphors, including (but not limited to) awareness of context, differential memory, differential concerns and interests, and their various subcases (Kövecses, 2005). We are attuned to the world in which we live most of the time. This means that we unconsciously monitor and pick out certain details of the world, including, of course, ourselves as part of it. This world consists, for us, of ourselves (our body), the physical environment, the physical and social aspects of the settings in which we act, and the broader cultural context. These are also aspects of the world that may have an influence on how the metaphors we use often vary along the dimension and in the form that we have isolated. Since all of these aspects of the world can vary in many ways, the metaphors we use can vary in many ways. Determining factors include the physical environment, social context, cultural context, and the communicative situation in which the use of metaphor takes place. To illustrate, let us take the last factor – the communicative situation. Awareness of the changing aspects of the communicative situation can influence the metaphors people produce. Metaphorical source domains for particular targets may arise from what I call the ‘pressure of coherence’. This means that speakers try (and tend) to be coherent with various aspects of the communicative situation in the process of creating metaphorical ideas. As an example, consider a headline from The Wall Street Journal Europe (January 6, 2003). It reads: (10) The Americanization of Japan’s car industry shifts into higher gear. What we have here is the PROGRESS IS MOTION FORWARD metaphor, a part of the Event Structure metaphor (see Lakoff, 1993). The specific target domain is the progress of the Japanese car industry. Since the topic is the car industry, it makes sense to use the motion of a car, and not the motion of some other entity capable of motion, in the metaphor. It seems to me that the pressure of coherence (with the topic at hand) is at work here. The phenomenon is not limited to headlines. Let us now take the following portion of a newspaper article from a Hungarian newspaper and its rough English translation: (11) Levelet írt Sepp Blatter a Nemzetközi Labdarúgó Szövetség (FIFA) svájci elnöke az ázsiai szövetség (AFC) vezeto˝inek, melyben elfogadhatatlannak mino˝ sítette a kontinens küldötteinek három héttel ezelo ˝ tti kivonulását a FIFA-kongresszusáról, ugyanakkor megígérte,
Metaphor, Culture & Discourse 19
hogy megpróbál segíteni az AFC gondjainak megoldásában – jelentette kedden a dpa német hírügynökség. Nagyon elkeserített az Önök viselkedése a Los Angeles-i kongresszusunkon. Önöknek, mint a labdarúgáshoz érto˝ szakembereknek tudniuk kellett volna, hogy az a csapat soha nem nyeri meg a mérko ˝zést, amelyik a lefújás elo˝tt levonul a pályáról – áll a levélben. (Zalai Hirlap [The Chronicle of Zala County], July 28, 1999) ‘Sepp Blatter, the Swiss president of the International Football Federation (FIFA), wrote a letter to the leaders of the Asian Football Association (AFC), in which he deemed unacceptable the behaviour of the association’s delegates three weeks ago when they left the FIFA Congress prematurely. On the other hand, he promised that he would try to help solve the problems with which AFC is struggling – the German news agency dpa reported. I was bitterly disappointed by your behaviour at our Congress held in Los Angeles. You, as experts on football, should have known that the team that leaves the field before the game is called off by the referee can never win the game – states the letter. ‘ [my translation, ZK] Here, the target domain is the politics of international football (or soccer in American English) that is conceptualized as the game of football itself. The delegates leaving the FIFA meeting ahead of time correspond to the football players who leave the playing field before the referee calls the game off. The pressure of coherence causes or enables the president of FIFA, Sepp Blatter, to choose the football game metaphor because the topic of the meeting is the international politics of football. The pressure of coherence seems to be pervasive in journalism in general. The Hungarian daily Magyar Nemzet (Hungarian Nation) carried an article some years ago about some of the political leaders of neighbouring countries who were at the time antagonistic to Hungary. One of them, the Slovak then-president, Meciar, used to be a boxer. This gave a Hungarian journalist a chance to use the following metaphor that is based on this particular property of the former Slovak president: (12) A pozsonyi exbokszolóra akkor viszünk be atlanti pontot éro˝ ütést, ha az ilyen helyzetekben megszokott nyugati módra “öklözünk”: megveto˝ távolságot tartva. A kolozsvári pankrátorral pedig közös ringbe sem szabad lépni. (Hungarian Nation, September 13, 1997) ‘We deal a blow worth an Atlantic point to the ex-boxer of Bratislava if we box in a Western style as customary in these
20
Zoltán Kövecses
circumstances: keeping an aloof distance. As far as the prize wrestler of Cluj-Napoca is concerned, we should not even enter the same ring with him.’ [my translation, ZK] What we have in all of these cases is that a certain aspect or property of either the source or the target domain forms the basis for building the respective source or the respective target domain. In the example of ‘Cowboys corral Buffaloes’, the generic-level meaning of corralling, that is, ‘controlling,’ is used to express the target domain of defeat, whereas in the last example of keeping an aloof distance (from Meciar) it is the property of one of the characters in the target domain that helps select the source domain for the target of confrontational international politics, that is, boxing. The physical setting as a potential cause of, or factor in, metaphor variation was studied by Boers (1999). He started out from the following general hypothesis. People will make more extensive use of a source domain when that particular source domain becomes more salient for them under certain circumstances. In other words, certain changes in the circumstances of the communicative situation may make people more aware of a particular source domain, and this may result in an increased use of the source domain in metaphorical conceptualization. The specific hypothesis was that the source domain of HEALTH will be especially productive of linguistic expressions in the winter because this is the time when, at least in countries of the northern hemisphere, people are more aware of their bodies through the more frequent occurrence of illnesses (such as colds, influenza, pneumonia, bronchitis). The particular target domain that was selected for the study was ECONOMY. Thus, according to the hypothesis, we can expect an increase in the relative salience of the ECONOMY IS HEALTH metaphor in the winter period. The salience of the HEALTH domain was assessed in terms of the frequency of health-related metaphorical expressions for economy. In order to test the hypothesis, Boers counted all the metaphorical expressions that have to do with economy and that are based on the HEALTH domain in the editorials of all issues of the English weekly magazine The Economist over a period of ten years. The study resulted in a sample of over one million words. Here is a selection of some of the metaphorical expressions that he identified: ‘healthy companies’, ‘sickly firms’, ‘economic remedy’, ‘symptoms of a corporate disease’, ‘a financial injection’, ‘arthritic markets’, ‘economic recovery’, and many others. The heavy presence of such and similar expressions shows that the economy is commonly talked and thought about in terms of bodily
Metaphor, Culture & Discourse 21
health. The question for the researcher was whether there was any fluctuation in the frequency of use of the HEALTH metaphor from season to season. Boers found that the frequency of the metaphor was highest between the months of December and March. The same result was found systematically for the ten years under investigation. During this period, the frequency of health-related metaphors for economy went up and stayed higher in the winter. This finding confirmed the hypothesis. When the HEALTH domain becomes more salient for people, they use health metaphors more extensively than when it is less salient. We can reinterpret Boers’ findings in the following way. Since the physical setting is part of the communicative situation, it may play a role in selecting particular metaphorical source domains. That is, in a given type of communicative situation people’s selection of metaphors may be influenced by the pressure of coherence principle. In the present example, wintertime is more likely to lead to the selection of health-related metaphors than to other metaphors, simply because such metaphors may be higher up in awareness than others. Further factors that play a role in producing differential experience include differential memory and differential concerns, or interests. As an example, let us look at how the latter influences our choice of metaphor in discourse. Intense professional interest may also lead a person habitually to think about and express target domains in terms of a source domain that is based on those professional interests. A good way of studying this form of variation is to look at letters in newspapers that are sent in to editors by readers. In Hungarian newspapers the authors of the letters often mention their profession. Consider the following letter by a Hungarian electrical engineer concerning the issue of Hungary’s new relationship with Europe in the late 1990s. (The quote is followed by my more or less literal translation of the original.) (13) Otthon vagyunk, otthon lehetünk Európában. Szent István óta bekapcsolódtunk ebbe a szellemi áramkörbe, és változó intenzitás”k ido ”nként, sal, de azóta benne vagyunk – akkor is, ha különféle ero ”le. (italics in hosszabb-rövidebb ideig, megpróbáltak kirángatni belo the original; Magyar Nemzet, [Hungarian Nation] June 12, 1999) ‘We are, we can be at home in Europe. Since Saint Stephen we have been integrated/ connected to this intellectual/ spiritual electric circuit, and with varying degrees of intensity, but we have been in it – even though various powers, for more or less time, have tried to yank us out of it’ [my translation, ZK]
22
Zoltán Kövecses
The target domain is Hungary’s new relationship to Europe in the wake of major political changes in the country in the 1990s. The interesting question is what the source domain is. As the passage makes it clear, many of the words used reflect the professional interest of the author of the letter: be integrated/ connected, electric circuit, with varying degrees of intensity are expressions that reveal electricity and electric circuitry as a source domain in the passage. The electrical engineer reasons on the basis of his knowledge of this domain. The concept of electricity and electric circuitry as a source domain is not obvious for this target and is certainly not the only one that could be used. My claim is that it is made available and its use is facilitated by the professional interest of the person who does the thinking about this particular target domain. Doctors, teachers, athletes, scientists, and so on, often take their source domains from their fields of activity to characterize and reason about the various target domains they encounter, talk, and think about. In sum, what we find in all of the cases discussed in this section is that when people use metaphors they tend to adjust them to various aspects of the communicative situation; they try to be coherent with such aspects of the context. In other words, people’s choice of metaphor seems to be influenced by what I have called the principle of ‘the pressure of coherence’. Future research will determine which aspects of the communicative situation can have this effect on the choice of metaphors.
4
Conclusions
Several conclusions emerge from the study of metaphor variation in naturally occurring discourse, such as media language, as presented in this chapter. First, media language is a useful area to look at in studying metaphor variation. This is because, by its very nature, media language comes with clearly identifiable parameters that can be taken into account in explaining why particular metaphors are used. This is a definitive advantage over the study of metaphor through introspective examples. Second, the study of metaphor variation in media language shows promise in solving (or at least getting clearer about) one of the important issues in the cognitive study of metaphor. This is the issue of the primacy of metaphors or cultural models in their interaction: do conceptual metaphors give rise to cultural models, or do cultural models give rise to conceptual metaphors? As the examples discussed in this chapter show, our cultural models can be both created by metaphors and at the same time can determine (or select) the metaphors we use in
Metaphor, Culture & Discourse 23
discourse. This result offers a more balanced view than either the one proposed by Johnson (1987) or Lakoff and Kövecses (1987) or Kövecses (1999) or the one proposed by Quinn (1991). Third, as several of the examples studied above indicate, an especially important principle of metaphor variation in media language emerges – that of ‘the pressure of coherence’. This principle, I suggest, makes the user of language adjust his or her metaphors to the surrounding context. It is likely that the principle can explain a large amount of metaphor variation in naturally occurring discourse on the basis of the interplay between universal embodiment, differential experience, and the changing context of communication. In this view, even universal embodiment can be seen as a special case of the pressure of coherence. That is to say, if there are no overriding factors, people can use certain universal metaphors for particular targets. However, in the majority of cases of metaphor use there seem to be overriding factors that lead groups of people and individuals to employ non-universal metaphors. This study, as well as Kövecses (2005), is merely a first step in identifying what these factors are and under what conditions they can override metaphors based on universal embodiment. As a matter of fact, it may turn out that reversing this reasoning would ultimately be a better solution. Instead of asking when we use non-universal metaphors instead of the universal ones, we might more profitably ask when we use universal metaphors instead of the many available locally-governed, non-universal ones. At this stage, it seems to me that we need more empirical evidence in order be able to confidently decide in which direction the principle of the pressure of coherence works.
References Aitchison, J. (1987). Words in the Mind. (Oxford: Blackwell). Boers, F. (1999). ‘When a bodily source domain becomes prominent’. In R. Gibbs and G. Steen (eds.), Metaphor in Cognitive Linguistics (pp. 47–56) (Amsterdam: John Benjamins). Chilton, P and Lakoff, G. (1995). ‘Foreign policy by metaphor’. In C. Schäffner and A. L. Wenden (eds.), Language and Peace. (pp. 37–59) (Brookfield, VT: Dartmouth Publishing Company). Johnson, M. (1987). The Body in the Mind (Cambridge: Cambridge University Press). Kövecses, Z. (2000a). Metaphor and Emotion. (Cambridge: Cambridge University Press). Kövecses, Z. (2000b). ‘The scope of metaphor’. In A. Barcelona (ed.), Metaphor and Metonymy at the Crossroads (pp. 79–92) (Berlin: Mouton de Gruyter).
24 Zoltán Kövecses Kövecses, Z. (2002). Metaphor. A Practical Introduction (Oxford: Oxford University Press). Kövecses, Z. (2005). Metaphor in Culture (Cambridge: Cambridge University Press). Lakoff, G. (1993). ‘The contemporary theory of metaphor’. In A. Ortony (ed.), Metaphor and Thought. (Cambridge: Cambridge University Press). Lakoff, G and Johnson, M. (1999). Philosophy in the Flesh. The Embodied Mind and its Challenge to Western Thought (New York: Basic Books). Lakoff, G. and Kövecses, Z. (1987). ‘The cognitive model of anger inherent in American English’. In D. Holland and N. Quinn (eds.), Cultural Models in Language and Thought. (Cambridge: Cambridge University Press). Musolff, A. (2001). ‘Political imagery of Europe: a house without exit doors?’ Journal of Multilingual and Multicultural Development, 21, 3: 216–29. Musolff, A. (2004). Metaphor and Political Discourse (Basingstoke: Palgrave Macmillan). Quinn, N. (1991). ‘The cultural basis of metaphor’. In J. Fernandez (ed.), Beyond Metaphor. The Theory of Tropes in Anthropology (pp. 56–93) (Stanford: Stanford University Press).
3 Three Kinds of Metaphor in Discourse: A Linguistic Taxonomy Gerard Steen
Introduction The study of metaphor has been revolutionized by the Cognitive Linguistic approach advocated by Lakoff and Johnson (1980, 1999; for a good introduction, see Kövecses, 2002). They have shown how the structure of language is replete with metaphor, for instance in conventional expressions for talk about argument in terms of war (Lotte defended her thesis; Sam attacked George with mild irony; Bush has proposed another untenable position) or (un)happiness in terms of a vertical scale (I feel in high spirits today; don’t let it bring you down). But they have also widened the scope of the analysis of metaphor beyond its linguistic form and meaning. The Cognitive Linguistic claim is that these conventional linguistic metaphors are reflections of underlying conventional mappings between distinct conceptual domains, or metaphor ‘in thought’, to the effect that we think of ARGUMENT AS WAR or HAPPINESS AS UP. Such conceptual cross-domain mappings are held to be part of people’s enduring knowledge structures and to be a constitutive factor of all kinds of cognitive processes, including language use, reasoning, and the exercise of the imagination in literature, the arts, and other domains of human performance. The empirical evidence for the Cognitive Linguistic (CL) view of metaphor as a conceptual phenomenon is mainly linguistic, and has been criticized as such for what it can tell us about the role in knowledge and cognitive processing by cognitive psychologists like Greg Murphy (1996, 1997), and in language processing by discourse analysts such as Cameron and Low (1999; for a comprehensive review, see Steen, 2007). Still, the CL 25
26 Gerard Steen
view of metaphor has kept inspiring much innovative research and debate in a range of disciplines. By comparison, much less impact outside its own tradition has been enjoyed by the Systemic Functional approach to metaphor in language, which is characterized by the notion of ‘grammatical metaphor’. Michael Halliday first proposed this idea in An Introduction to Functional Grammar (Halliday, 1985), where he suggested that a grammatical construction like The fifth day saw them at the summit is a metaphorical alternative to the less metaphorical They arrived at the summit on the fifth day: The former expresses as a mental process of perception what can be less metaphorically expressed as a material process. In a more recent work, co-authored with Christian Matthiessen, Halliday aims to broaden the purport of linguistic research beyond language as such, and in a long and fundamental chapter on grammatical metaphor seeks to establish explicit connections with the CL approach (Halliday and Matthiessen, 1999). The notion of grammatical metaphor is now presented as complementary to the phenomena typically discussed in Cognitive Linguistics, which are classified by Halliday and Matthiessen as lexical metaphor (1999: 232–3). To illustrate, The fifth day saw them at the summit is grammatically metaphorical because it uses the grammatical roles of perception to express a process that has nothing to do with perception but is a material process; but it also contains the metaphorically used lexical item see. This publication has in turn led to another, edited, volume in the Systemic-Functional Grammar (SFG) school entirely devoted to grammatical metaphor, with a number of chapters contemplating the connections between the CL and SFG schools in more detail (Simon-Vandenbergen, et al., 2003). These SFG authors have accepted that the CL notion of metaphor as a conceptual crossdomain mapping may provide a unifying platform for the description of both grammatical as well as what they call ‘lexical’ metaphor (again, see Halliday and Matthiessen, 1999: 233). What is not given much attention in either the CL or the SFG approach is how to deal with other forms of metaphor, such as simile and analogy. These are diverging linguistic manifestations of cross-domain mappings in thought and constitute an interesting class of phenomena since they do not appear to require metaphorical language use in the traditional sense of the term (cf. Goatly, 1997). An example would be Shakespeare’s extended comparison in sonnet 18, beginning with the famous line Shall I compare thee to a summer’s day? (Steen and Gibbs, 2004). This type of explicit metaphorical comparison involves a mapping between two conceptual domains that are each presented in their own right,
Three Kinds of Metaphor in Discourse
27
without exhibiting metaphorical language use that is comparable to either grammatical metaphor or lexical metaphor. Psychologists have examined the difference between regular metaphor on the one hand and simile, analogy, and the like on the other, but this has not had much influence on the discussion in linguistics of metaphor in language and thought (e.g. Bowdle and Gentner, 2005; Glucksberg, 2001; Chiappe and Kennedy, 2001). It is my aim in this chapter to examine and develop the relations between these proposals about metaphor in language from a discourseanalytical perspective. If we assume that metaphor in discourse is the linguistic manifestation of a conceptual cross-domain mapping, as many metaphor researchers do, the question arises how the three types of metaphor distinguished above can be understood within that theoretical framework. I believe that the study of discourse, defined as concrete events of language use, is able to refine our view of the nature and function of the various phenomena involved. I will therefore offer a theoretical consideration of the position of metaphor in discourse in section 1, and then move on to the ways metaphor can be operationalized for the study of discourse in section 2. Section 3 will then analyse the three types of metaphor against this light, and offer a new linguistic taxonomy for metaphor in discourse that builds on from, and integrates, the developments evoked above.
1 Conceptualizing metaphor in discourse The problem with terms like ‘language’, ‘thought’ , and ‘discourse’ is that they are used in subtly different ways in several schools of research. That is why I first need to explain that the perspective I adopt on metaphor in discourse is part of the one advocated by cognitive psychologists such as Van Dijk and Kintsch (1983) and their colleagues (e.g. Van Oostendorp and Goldman, 1999; Kintsch, 1998; Weaver, Mannes, and Fletcher, 1995; Perfetti, 1999; cf. Steen, 1994; 2004, 2007). The strength of this approach lies in its empirical concern with the minute details of the short-term processes of discourse production and comprehension that can be observed in concrete language users, as well as the resulting cognitive representations and subsequent processing in terms of, for instance, metaphor recognition, interpretation, and appreciation (Gibbs, 1994, 1999a). One basic characteristic of the cognitive-psychological approach is that it assumes that people need several distinct mental representations of discourse during processing. A productive distinction has been made
28
Gerard Steen
between four types of cognitive representation that people build during production and comprehension: 1. The surface text, which involves the linguistic representation of the message exchanged in communication; 2. The text base, which involves a conceptual representation of the message in the form of a linearly and hierarchically ordered series of propositions; 3. The situation model, which is another conceptual representation of the message, but this time in the form of a non-propositional format which includes all kinds of information; 4. The context model, which is a conceptual representation of the discourse situation and its participants, medium, message-as-tool, and so on. Empirical research has shown that language users need to construct at least parts of these distinct types of mental representations during language use processes to be successful in their communicative behaviour (e.g. Kintsch, 1998; Perfetti, 1999). I will take this behavioural division of the field as my theoretical frame of reference for situating a number of phenomena of metaphor in discourse, including lexical metaphor, grammatical metaphor, and other forms of metaphor such as simile and analogy. What is important from this perspective is the following: 1. When language users process parts of the surface text to build a text base, they activate concepts related to words, but they also activate concepts for information that has to be recovered from grammatical constructions like substitution and ellipsis (e.g. Bovair and Kieras, 1985). Thus, there should not be a difference between the propositional representations in the text base of the following two halves of the first line of a poem by Lord Alfred Tennyson: Now sleeps the crimson petal, now the white, even though it is clear that the mental representations in the surface text of their corresponding verbal expressions should be radically different (Steen, 2002). In the text base, both discourse units should exhibit a propositional structure that turns on the metaphorical use of the concept sleep with the non-metaphorically used argument petal (Steen, 2002). The text base hence captures the propositional content of a discourse, whereas the surface text deals with the expressive structure of language (for an overview and references, see Perfetti, 1999).
Three Kinds of Metaphor in Discourse
29
2. When language users process surface text and the text base to build a situation model, they focus on the referential function of the concepts in their mental model of the world evoked by the text. Thus, given that the line from Tennyson is about a garden scene, the concept petal directly refers to an entity
in that garden, whereas the concept sleep indirectly refers to some action performed by that petal. It is this indirectly evoked action which needs to be inferred by the reader of the poem to complete the semantic structure of the utterance, and this part of the mental model of the message. This is presumably done by setting up a cross-domain mapping from the source domain sense of sleep to the intended but implicit target domain sense. How this is done during online processing has been the matter of some debate in psycholinguistics (e.g. Bowdle and Gentner, 2005). 3. When language users process language to build a context model, they may be said to focus on the rhetorical function of the discourse: the topic of the message and what the author may want to say about that topic or do with that topic (e.g. Mann and Thompson, 1988). At utterance level, this means that some information is treated as given and other information as new, and that all of the information is assumed to have pragmatic function in terms of, for instance, speech acts and conversational maxims. The rhetorical function of a line such as Now sleeps the crimson petal, now the white, would have to be based on a functional interpretation of the sequential ordering of the two events by the repeated initial component now, in combination with the shifting focus from the crimson petal to the white petal. This may be theorized to take place in the context model (e.g. Van Dijk, 1997, 2006). In earlier publications, I have shown how language users can display distinct reactions to the same metaphors in discourse depending on whether they concentrate on their role in the surface text, situation model, and context model (Steen, 2006), and how elicited metaphor recognition by language users can be affected by metaphor properties that can be interpreted as part of the surface text, text base, and context model (Steen, 2004). Even though there is much talk in both CL and SFG about the processes of language use, it is doubtful that there is much substantial empirical research in linguistics on metaphor as a cross-domain mapping which pertains to the processes that lead to the construction of a surface text, text base, situation model, or context model in individual language users’ minds. Psycholinguist Ray Gibbs (1999b) has proposed that CL research on metaphor generally allows for four alternative
30 Gerard Steen
hypotheses about the role of cross-domain mappings, or metaphorical thought, in language use: 1. Metaphoric thought might function automatically and interactively in people’s on-line use and understanding of linguistic meaning. 2. Metaphoric thought might motivate individual speakers’ use and understanding of why various words and expressions mean what they do, but does not play any role in people’s ordinary on-line production or comprehension of everyday language. 3. Metaphoric thought might motivate the linguistic meanings that have currency within linguistic communities, or may have some role in an idealised speaker/hearer’s understanding of language. But metaphoric thought does not play any part in individual speaker’s ability to make sense of, or process, language. 4. Metaphoric thought might play some role in changing the meanings of words and expressions over time, but does not motivate contemporary speakers’ use and understanding of language. (Gibbs, 1999b: 42–3) Gibbs has adopted a similar position towards other proposals in CL (Gibbs, 2000, 2005, 2006). I believe that the same alternative hypotheses can be proposed for the notion of grammatical metaphor in SFG, and that the psycholinguistic research into simile and analogy has not been adequately connected to this debate. In what follows I will develop my views about this configuration of problems while assuming the opposition between a behaviour-oriented perspective as pursued by Gibbs and other psychologists on the one hand and a linguistic or more broadly semiotic perspective as pursued by most linguists and discourse analysts on the other hand (Steen, 2007). This means that we will examine the meaning of metaphor in discourse without making any systematic claims about how people process metaphor in discourse. The analysis of metaphor in discourse may ultimately be placed in a wider context of the psychological study of cognition, but it first has to succeed in clarifying the differences between various uses of metaphor in discourse from a systematically linguistic or semiotic vantage point. That is the aim of the following sections.
2
Operationalizing metaphor in discourse
Even though the theoretical definition of metaphor in CL and in SFG are not necessarily completely identical, I will take seriously the attempt
Three Kinds of Metaphor in Discourse
31
by Halliday and Matthiessen (1999), as well as their followers in SimonVandenbergen et al. (2003), to build a bridge from SFG to CL when it comes to metaphor research. Therefore I will set out from the CL conceptualization of metaphor as a cross-domain mapping and now turn to the criteria for operationalizing metaphor in discourse. I will do so by taking a semiotic perspective on discourse instead of a behavioural one in which cognitive processes and representations are at the centre of attention, for the reasons given above. The question is how the three phenomena distinguished in this chapter, lexical metaphor, grammatical metaphor, and other forms of metaphor, relate to this matter of operationalization. Since the other forms of metaphor have not received enough attention yet to be able to designate them by a more positive description, I will, for this moment, abide by the slightly odd terminology introduced here (cf. Steen, 2007). The operational criteria for metaphor in discourse need to be targeted to a specific level of reality. In CL, the distinction between metaphor in language versus metaphor in thought has been fundamental from the beginning. In this chapter I will concentrate on the area of linguistic form. In addition, although Cognitive Linguists do not always make a sharp distinction between the study of metaphor in grammar versus usage, I will also limit myself to the study of usage. This area is characterized by its attention to any linguistic form that can express a cross-domain mapping, and any cross-domain mapping in conceptual structure that can be judged to be metaphorical. This has to be opposed to the conventional metaphors in linguistic form and conceptual structure that are part of the study of grammar (Steen, 2007). Given these preliminaries, let us now examine more closely the operationalization of linguistic forms of metaphor. For this phenomenon, the publications by Lakoff (1986, 1993) are essential, and have been generally accepted by many Cognitive Linguists. I have argued that they amount to the claim that metaphorical meaning is indirect meaning which may be understood by means of some form of similarity or comparison (Steen, 2007). I have also argued that understood, in that definition, needs to be taken in its loose sense of ‘interpreted by the analyst’, not in its cognitive-psychological sense of ‘processed by every language user’). Thus, the word sleep in the line from Tennyson is used indirectly in that it is not used in its basic sense pertaining to humans or animals, but requires some sort of mapping from its basic sense to some intended contextual sense that is adequate for flowers and plants. Although the theoretical situation is complex, recent publications in Cognitive Linguistics such as Kövecses (2002) and Dirven and Pörings
32 Gerard Steen
(2003) may be invoked to support this view of the CL definition of linguistic metaphor as based on indirectness-by-similarity. The more recent attention to metonymy, in particular, has increased the popularity of the old structuralist opposition between indirectness-bysimilarity for metaphor and indirectness-by-contiguity for metonymy (Barcelona, 2000; Panther and Radden, 1999; Panther and Thornburg, 2003; cf. Steen, 2005). Some analysts of discourse who are inspired by CL have adopted this operationalization of metaphor in discourse by indirectness-bysimilarity (e.g. Chilton, 1996; Koller, 2004; Pragglejaz Group, 2007). However, other discourse analysts have preferred to work with another set of criteria for metaphor in discourse, even though they have also accepted some or many of the tenets of the CL view. Thus, Cameron (2003), Charteris-Black (2004) and Caballero Rodríguez (2006) have all published studies in which metaphor in discourse is operationalized by some form of incongruity between the term(s) from the source domain and the target domain, which is then resolved by some form of semantic transfer that accommodates the source domain term(s) within the target domain (see also the chapters by Chilton, Koller, and Charteris-Black in this volume). Again, the word sleep causes incongruity in a context that is about flowers, which cannot sleep; this incongruity has to be resolved by transfer of those semantic features from sleep which are relevant to the presumed action of the petal. The CL terms ‘indirectness’ and ‘similarity’ are not prominently used by these discourse analysts. To me the two sets of criteria, of indirectness plus similarity and of incongruity plus semantic transfer, are in principle compatible. They seem to reflect two distinct but related perspectives on metaphor in discourse (cf. Steen, 2007): 1. a paradigmatic, sign-oriented perspective: words like attack and defend can be used directly and indirectly, depending on the context of war (directly) or argumentation (indirectly); when they are used indirectly, mapping based on some form of similarity will have to motivate their intended meaning; 2. a syntagmatic, utterance-oriented perspective: words like attack and defend can cause congruity or incongruity in the utterance, depending on the utterance context of war (congruous) or argumentation (incongruous); when they cause incongruity, meaning transfer from the incongruous sense to the intended, congruous meaning will have to clarify their contextual meaning.
Three Kinds of Metaphor in Discourse
33
But even though the two approaches may be compatible, it may still be true that the former is slightly more suitable for the study of grammar, which focuses on conventionalized signs, whereas the latter may be more suitable for the study of discourse, which focuses on any utterance in usage. If proponents of either approach clarify their position to the other approach, it will be clear what claims they are making about metaphor from a more general perspective.
3
Three types of metaphor in discourse
If we adopt these operational definitions of metaphor as our starting point, the question arises as to how they can cater for the phenomena of grammatical metaphor and other forms of metaphor. Here are two examples from concrete usage events discussed by Koller in her study of business discourse: Like football and trench warfare, this is a contest of sweat, mud, and inches. (Koller,2004: 101, from Fortune) [Mergers] are, like second marriages, a triumph of hope over experience. (Koller, 2004: 155, from The Economist) Both examples involve direct, not indirect, presentations of the respective source domains, football and trench warfare as well as second marriages. The reader presumably needs to set up a situation model in which the referent of the business contest is put side by side with the referents of football and trench warfare in the first example, and in which the referent of mergers is put side by side with the referent of second marriages in the second example. The criterion of indirect meaning therefore seems to fail at this level of analysis. The criterion of similarity, however, is clearly appropriate, since both utterances contain the signal like which suggest that some similarity has to be constructed. When we look at these examples in terms of incongruity and semantic transfer, it can be seen that there is some form of incongruity between the source domains and target domains, but that it is a form of incongruity which has been acknowledged by the writer, who utilizes a construction of juxtaposition between the two domains. The need for semantic transfer is signalled by the same token, the fact that we have the lexical unit like. The incongruity criterion appears to be applicable
34
Gerard Steen
to simile, which suggests that this is a notion that can capture both direct and indirect expressions of metaphor alike. It can accommodate metaphor which is expressed by metaphorical language use, as in the sleep example from Tennyson, as well as metaphor which is expressed by other linguistic means, including all forms of overt comparison. What may be concealed by the application of the incongruity criterion, however, is the fact that regular metaphor and simile are forms of indirectness as well as of incongruity which pertain to two different mental models of the discourse. If we take indirectness as referential indirectness, which is the way in which it has been typically used in CL, then it pertains to metaphorical mappings which need to be constructed and resolved in the situation model, or the mental picture of the world evoked by the text. That is where the tension (or incongruity) between petal and sleep, for instance, has to be transcended in order to achieve a complete mental representation. By contrast, simile involves a different kind of indirectness: when journalists write about mergers by comparing them to second marriages, they speak about one thing in terms of another in a different way than when they use metaphorical language; they actually signal to their addressee that they are speaking of one topic by looking at it from the perspective of another topic. This can be seen as an aspect of rhetoric, where a message does not hide its dependence on an alien conceptual domain, but in fact presents it as a distinct referential component of its communicative strategy. In simile, the situation model hence contains two referential domains that are placed side by side in order to be compared, all of which is meant to form one coherent referential whole, guaranteed by signals such as like. The context model has to privilege one of the two domains presented in the situation model as the target domain, which is the overall or local topic of the text; that target domain is represented as talked about in terms of an alien, subservient source domain which requires metaphorical connecting. In other forms of metaphor, then, the metaphorical tension has to be constructed and resolved in the context model, not the situation model. Regular lexical metaphor, therefore, constitutes a form of referential indirectness or incongruity, whereas simile and other forms of overt metaphorical comparison constitute forms of topical or rhetorical indirectness and incongruity. Referential indirectness and incongruity pertain to metaphorical language use in the sense that terms are used in other senses than their basic sense; topical or rhetorical indirectness and incongruity pertain to language use that is not metaphorical in the same way, but still requires the construction of cross-domain mappings
Three Kinds of Metaphor in Discourse
35
at another level (the context model) to achieve their metaphorical function. What both phenomena share is that a cross-domain mapping is inevitable. It has to be constructed by the analyst in order to make the discourse coherent, either at a referential level (indirect metaphor) or at a rhetorical level (other forms of metaphor). If such a cross-domain mapping is not assumed, the word sleep and concept ‘sleep’ remain unconnected to the rest of the discourse in the situation model, and the words second marriage and concept ‘second marriage’ remain unconnected to the rest of the discourse in the context model. These are mandatory semantic operations if the discourse as a whole is to remain intact. This is precisely what differentiates regular metaphor and other forms of metaphor from grammatical metaphor. Returning to Koller’s examples above, there are several expressions which might qualify as grammatical metaphor, of which I will take contest as my test case here. Before we get to grammatical metaphor, though, it should be pointed out that contest may also be seen as metaphorical from a referential perspective, that is, it also constitutes a case of lexical metaphor. This is because we are not talking about ‘a competition, especially in which people’s skill in a particular activity or sport is tested’, but about ‘a situation in which two or more people or groups are competing to gain power or an advantage’ (Macmillan English Dictionary for Advanced Users). A metaphorical mapping from the sports sense to the more abstract sense has to be assumed for the situation model to incorporate the indirectly or incongruously used term contest into the discourse. But contest is also a grammatical metaphor. A process has been portrayed as a thing. It is the grammatically metaphorical wording of an idea that might also have been expressed more congruently in a clause with two companies contending for an advantageous outcome in an economic struggle. In SFG terms, there is a mapping from the more congruent clause to the grammatically more metaphorical nominal group which crosses two skeletal or abstract conceptual domains: process is thing. The metaphorical mapping between process and thing, however, is not between two senses or meanings that are operative in the reconstruction of any of the mental models needed for the conceptual representation of the discourse. The term contest does not mean ‘thing’ at any of the levels of discourse which we have considered, not even in the surface text. It means either ‘a competition, especially in which people’s skill in a particular activity or sport is tested’ or ‘a situation in which two or more people or groups are competing to gain power
36 Gerard Steen
or an advantage’, and that is how it is presumably used as a linguistic sign for the activation of the corresponding concept (text base), referent (situation model), and topic (context model). The noun contest does not exhibit an indirect ‘thing’ use, for the same word cannot be used in another ‘process’ meaning that is more direct. This should be contrasted with the way in which it is a lexical metaphor: its indirect ‘power / advantage’ sense has to be compared with its more direct ‘competition’ sense. Nor does the term contest display any incongruity between the presumably more direct (or congruent) ‘process’ sense and the more indirect (or metaphorical) ‘thing’ sense which ought to be exhibited by the rest of the utterance. This should also be contrasted with the way in which contest is a lexical metaphor: there is an incongruity between its more direct ‘competition’ sense and the more indirect ‘power/advantage’ sense which is predominant in the rest of the utterance. If grammatical metaphor is metaphorical, it is so in a radically different way than the other two types of metaphor. I think this is because it is metaphorical in comparison with another (more congruent) possibility of expression that might be afforded by the language system. It is metaphorical when it is seen as part of the realization possibilities of the grammar as a grammar. But it is not necessarily or functionally metaphorical in relation to the dominant mode of expression (and its referential and topical/rhetorical functions) in the unique discourse in which it is located. The latter is what characterizes regular metaphor and other forms of metaphor. They are metaphors that simply have to be resolved in usage. Grammatical metaphor, by contrast, is a type of metaphor that can be reconstructed by the linguist as a grammatical reflection on usage.
4
Conclusion
In this chapter I have presented a first attempt at a linguistic taxonomy for three manifestations of metaphor that have received attention in diverging traditions. All traditions have discussed these three types of metaphor as part of discourse, or usage events. They have looked at the symbolic structure of these types of metaphor, and analysed their linguistic forms in relation to assumed underlying cross-domain mappings. This may have given the impression that the three types are functionally equivalent. I have shown here that there are important differences between the three types. Regular, ‘lexical’, metaphor is characterized by indirect language use which needs to be incorporated into the overall discourse
Three Kinds of Metaphor in Discourse
37
by a cross-domain mapping at the level of projected referents in the situation model. Other forms of metaphor such as simile are typically characterized by direct language use, which sets up an opposition between two conceptual domains within the referential situation model which needs to be resolved in terms of the rhetorical purposes of the sender in the context model. Both of these types of metaphor require the construction and resolution of a cross-domain mapping in the discourse in order to preserve its overall coherence. By contrast, grammatical metaphor is metaphorical by some form of contrast with possible alternative expressions that are also afforded by the language system. The contrast between the indirect or incongruous metaphorical meaning on the one hand and the more direct, congruous meaning on the other does not reside in the discourse itself. It is located in the analytical contrast between modes of expression that are facilitated by a grammar of choice. The metaphorical – that is, contrastive and comparative, two-domain – nature of grammatical metaphor does not appear to play a mandatory role in the (re)construction of the meaning of a discourse as a unique event of language use. My prediction would therefore be that language users in cognition do not show cross-domain mappings between, for instance, processes and things in the same way as they do for lexical metaphor and other forms of metaphor. But this would be an empirical question for further research on behaviour.
References Barcelona, A. (ed.) (2000). Metaphor and Metonymy at the Crossroads: A Cognitive Perspective (Berlin: Mouton de Gruyter). Bovair, S. and Kieras, D. (1985). ‘A guide to propositional analysis for research on technical prose’. In B. Britton and J. Black (eds.), Understanding Expository Text (pp. 315–62) (Hillsdale, NJ: Lawrence Erlbaum). Bowdle, B. F. and Gentner, D. (2005). ‘The career of metaphor’, Psychological Review, 112, 1, 193–216. Caballero Rodriguez, M. d. R. (2006). Re-viewing Space: Figurative Language in Architects’ Assessment of Built Space (Berlin: Mouton de Gruyter). Cameron, L. and Low, G. (eds.) (1999). Researching and Applying Metaphor (Cambridge: Cambridge University Press). Cameron, L. (2003). Metaphor in Educational Discourse (London: Continuum). Charteris-Black, J. (2004). Corpus Approaches to Critical Metaphor Analysis (Basingstoke: Palgrave Macmillan). Chiappe, D. L. and Kennedy, J. M. (2001). ‘Literal bases for metaphor and simile’, Metaphor and Symbol, 16, 3 & 4, 249–76. Chilton, P. (1996). Security Metaphors: Cold War Discourse from Containment to Common House (New York: Peter Lang).
38
Gerard Steen
Dirven, R. and Pörings, R. (eds.) (2002). Metaphor and Metonymy in Comparison and Contrast (Berlin: Mouton de Gruyter). Gibbs, R. W., Jr. (1994). The Poetics of Mind: Figurative Thought, Language, and Understanding (Cambridge: Cambridge University Press). Gibbs, R. W., Jr. (1999a). Intentions in the Experience of Meaning (Cambridge: Cambridge University Press). Gibbs, R. W., Jr. (1999b). ‘Researching Metaphor’. In L. Cameron and G. Low (eds.), Researching and Applying Metaphor (pp. 29–47) (Cambridge: Cambridge University Press). Gibbs, R. W., Jr. (2000). ‘Making good psychology out of blending theory’, Cognitive Linguistics, 11, 3&4: 347–58. Gibbs, R. W., Jr. (2005). Embodiment and Cognitive Science (New York: Cambridge University Press). Gibbs, R. W., Jr. (2006). ‘Introspection and cognitive linguistics: should we trust our own intuitions?’ Annual Review of Cognitive Linguistics, 4, 135–52. Glucksberg, S. (2001). Understanding Figurative Language: From Metaphors to Idioms (Oxford: Oxford University Press). Goatly, A. (1997). The Language of Metaphors (London: Routledge). Halliday, M. (1985). An Introduction to Functional Grammar (2nd edn) (London: Edward Arnold). Halliday, M. and Matthiessen, C. M. I. M. (1999). Construing Experience Through Meaning: A Language-based Approach to Cognition (London: Cassell). Kintsch, W. (1998). Comprehension: A Paradigm for Cognition (Cambridge: Cambridge University Press). Koller, V. (2004). Metaphor and Gender in Business Media Discourse: A Critical Cognitive Study (Basingstoke: Palgrave Macmillan). Kövecses, Z. (2002). Metaphor: A Practical Introduction (Oxford: Oxford University Press). Lakoff, G. (1986). ‘The meanings of literal’, Metaphor and Symbolic Activity, 1,4: 291–6. Lakoff, G. (1993). ‘The contemporary theory of metaphor’. In A. Ortony (ed.), Metaphor and Thought (2nd edn: pp. 202–51) (Cambridge: Cambridge University Press). Lakoff, G. and Johnson, M. (1980). Metaphors We Live By (Chicago: Chicago University Press). Lakoff, G. and Johnson, M. (1999). Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought (New York: Basic Books). Mann, W. and Thompson, S. (1988). ‘Rhetorical structure theory: toward a functional theory of text organization’, Text, 8, 3: 243–81. Murphy, G. (1996). ‘On metaphoric representation’, Cognition, 60, 2: 173–204. Murphy, G. (1997). ‘Reasons to doubt the present evidence for metaphoric representation’, Cognition, 62, 1: 99–108. Panther, K.-U. and Radden, G. (eds.) (1999). Metonymy in Language and Thought (Amsterdam: John Benjamins). Panther, K.-U. and Thornburg, L. (2003). Metonymy and Pragmatic Inferencing (Amsterdam: John Benjamins). Perfetti, C. (1999). ‘Comprehending written language: a blueprint of the reader’. In C. Brown and P. Hagoort (eds.), The Neurocognition of Language (pp. 167–208) (Oxford: Oxford University Press).
Three Kinds of Metaphor in Discourse
39
Pragglejaz Group. (2007). ‘MIP: A method for identifying metaphorically used words in discourse’, Metaphor and Symbol, 22, 1: 1–39. Simon-Vandenbergen, A.-M., Taverniers, M. and Ravelli, L. (eds.) (2003). Grammatical Metaphor (Amsterdam: John Benjamins). Steen, G. J. (1994). Understanding Metaphor in Literature: An Empirical Approach (London: Longman). Steen, G. J. (2002). ‘Metaphor identification: a cognitive approach’, Style, 36, 3: 386–407. Steen, G. J. (2004). ‘Can discourse properties of metaphor affect metaphor recognition?’ Journal of Pragmatics, 36, 7: 1295–313. Steen, G. J. (2005). ‘Metonymy goes cognitive linguistic: introduction to special issue’, Style, 39, 1: 1–11. Steen, G.J. (2006). ‘Discourse functions of metaphor: an experiment in affect’. In R. Benczes and S. Csábi (eds.), The Metaphors of Sixty: Papers Presented on the occasion of the 60th Birthday of Zoltán Kövecses (pp. 236–44) (Budapest: School of English and American Studies, Eötvös Loránd University). Steen, G. J. (2007). Finding Metaphor in Grammar and Usage: A Methodological Analysis of Theory and Research (Amsterdam: John Benjamins). Steen, G. J. and Gibbs, R. W., Jr. (2004). ‘Questions about metaphor in literature’, European Journal of English Studies, 8, 3: 337–54. Van Dijk, T. A. (1997). ‘Cognitive context models and discourse’. In M. Stamenov (ed.), Language Structure, Discourse and the Access to Consciousness (pp. 189–226) (Amsterdam: John Benjamins). Van Dijk, T. A. (2006). ‘Discourse, context, and cognition’, Discourse Studies, 8, 1: 159–77. Van Dijk, T. A. and Kintsch, W. (1983). Strategies of Discourse Comprehension (New York: Academic Press). Van Oostendorp, H. and Goldman, S. (eds.) (1999). The Construction of Mental Representations during Reading (Mahwah, NJ: Lawrence Erlbaum). Weaver, C. A., III, Mannes, S. and Fletcher, C. A. (eds.) (1995). Discourse Comprehension: Essays in Honor of Walter Kintsch (Hillsdale, NJ: Lawrence Erlbaum).
4 Reading Sonnet 30: Discourse, Metaphor and Blending Paul Chilton
Introduction Close analysis of small texts can reveal the potential conceptualizations prompted by linguistic material. Such texts are in themselves ‘discourse’ but also depend on the surrounding context, which is also discourse. This chapter introduces three levels of discourse and, using Blending Theory, shows how cross-domain conceptual relationships of various kinds are linked to the three levels. We will observe a small text, Shakespeare’s sonnet 30, in which concentrated cognitive effects are stimulated, but which is simultaneously dependent on the cultural environment of discourses.1 This means that the methodology focuses on the complexity of a single discourse event. In the history of science, intense investigation of single so-called ‘golden events’ (Galison 1997) has been contrasted with the large-scale search for generalization, a distinction that Coulson and Oakely (2000) have compared to the different approaches to linguistic and associated cognitive phenomena. While this chapter focuses on a ‘golden event’, no linguistic event can be separated from its context. This means that the starting point will be certain theoretical notions about discourse, and the aim will be to describe how metaphorical effects are related to discourse. By metaphorical ‘effects’ here, I mean the emergence in the course of discourse processing of relationships of various kinds across conceptual domains. Because this chapter is concerned with discourse processing, conceptual emergence and coherence, it is methodologically natural to couch the analysis in terms of Blending Theory (BT) which is explicitly a theory of processing, as well as in terms of Conceptual Metaphor Theory (CMT).
40
Reading Sonnet 30 41
1
Discourse
In the metaphor research literature, explicit recognition that metaphorical expressions recruit conceptual frames embedded in culture seems to have come a little late, although the notion is adumbrated clearly enough in Lakoff and Johnson (1980) in their chapter on metaphor and cultural coherence. Gibbs (1999) makes the crucial point that cultural experience is just as embodied as supposedly individual primary metaphors. Another way of investigating the cultural basis of metaphors and their expression in language use is to reframe the issue in terms of ‘discourse’. The term covers various notions current in the social sciences and humanities that are more or less relevant to the issue under discussion here. I shall distinguish three broad understandings of the term. 1.1
Microdiscourse: linguistic prompts to conceptualization
The term ‘microdiscourse’ is used here to denote those aspects of linguistic organization that provide conceptual coherence for a discourse processor. Classic examples are deictic expressions and anaphora. Formal and computational pragmatics has concerned itself with the coherence of stretches of discourse and produced sophisticated models such as Kamp and Reyle’s Discourse Representation Theory (Kamp and Reyle, 1993). Later developments of this approach have sought to formalize ‘rhetorical’ aspects of discourse (Asher and Lascarides, 2003). The idea that discourse processors seek conceptual coherence has been a central underlying assumption of this kind of work. Not surprisingly, the question of how metaphor is recognized, understood and processed has been a continuing concern in several contiguous traditions, including the philosophy of language (see for instance, contributions in Sacks, 1980), in pragmatics (see Sperber and Wilson’s 1986 revision of Grice) and in the logical and computational tradition already mentioned (see Asher and Lascarides, 2001, 2003). In Cognitive Linguistics, work on discourse processing is comparatively under-developed. However, Blending Theory (e.g. Fauconnier and Turner, 2002) and subsequent work have developed detailed principles and models claiming to outline how complex conceptualization, including metaphorical effects, emerges from the online processing of linguistic and other stimuli, i.e. discourse. Langacker (2001) has outlined a model of on-line processing that is complementary to BT, and indeed makes use of concepts such as integration (cf. coherence) and mental spaces. Because BT is explicitly a theory about discourse
42
Paul Chilton
processing, the following discussion will be cast in terms of that theory. However, discourse processing and thus BT may draw on conventionalized metaphorical mappings and metaphorical expressions. In the analysis of sonnet 30, I shall take it for granted that the microdiscourse level, and in particular the principle of coherence, is a prerequisite. 1.2
Discourse as a dynamic system
As Cameron and Deignan (2006) argue, discourse can be seen as a complex dynamic system in social groups of various kinds and sizes. It is in such groups that linguistic expressions emerge in particular periods and are liable to become entrenched and conventionalized. This is a phenomenon that applies to all types of linguistic expressions, not merely the metaphorical expressions that Cameron and Deignan discuss. For example, metaphorical expressions may emerge and conventionalize within a small group or professional practice, such as classroom discourse. On a longer timescale, Fauconnier and Turner (2002) argue that cultural artefacts such as complex numbers are based on a succession of integrated blending operations until they become conventionally stabilized. In general, the position adopted here is that language forms are always prompting for particularized meanings in contexts of use, some of which become entrenched, conventionalized, and in their turn serve for the prompting of particularized meaning. It is a feature of the text we will examine that some of the expressions are being used in a wider discourse setting in which their meanings may be in the process of dynamic diversification in the speech community. 1.3
Macrodiscourses
The process sketched above can lead to ‘discourse practices’ and ‘discourse formations’ – terms that are associated with the study of discourse at the level of societies and cultures. At this level ‘discourses’ are relatively stable ways of conceptualizing objects, events and relationships. Discourses in this sense also assign speaking and listening roles. Furthermore, language is not the sole vehicle of social signification but involves also, for example, architectural dispositions. Viewing discourse in this way means that it may be closely linked with social and political institutions, their characteristic linguistic forms being bound up with both conceptualization and interaction among individuals and groups. Such an approach has been adopted in Critical Discourse Analysis (CDA), whose proponents are preoccupied with discursive manifestations of power. (e.g. Fairclough 1989; Wodak, 1996). In principle, CDA ought to be able to demonstrate a relationship between the microdiscourse structures of any text and the macrostructures of a
Reading Sonnet 30 43
society. Such a relationship does turn out to be relevant for an analysis of the functioning of metaphor in sonnet 30.2 However, CDA can tell us nothing about the intricate conceptual effects that the linguistic structure of the text can trigger.
2
Sonnet 30
Variant texts are capable of producing variant conceptual effects (or meanings). The version used here is that of Shakespeare (1609). This means its orthographic and other characteristics will link it with the linguistic and discourse setting contemporary to it. 1 When to the Sessions of sweet silent thought, I sommon vp remembrance of things past, I sigh the lacke of many a thing I sought, And with old woes new waile my deare times waste: 5 Then can I drowne an eye (vn-vs’d to flow) For precious friends hid in deaths dateles night, And weepe a fresh loues long since canceld woe, And mone th’expence of many a vannisht sight. 9 Then can I greeue at greeuances fore-gon, And heauily from woe to woe tell ore The sad account of fore-bemoned mone, Which I new pay as if not payd before. 13 But if the while I thinke on thee (deare friend) All losses are restord, and sorrowes end. I will assume that readers will seek maximal conceptual coherence consistent with the linguistic prompts in processing a text of this kind, perhaps as a result of cultural norms. The important point, however, is that in achieving such coherence readers will experience a variety of conceptual effects, some of which may be categorized as metaphorical. How do we go about describing such effects? Here I will adopt an introspective process structured by the theories of CMT and BT, but will also attempt to expose some of the backstage processing (to use Fauconnier’s term) that yields the felt effects. To do this it will be necessary to refer to several levels of linguistic structure, segmental and suprasegmental aspects of phonology, lexical relations, grammatical constructions and text organization. This has to be done not by separating each level but by relating them as necessary to explain the effects. The reading experience of sonnet 30 is dynamic: integrated conceptual spaces are successively built up, modified and partially resolved.
44
Paul Chilton
3
Linguistic prompts
Many metaphor effects in this text are not triggered by semantic or conceptual clashes evident in argument–predicate structure as coded in linguistic form. For example, in line 14, the clause ‘All losses are restord’ is perfectly comprehensible without appeal to metaphorical reading. However, if a reader is seeking to maximize coherence in the light of the surrounding text, then a conceptual relationship is available between the generic sense of ‘loss’ and a specific financial or legal sense. There is thus a kind of cross-space mapping which can be called a metaphor effect. This does not mean that grammatical constructions are not an essential element. In fact they make a crucial contribution to the overall conceptual response of readers. The emphasis on the conceptual (and indeed on the propositional) has tended to obscure the linguistic coding of metaphor, although traditional studies have offered important insights (e.g. Brooke-Rose, 1958). Cognitive Linguistic principles should point us toward more detailed study of such matters, but the point cannot be elaborated here. Conversely, some metaphor effects are prompted not by grammatical structures at all but by semantic properties of lexical items or by phonological properties. Such structures can prompt for metaphor effects, although they are not detectable as metaphorical expressions by the test of semantic anomaly or grammatical form. 3.1
Polysemy and ambiguity
Polysemy is a synchronic language-wide phenomenon produced diachronically. Ambiguity is a discourse phenomenon – the reader has to ‘entertain’ two meanings simultaneously, that is, either oscillate between them or blend them. There is a crucial polysemy and ambiguity in line 1 of sonnet 30, as explained in more detail below. The word sessions has emerging polysemy in the contemporary English of Shakespeare and in the text two meanings can be simultaneously called up. These two senses are then related to one another as a metaphor effect produced by blending, specifically lexical compression. Ambiguous lexical items can thus give rise to metaphor that is not detectable by existing automatic corpus-searching procedures. 3.2 Iconic blends Phonological structures (segmental and suprasegmental) give rise to iconic effects, in conjunction with grammatical and lexical cues. To
Reading Sonnet 30 45
be more concrete, sound similarities between pairs of units blend with grammatical and conceptual pairs.3 Similar pairings occur throughout sonnet 30, and will be mentioned also below, but here is one example: Then can I greeue at greeuances fore-gon (line 9) Rhythmically the line has a caesura splitting it after ‘greeue’ ‘grieve’. Thus greeue occurs in the first hemistich and greeuances in the second, in opposing halves or mirrors. They are similar phonologically but, again, they are not totally similar semantically, because of the quirks of the historical meaning extension of greeue and greeuance. The latter evokes a legal frame (as well of course as a non-legal emotional one); the former evokes only an emotional meaning, especially related to loss. Moreover, ‘greeue’ is a present act, while ‘greeuances’ are past, or ‘foregon’. To make things more complex, ‘fore-gon’ is a case of the kind of ambiguity noted above. It not only means ‘gone by’, but voluntarily ceded. So why grieve? This confluence of phonological and semantic effects yields complex cognitive effects. These are probably not what one would describe as metaphorical, but the structures do prompt a fusion of times and not-quite-compatible concepts that produces (in the blend space) a sense of paradoxicality and perhaps, therefore, absurdity and futility. We shall see more instances of this below, but let us also note the pairing with ‘fore-bemoned mone’ (line 11), where similar effects are triggered by partly similar means.
4
Metaphor in relation to discourse
Many metaphor-inducing expressions are conventionalized. To be conventionalized does not necessarily mean to be in the lexicon of every speaker of a language or that its use is not constrained by situations of use. It may be more accurate to say that an expression is conventionalized in certain discourses that are in certain genres sanctioned by the social system. This observation may enable us to make a distinction between entrenched (a cognitive notion) and conventionalized (a societal notion). It is also important to distinguish metaphor from metaphorical expression, in the way usual in CMT. A metaphorical expression, or a metaphor triggered by various possible linguistic forms, may be entrenched, in the sense that its form is accessible as a unit from memory and may not necessarily require pragmatic computation on each occasion of use. However, consistently with Cameron and Deignan’s conception of ‘dynamic discourse’, we can say
46
Paul Chilton
that a metaphorical expression is conventionalized to a certain genre or discourse practice. The same may or may not be true for the conceptual metaphor underlying the metaphorical expression, depending on the discourse in question. In sonnet 30 we can say that there are conventionalized and entrenched metaphors, although their metaphorical expression may not be conventional. And ‘deaths dateles night’ is an expression of DEATH IS NIGHT . These expressions link the poem with the literary discourse environment. It is possible also that they can be read as ironic, but the point cannot be pursued here. There are two metaphors that structure the sonnet, although these are interwoven with other kinds of cross-domain relations, as we shall see in the next section. The first is: VALUED EXPERIENCES ARE VALUED POSSESSIONS. The frequent mention of this metaphor in the CMT and BT literature is testimony to its entrenched and conventionalized character (e.g., Lakoff and Johnson, 1980; Lakoff and Turner, 1989; Grady, 1999) In societies that have it, money is a valued possession or a valued token for possessions, so money is subsumed under the same generic metaphor. Since life is the most fundamentally valued experience and life is the passing of time, a life, one’s time, is a valued possession and also money. It could be argued that such metaphors are culturally embedded: their use is dependent on basic economic structure and cultural values and attitudes. In sonnet 30 this metaphor, minimally evoked by lexical triggers (deare, precious, losses, waste), is integrated with a metaphor from a more socially specific domain. This second metaphor is prompted by the ambiguity of the word ‘Sessions’ and is connected to macrodiscourses in a way that requires further examination. A twenty-first-century reader may process line 1 by accessing the contexts in which session frequently appears in English: She had a session with her counsellor. Indeed, it is possible to blend session with almost any noun denoting an activity: a drinking session, poker session, a running session, rap session, recording session.4 There are two grammatical constructions that prompt for such blends: noun-compounding with session as the second and head noun; and the NP of NP construction. Thus in the latter case we can say: a session of poker. These involve blends because the second noun provides input that fills out the input provided by session. In contemporary English this input appears to consist of something like ‘bounded activity’ A ‘session of’ fulfils a partitive or unitizing role, such that a session of delimits a non-delimited activity such as drinking with friends or arguing with partners. The examples that appear to suggest a particular
Reading Sonnet 30 47
association with musical performance, dialogue, sport are not part of the semantics of session but are due to conventionalization emergent in specific cultural settings. None of these cases strike one as metaphorical. However, the modern meaning of session includes examples such as session of the court, parliament is in session, which relate to social institutions. We can also say a sitting of the court, or parliament is sitting, where sitting corresponds to the original conceptual (metonymic) motivation of session (Latin sedere). In England in 1609, readers are quite likely to have had in mind this kind of entrenched and conventionalized meaning for session, a meaning grounded in contemporary social structures and practices and their constitutive discourses. We therefore can reasonably assume that sonnet 30 sets up a metaphor CERTAIN KINDS OF SOLITARY THOUGHT ARE SESSIONS OF A LAW COURT. But, it is also possible that the word was in the process of developing its more generalized modern meaning, for we find it in other literary texts in a similar expression, for example, from John Dickenson (1594): Being in these dumpes he held a session in his thoughtes.5 From this example, we cannot really be sure whether a contemporary reader of Dickenson’s text already had an entrenched bleached meaning for session, or whether the reader would spontaneously but online infer the meaning by blending input from the ‘legal sessions’ frame with the ‘thought’ frame. The similarity with Shakespeare’s wording is striking. Maybe they were both using an entrenched or emergent blend, specifically linked with thought. In any event, the use of the capital letter in the 1609 spelling of sonnet 30 (the only use of a capital for the initial letter of a noun in the text) can be seen as a prompt for the legal institution frame. The legal ‘Sessions’ frame would clearly be established in the longterm memory of readers of sonnet 30. In England at this period (and up to the revision of the court system and its nomenclature in 1972) courts sat (were in session) at periodic intervals in particular geographic locations. For example, the Petty Sessions were regular local courts presided over by two justices of the peace within a particular district to deal with minor offences and Quarter Sessions were courts that sat in the counties four times a year, with limited criminal and civil jurisdiction and certain administrative functions. In Elizabethan society, which was highly litigious, such courts dealt with various forms of grievance, but cases heard at Sessions frequently had to do with the examination of accounts, payment of debts and the recovery of financial and other material losses.
48
Paul Chilton
Now, social institutions such as the Sessions are produced, sustained and constituted through discourse in the various senses discussed earlier. They are chronotopically embodied – that is, they are a part of physical situated experience of places and times. Sessions are only possible in certain buildings in certain cities at certain points of the annual cycle. Further, specific speech roles and styles of speech are prescribed. This information is included in the cognitive frame for session, as well as emotions such as guilt, sense of grievance or resentment, satisfaction, etc. Moreover, these cognitive frames are not a matter of individual cognition, but a function of inter-subjective collective cognition. In terms familiar in CDA, they are part of a ‘discursive practice’ that is ‘articulated’ with a whole ‘discourse formation’ constituting the power of the state in early modern England. We noted above that metaphor expressions frequently emerge and come into circulation in discourse within a particular community of speakers. Historically this may be linked to the establishment of fundamental social institutions – the Sessions are indeed an instance. In such cases basic image schemas frequently provide source domains: sitting acts metonymically for an entire socio-conceptual frame. But in sonnet 30 we have a further link in the dynamic chain of metaphor formation in a community of speakers. It is the socio-conceptual frame ‘Session’ itself that is recruited as source domain input to a new conceptual integration, guided by the linguistic structures of the text. This is the point at which we need to focus on the way these structures are functioning in sonnet 30. On the CL hypothesis about linguistic meaning, the word session has a meaning by virtue of accessing a knowledge frame (cf. Evans, in press; also Feldman, 2006), illustrated in Table 4.1.
Table 4.1 The Sessions frame Place time participants: judge, plaintiff, defendant, jury court summons defendant court proceedings: charges account of events losses claimed losses restored or not restored, etc.
Reading Sonnet 30 49
Table 4.2 The debt frame Money owed pay-by date repay or cancel debt or fail to do so count (tell) the money, etc.
Since cases heard at Sessions were frequently about material losses, especially financial ones, we need also to outline the debt frame (see Table 4.2 ). The text of the sonnet calls up these frames in unobtrusive ways, but they are necessary for the achievement of maximal coherence in reading the text. The word ‘Sessions’ in the first quatrain does more than quantify ‘thought’, enabling us to think of repeated occasions of thinking; it also and perhaps predominantly calls up the court frame, as we have seen. However, this frame is sustained in minimal ways and mainly by the use of contextually ambiguous expressions. Thus in quatrain 1, ‘sommom vp’ is understandable both in its legal sense and in its extended bleached sense. If ‘Sessions’ activates the legal sense, then in line 2 the contemporary legal sense of ‘remembrance’ (summarizing in court) may also be activated alongside its ordinary meaning denoting the mental process of remembering, as well perhaps as its sense of ‘commemoration’. A metaphorical correspondence between court Sessions and thought sessions is thus active. Lines 3 and 4 activate the metaphor time is money, a metaphor that may have been already conventional or in the process of becoming conventionalized. However, there is more, since in the context of the sonnet as a whole, ‘lacke’ and ‘waste’ overlap semantically with ‘expence’ and with the crucial word ‘loss’ in line 14. All these are consistent with the conventional metaphor VALUED EXPERIENCES ARE VALUED POSSESSIONS, its special case TIME IS MONEY , and the more specific case of the debt frame. Quatrain 2 continues to activate both the court frame, and the debt frame . The vocabulary of emotional loss is blended with these frames, by way of a lexical set that is linked on several levels. ‘Woe’ and ‘waile’ semantically overlap with with ‘weepe’ in the second quatrain (and the repetition of ‘woe’ in the same line and in line 10); ‘woe’ also phonologically overlaps with the final syllable of ‘sorrows’ in line 14. The key point here is that this vocabulary is linked to the court and debt frame by way of lexical polysemy and grammatical construction. Thus, ‘greeue’ is
50 Paul Chilton
linked to this set of emotion words but also punningly to the legal and other sense of ‘greeuance’ , as noted earlier. In addition, the lexical item cancel in ‘canceld woe’ (line 7) evokes the debt and court frames, as does the ambiguous expression ‘account’ in ‘tell ore/The sad account’. To ‘tell ore’ the account is to count it, re-count it and recount it – the multiple sense being fused in the online blend, the reader being required to call up the debt and court frames specifically in line 12, where the account is ‘payd’ again. To what extent are source domains (debt and court) coherent with the target domain (bouts of melancholic thought)? The topologies stipulated by the court input and that stipulated by the poem for the thought input do not completely match. In the court frame there is a complainant, whose counterpart is ‘I’ who complains (moans, wails about things he has had but no longer has). However, one cannot go to a court to complain about things that one has not got because one has not been successful in seeking them. Nor can one complain about the loss of things that have simply ‘vannisht’. Nor indeed can one complain about loss of money that one has spent or ‘wasted’. Furthermore, the judge or court officer who ‘summons’ an individual cannot be the same as the complainant, while in the sonnet ‘I’ both summons and complains. There are also elements that have no counterparts: there is no judge, or verdict or sentence. Now it might be said that this being so, we do not have metaphor in this sonnet at all. But intuitively this seems to belie the reading experience. Blending Theory provides a more satisfactory framework of analysis. There are elements of cross-domain mapping – the court Sessions is indeed somehow compressed with the melancholic thought session and absence of valued life experiences is somehow compressed with loss or absence of valued objects. This kind of integration is what BT calls ‘double-scope’ integration. Different elements from the inputs are mapped to the blend space. Figure 4.1 is a partial sketch of the integration network . The generic space is a schematic structure such as: absence of desired object – desired re-possession of desired object. The blend inherits some structure from the court and debt input. Bringing a grievance concerning financial loss to a court can result in restoration of that loss, but it does not inherit a judge whose order causes restoration of the loss – a crucial point when we come to the final couplet of the sonnet. The same frame also projects the emotional sense of grievance, fusing with the emotional content of the ‘thought’ space that includes the melancholic recall of
Reading Sonnet 30 51
absence of desired object
desired re-possession of desired object
GENERIC SPACE INPUT SPACE 1
INPUT SPACE 2
plaintiff place, time telling account
I thinking recalling separate
losses, grievances
past experiences (love, frustration, resentment, woes)
judge defendant sentence
I/plaintiff
Sessions/thinking
recounting/remembering lost possessions/past experiences
BLENDED SPACE
Figure 4.1 Partial integration network for debt-court-thought sessions blend for lines 1–12 of sonnet 30.
past losses and past non-achievements. This is a rough outline of part of the conceptual model produced by processing the discourse of lines 1 to 12. But it is only a part. Section 5 attempts to outline another crucial conceptual network that is integrated with it. Section 6 outlines the cognitive effects of the couplet which produce a dramatic effect with respect to the preceding lines.
52
Paul Chilton
5 Temporal and counterfactual blends There are several time spaces and some time compressions in this small text. The structure When ... then ... then sets up a mental space that contrasts a certain kind of mental activity – remembering past experiences – with the absence of such activity (the times when ‘I’ is not engaging in remembering the past). Since when can be read as whenever (habitual aspect), separate sessions, separated in time, are compressed. The then clauses are all modalized by ‘can’, readable as expressing potentiality: I am enabled by the circumstance or it is possible that. Thus the potential is embedded in the habitual. A particular potential action, weeping, simultaneously evokes its habitual absence via the semantics of ‘vn-vs’d’ (‘unused’ ) which fuses occurrence with a background of non-occurrence. This kind of semantics has been described by Fauconnier and Turner (2002: 217–47). In sonnet 30 it is repeated lexically in ‘lacke’, ‘sought’, ‘waste’, ‘canceld’, ‘expence’, ‘fore-gon’ (in both senses) and in the if-construction (line 12), ‘as if not payd before’. Counterfactual if-constructions involve simultaneous conceptual representation of the occurrence and non-occurrence within the subject’s reality space (cf. Chilton 2005 for a more formalized account of counterfactuals). Another kind of irrealis space is set up in lines 13–14, built both by ‘But’, which instructs us to set up a space with content disanalogous to what has just preceded, and also by ‘if’, which is this time not counterfactual but potential, analogous with the potentiality of the earlier can-based space. The two parts of the conditional construction in the couplet also invite a causal interpretation, linking two events as cause and effect. The semantics of verbs of remembering inherently compress a present mental action with an event located in a past relative to the rememberer. Event and emotions that are past and completed are repeated mentally in the present. Embedded in these re-called event spaces are losses and lacks, which in their turn are compressions. Loss evokes the present absence of a previously possessed object and simultaneously its possession in a past time relative to the loser. A sought object presupposes the absence of an object that is desired but not yet achieved in the future relative to the person desiring. In the sonnet ‘I’ remembers a past lack, a state in which ‘I’ desired something in a past future, a state that is now, in the present, the past desire of an object that was never possessed. Figure 4.2 shows in a rather impoverished way something of the conceptual complexity that may be experienced when processing the
Reading Sonnet 30 53
Event structure
PRESENT EVENT: REMEMBERING
PAST EVENT
mental representation e9
Past event e I
I
sigh
seek
lack of thing
thing
E I seek sigh thing
REMEMBERING = BLEND
Figure 4.2
Temporal-counterfactual integration in quatrain 1 of sonnet 30
discourse that is quatrain 1. It leaves out the fact that line 1 assigns the entire network to a base space, triggered by ‘to the Sessions/sessions of sweet silent thought, I sommon vp’. What this inserted network does is set up a remembering blend, prompted by ‘sommon vp remembrance’.6 There is a past event space (e) and a present space for the mental representation (e′) of I. In the past event space, there is an event of seeking to possess a desired object. The verb seek itself sets up a blend, as noted above. The thing is not real, but imagined as a possible acquisition in the relative future. In the present space the thinking subject I is identical
54
Paul Chilton
with I in the past event space. In the past space, I does not know if he will possess the object. In the present space, he knows he does not. The semantics of the word lack also involves a present–absent blend. There are identity mappings for I in each space and for thing – presumably it is the same thing or things in both spaces. However, the present has the action of sighing, while the past space does not. The thing in both input spaces is an embedded irrealis mental space, slightly different in each case. The blend is a typical remembering blend, fusing the past event with the present mental activity. Specific fusions include the present and past selves. In the blend I sighs (for something past) and seeks (for something future) simultaneously – an effect strengthened (a) by the alliteration and orthographic similarity of sigh/sought and (b) by the rhythmic trick that puts these words in parallel rhythm spaces on each side of the caesura. This is not all. The crucial effect is the conceptual blend around thing which is prompted inter alia by the grammatical construction in which it occurs. It is (part of) the complement of presenttensed sigh and the complement of past-tensed sought. It is also a fusion of two irrealis spaces – in the past space it was not-yet-present for the speaker (and as we learn, never present), and in the present space it is absent but has counterfactual presence. Both these slightly different modal spaces are compressed in the blend. Furthermore, the blending mind can infer a causal relation between the blended thing concept and the present act of sighing. I have attempted to describe the detail here because it shows the dependency of the conceptual integration on a variety of linguistic features. That is to say, the discourse structure can lead to conceptual compressions of at least some features of language that defy the usual duality of language, viz. the arbitrariness of the relationship between phonology and meaning. Sounds, syntax and semantics get blended in poetic texts.
6
Running the couplet blend in the culture
The couplet has the magical effect of resolving, cancelling the previous conceptual network structure. It is a dynamic effect of the reading process, which requires that one make a mental model of the entire small text and seek coherence. Working memory holds the conceptual structure evoked by the three quatrains. The couplet builds a new structure that dissolves them. For this to happen, both the quatrain structure and the couplet structure must meet in a hyper-blend space. When they are held together like this they interact in a kind of dialectic that makes
Reading Sonnet 30 55
them antithetical but synthesizes them. This is indeed the art of the concetto cultivated in the Renaissance sonnet. What can happen in the blend? ‘Running the blend’ may involve further input from surrounding discourses as well as construal based on the linguistic forms of the metaphorical expressions in the text, pattern completion and inferencing. The passive construction occurs only once in the text, in the final line, and without an expressed agent. Readers automatically seek to infer agents on the basis either of co-text or background cultural knowledge – an example of pattern completion. Furthermore, in the metrically parallel half-line, intransitive ‘end’ does not semantically entail agency either. These are linguistic prompts that require resolution in normal discourse, and here they contribute to ‘running the blend’. Restoring a loss or recovering a debt, concepts supplied by the debtcourt input frame, are technically called ‘redemption’. In contemporary theological discourse, writers were often aware of this metaphorical basis to the doctrine of ‘redemption’. Fauconnier and Turner (2002: 259–60) discuss redemption as a conceptual blend: past wrongs are both present and absent in the blend. They regard this as an entrenched and conventional blend in the culture. But which culture? In fact, they treat the redemption frame as one in which it is always the individual who works his or her own redemption. This is clearly not the case in the possible blend scenario I am about to suggest for sonnet 30. But this is not a counter-indication to the possibility that readers run surrounding discourses into the blend. What it means is that Fauconnier and Turner are assuming a particular cultural discourse of redemption, probably that of the American myth of individual freedom and the self-made man. This is another way in which discourse and culture are inextricably bound up with metaphor and blending, in this instance with the process of analysis itself. By contrast, what Shakespeare’s discourse (in the sense of both microand macrodiscourse) does is different, and it is partly a reflection of the cultural environment and partly not. A reader might simply ascribe no agent to the redemption, but this is unlikely, since conditional if and conjunctive and generally implicate causality. The reader then seeks to infer an agent, that is, a redeemer who is not I. Where does a reader look? The first place to look is the micro-text, where the ‘deare friend’ is a candidate. But who, the reader may ask in running the blend, is this dear friend? The preceding sonnets famously point to a young man, and the title page famously refers to ‘Mr W.H.’ Or maybe it is the dark lady of the later sonnets in the sequence, or maybe the poet’s wife.
56 Paul Chilton
This is all discourse material that may enter the running of the blend. Piously minded readers could interpret the entire sonnet as a religious meditation in which the unnamed dear friend is the Redeemer.
7
The metaphor–discourse nexus
At the microdiscourse level, identifying metaphorical expressions alone cannot explain the cognitive effects of sonnet 30: Blending Theory seems to take us further. One crucial reason why BT is required is the role played by ambiguous discourse expressions. Of course, one kind of reading practice might involve choosing alternate senses. But choosing one or the other (e.g. either based on the Sessions frame, or one based on sessions as temporal quantifier) leads to an unsatisfactory reading. Why unsatisfactory? Because it does not maximize coherence. Selecting the temporal sessions reading leads to lack of coherence – why the legal and financial metaphors? Choosing the debt-court frame means looking for consistent cross-domain correspondences – but they are not all systematically mapped, so gaps are filled by conceptual integration. The assumption here is that readers seek maximal conceptual coherence of small texts. At the level of macrodiscourses, we have seen that a crucial part of the metaphoric structuring of sonnet 30 recruits knowledge of discourses that constitute the historical socio-political context – the legal institutions of Tudor and Stuart England. The discourses maintain cognitive frames that can be called up to maintain the macrodiscourse of governance. What can we say about the relationship between poetic text, metaphor and the socio-political discourse? A CDA response might be: such elite texts naturalize and legitimize the power of the contemporary state apparatus. But such interpretations are not a logical consequence of the analysis; they are an ideological choice. More rational is to consider alternative interpretations. For instance, one might equally say, the text exploits the institutions of power for its own ends, the elucidation of complex individual states of mind that are not individual but recognizable as universally human. Indeed, the Cognitive Linguistics approach would argue that the cognitive operations we have examined are reflections of the universal potential of the human mind.
Notes 1. Sonnet 30 and other Shakespeare sonnets have drawn the attention of a number of linguists and linguistically oriented critics (Levin, 1962; Jakobson and Jones, 1970; Lakoff and Turner, 1989: 26–34; Turner, 1991: 89)
Reading Sonnet 30 57 2. Chilton (1977) shows precisely this – a relationship between metaphorical expressions and macrodiscourses in the case of a collection of sonnets roughly contemporary with Shakespeare’s. 3. Levins (1962) points about ‘coupling’ in sonnet 30, and Jakobson and Jones (1970) on ‘parallelisms’ in sonnet 129. 4. The OED gives the 1920s as the date for the appearance of such examples. 5. John Dickenson, Arisbas, Euphues amidst his slumbers: or Cupids journey to hell 1594 (Grosart 1878), cited OED online, entry for session. 6. ‘Sommon vp’ involves two metaphors, one dependent on the court frame, the other based on an entrenched metaphor involving the UP –DOWN image schema.
References Asher, N. and Lascarides, A. (2001). ‘Metaphor in discourse’. In P. Bouillon and F. Busa (eds.), The Language of Word Meaning (pp. 262–90) (Cambridge: Cambridge University Press). Asher, N. and Lascarides, A. (2003). Logics of Conversation (Cambridge: Cambridge University Press). Brooke-Rose, C. (1958). A Grammar of Metaphor (London: Secker and Warburg). Cameron, L. and Deignan, A. (2006). ‘The emergence of metaphor in discourse’, Applied Linguistics, 27, 4: 671–90. Chilton, P. (1977). The Poetry of Jean de La Ceppède: A Study in Text and Context (Oxford: Clarendon Press). Chilton, P. (2005). ‘Discourse space theory: geometry, brain and shifting viewpoints’, Annual Review of Cognitive Linguistics, 3: 78–116. Coulson, S. and Oakley, T. (2000). ‘Blending basics’, Cognitive Linguistics, 11, 3–4: 175–96. Dickenson, J. (1878). Prose and Verse, ed. by A. B. Grosart (Manchester). Evans, V. (in press). ‘A lexical concepts and cognitive models approach to spatial semantics’. In V. Evans and P. Chilton (eds.), Language, Cognition and Space: State of the Art and New Directions (London: Equinox). Fairclough, N. (1989). Language and Power (London: Longman). Fauconnier, G. and Turner, M. (2002). The Way We Think (New York: Basic Books). Feldman, J. E. (2006). From Molecule to Metaphor: A Neural Theory of Language (Cambridge MA: MIT Press). Galison, P. (1997). Image and Logic: A Material Culture of Microphysics (Chicago: University of Chicago Press). Gibbs, R. W. Jr. and Steen, G. J. (eds.) (1999). Metaphor in Cognitive Linguistics (Amsterdam: John Benjamins). Gibbs, R. W. Jr. (1999). ‘Taking the metaphor out of our heads and putting it into the cultural world’. In R. W. Gibbs Jr. and G. J. Steen (eds.), Metaphor in Cognitive Linguistics (pp.145–66) (Amsterdam: John Benjamins). Grady, J. (1999). ‘A typology of motivation for conceptual metaphor: correlation vs. resemblance’. In R. W. Gibbs Jr. and G. J. Steen (eds.), Metaphor in Cognitive Linguistics (pp. 79–100) (Amsterdam: John Benjamins). Jakobson, R. and Jones, L. G. (1970). Shakespeare’s Verbal Art in Th’Expence of Spirit (The Hague: Mouton).
58 Paul Chilton Kamp, H. and Reyle, U. (1993). From Discourse to Logic (Dordrecht: Kluwer). Lakoff, G. and Johnson, M. (1980). Metaphors We Live By (Chicago: Chicago University Press). Lakoff, G. and Turner, M. (1989). More Than Cool Reason (Chicago: Chicago University Press). Langacker, R. W. (2001). ‘Discourse in cognitive grammar’, Cognitive Linguistics, 12, 2: 141–88. Levin, S. (1962). Linguistic Structures in Poetry (‘s-Gravenhage: Mouton). Sacks, S. (ed.) (1980). On Metaphor (Chicago: University of Chicago Press). Shakespeare, W. (1609). Sonnets, ed. by T. Thorpe (London: Aspley imprint). Turner, M. (1991). Reading Minds (Princeton, NJ: Princeton University Press). Wodak, R. (1996). Disorders of Discourse (London: Longman).
5 Collecting Political Meaning from the Count of Metaphor Christ’l De Landtsheer
Introduction This chapter presents a political semantic method and approach to metaphor. The study that is itself rooted within political psychology and persuasion has continued in the tradition of discursive approaches that are particularly concerned with developing methods. Again the assumption is that metaphors in discourse can be studied as political phenomena. We start with an overview of significant contributions and elaborate the role and effects of metaphor in political discourse. The method of the metaphor power model for the (quantitative) political semantic approach is detailed in the second Section. At the heart of the model are the identification and categorization of metaphor. With the help of many examples it is explained how particular political meaning is derived from the calculation of metaphor power in discourse. The third Section discusses some results and implications from empirical studies with the metaphor power model.
1
Metaphor in political discourse
The concepts of metaphor and politics have for the last decades separately been the subject of a surge in intensive research. From an account of political metaphor research we can conclude that little previous effort went into examining the role of metaphor in politics and that the methodology used in investigating the subject received next to no interest (Beer and De Landtsheer, 2004). 1.1 Political metaphor studies Since Lakoff and Johnson argued in their 1980 book Metaphors We Live By that thinking itself is metaphorical, scholars have been increasingly 59
60
Christ’l De Landtsheer
aware of the significance of metaphor in everyday life. With Cognitive Linguistic metaphor theory, a new paradigm was born. Lakoff and Johnson’s book, translated into many languages, launched the discussion among scientific and political elites as to how metaphors generate (political) meaning. If metaphor is at the heart of cognitive framing, then it should be crucial to political study. Scholars in linguistics apparently move toward investigating politics, and political researchers start studying political metaphor. Surprisingly, the amount of literature addressing political metaphor has not until now reflected the importance of the subject. Even though the majority of scholars highly value this type of research, interdisciplinarity has turned out to be complicated, time-consuming, and in the end not always rewarding. Academics produce boundaries and often do not transcend them for discipline-focus reasons. The general discussion about how language and style are likely to affect politics or public opinion (and vice versa) is, especially among social scientists, more popular than the study of political metaphor. One can conclude that most researchers into the language of politics were directly influenced by rhetoric, the oldest approach to persuasion. Aristotle, who described rhetoric as the faculty of observing in any given case the available means of persuasion, was however seriously interested in the role of metaphor in the persuasive process (Hellsten, 2002; Roberts, translation of Aristotle, 2004: 24). Since at least the 1930s, linguistic persuasion in politics has been explored by a number of linguistic philosophers, essayists, columnists, and social scientists. It was in fact the linguistic philosophers of the School of General Semantics who put forward the idea of political and public opinion effects of linguistic operations (Korzybsky, 1933). Orwell (1949) had, as a writer and a journalist, a great impact upon literary and scientific writers with his thesis that political language can covertly influence political thought. Within the social sciences, Lasswell and his disciples (1949, 1952) focused upon the subject as they investigated features and effects of fascist and communist totalitarian political language. Their pioneering work in political psychology and propaganda gave particular attention to methods for analysing political language. One can conclude from an overview of political language research traditions that their so-called quantitative-semantic and political-semantic approaches are the start of the scientific analysis of content and style in politics. Scholars in political semantics clearly preferred qualitative or quantitative empirical cases in which language and symbols were studied for their political meaning. Lasswells’ (1949) essay in which political
Collecting Political Meaning
61
style during crises was described as increasingly emotive and ornamental, is of particular interest for metaphor research. As Lasswell (1949: 38) wrote, certain changes in style may indicate the gradual decline of democratic feeling, or reveal the groundswell of gathering crisis. Of further methodological importance for the political study of metaphor is an analysis of international quality newspapers by De Sola Pool (1956) which found that political symbols of war and totalitarianism show less variety compared to symbols of peace and democracy. The Yale Studies of Persuasion (Hovland et al., 1953) have continued in the tradition of the pioneers while focusing upon effects of rhetoric with experimental social psychology studies (for an account of political language concepts and research, see De Landtsheer, 1998a; Billig, 2003; Wodak and De Cilia, 2006). It will be clear that all of the above mentioned studies, especially those from propaganda and persuasion studies, have somehow contributed to the elaboration of the metaphor method and model that are presented in this chapter. 1.2 Metaphors and their (political) effects It has been argued by a number of scholars that metaphors have considerable political effects. Metaphors are crucial to the language of foreign policy (Shimko, 2004, on US Desert Storm discourse); they may undermine long-existing political constellations (Baranov and Zinken, 2003, on ‘Perestrojka’ and ‘Wende’); they can prepare for new political structures (Chilton and Ilyin, 1993; Schäffner, 1993, on European integration). Metaphors can at times activate prejudice or act as ‘sleeping poison’ (Musolff, 2007, on anti-Semitic imagery). The great questions, what metaphors and their possible effects consist of, and what role is played by emotion, have to be addressed first in order to understand how and when metaphors can become powerful political devices. 1.2.1 Metaphors We will detail metaphor only as far as is necessary for understanding the political-semantic approach to metaphor (for an account of metaphor concepts and research, see Beer and De Landtsheer, 2004). First, one can notice that there seems to exist a common-sense, unproblematic meaning of ‘metaphor’ that is based on its etymological meaning. Metaphor comes from Greek meta and pherein, and it means ‘to transfer’ or ‘to carry beyond’ (Hellsten, 2002; Luke, 2004). A metaphor transfers a word or word group (denominated as the vehicle, the source, or the focus) into a context (denominated as the tenor, the target
62 Christ’l De Landtsheer
or the frame) that is not the word’s or the word group’s natural or common context (Beer and De Landtsheer, 2004; Richards, 1936). Aristotle defines metaphors as applications of alien names. The metaphorical nature of words or word groups depends upon their strangeness to the context, their surprisingness or even the conflict between metaphor content and metaphor context (Koeller, 1975; Mooy, 1976). According to the political-semantic approach to metaphor, the conflict nature of metaphor is crucial for its political meaning and effects. Second, recent metaphor literature suggests that one has to distinguish between two types of metaphor research. Both types of research have their own views on which metaphors should be studied and which effects should be expected (Steen, 2002). Linguistic research is the first type and its approach to metaphor resembles the above described common-sense unproblematic view of metaphor. Linguistic research examines the effects of all metaphors that can be found in all kinds of everyday language (see, for example, Gibbs, Leggitt and Turner, 2002). Conceptual Metaphor Theory (for example, Lakoff and Johnson, 1980; Kövecses, 2000) provides us with a second type of research. Researchers generally pick out a few so-called conceptual metaphors and examine how they are being reproduced in various forms. These metaphors would guide our thinking and act as important framing devices. Third, there is a combination approach possible between the linguistic and the Cognitive Linguistic view. According to the political-semantic approach to metaphor, all metaphors used in everyday language are of interest because of their ‘merely rhetorical’ effects. Metaphors from various source domains are assumed to direct thinking and perception, and they frame reality according to these source domains (Beer and De Landtsheer, 2004; De Landtsheer and Vertessen, 2006). 1.2.2 Metaphor effects What effects does metaphor use have, what turns metaphors into effective persuasive tools, and why are metaphors important in politics? The introduction of an alien word can have a number of effects, and thinkers and researchers of metaphor have stressed various of these. Aristotle distinguished between direct and desired effects that result from the replacement of words by unexpected words. Orators who possess this precious gift – Aristotle considered it to be an innate talent to pick at all times the most beneficial words – are, for example, able to hide particular perspectives while stressing others, or to obtain the audience’s attention by simplifying complex problems. Nobody seems to deny that metaphors ease persuasion and that the rhetorical effects
Collecting Political Meaning
63
can be very powerful. However, some researchers stress indirectness and unconsciousness of metaphor effects. By representing things by other things, metaphors can penetrate the mind more deeply than other, more overt, rhetorical effects. Metaphors create new meaning, rather than simply transporting meaning, and by doing so metaphors can obfuscate and potentially change thinking (Beer and De Landtsheer, 2004; Hellsten, 2002; Sopory and Dillard, 2002; Grey, 2000). A crucial insight for studying political metaphor is that words acquire meaning independently of their mere content. A speech with words used only in a literal way would sound dry or monotone and lack emotive appeal (Montero, 2005; Gibbs et al., 2002). Tropes like metaphor and metonymy (figures of speech) are, other than straight logical argument, used by politicians to construct a persuasive version of reality in which they adjust style and content to each other (Cooper, cited in Sutherland, 2005: 198; Hastings, 1998: 209). 1.2.3
Emotion and politics
A driving concept underlying and causing both rhetorical and cognitive effects of metaphor is emotion. Politicians turn to metaphors because they arouse emotions that guide what we think of and how we think of it. The experiment by Gibbs et al. (2002) showed metaphorical language to be more emotive than literal language. These vivid image words (Emrich et al., 2001) create a sense of intimacy centred on the sharing of emotions. Political leaders and journalists can use the potential of metaphor in one or more directions, to enhance the perceived greatness of the speaker, to reassure the audience, or in contrast, to increase anxiety or raise anger. 1.3 Metaphors as political devices For fascist ideologies, anxiety and emotive myth play an important role in influencing people’s minds. Fascism or extreme right-wing politics can only be successful if social stress levels in society are elevated. Hitler’s medical language is a clear example of how thought can be invaded and even corrupted by the deliberate use of emotive language. By describing threats to the Nazi state as dangerous diseases the fascist national socialist ideology prepared the German mind for the mass killings of the Jewish people. Musolff (2007) argues that Hitler in his medical metaphors as early as 1924–5 in Mein Kampf deliberately put forward the illness–cure scenario as a blueprint for genocide. The images of disease arouse anxiety and suggest necessary acceptance of the therapy proposals (Gaus, 2004; Beer and De Landtsheer, 2004; De Landtsheer, 1987; Edelman, 1977).
64
Christ’l De Landtsheer
The use of strong emotional and metaphorical language is, however, not limited to fascism. Metaphors and emotion are used as important persuasive tools in the democratic political struggle by all political groups. The language of democracy and that used by democratic politicians or journalists can be expected to use metaphor’s emotive potential to reassure rather than to raise anger or anxiety. Windt and Ingold (1987) conclude that political rhetoric in democracies occupies the middle ground between dogma and demagoguery. They distinguish between demagogues seeking to please with their impressive language that lacks content and dogmatic leaders enclosing themselves in ideological-expressive language. Exceptional circumstances such as war and severe political or economic crisis can affect the language of democracy and of moderate politicians. Taran (2000: 139, 140) compares ‘centre-politicians’, who in their factual analyses of the current situation prefer logical thinking, with ‘marginal’ leaders, who use more metaphors and mythical thinking. Political extremists, being the least satisfied with the political situation, try to destroy the semantic construction of the political discourse. Metaphor may as a consequence, appear the perfect tool for extremists from the left and the right to unsettle the existing order. Summarizing the above, metaphors can arouse a broad spectrum of emotions, ranging from negative to positive and from superficial to explicitly moving. Metaphor analysis can reveal variation due to the different roles emotion plays in various ideologies or in different circumstances. Because metaphor is the ‘literal language of the soul’ (Woodman, 1997), metaphors play an important role in the framing of the outside world or in creating utopian images of how the world should be. Metaphors are powerful political style elements that have in common with ideology the subjective and often polemic view of reality (Gay, 1992). The following Section details a method for investigating the emotive power of discourse through the use of metaphor in political language. Finally, Section 3 will describe some of the significant differences in political metaphor use that could be established with the help of this method.
2.
Metaphor power analysis
The metaphor power (MP) method introduced in this chapter consists of a quantitative metaphor content analysis that I describe as a political-semantic metaphor analysis (De Landtsheer, 1994). It is assumed
Collecting Political Meaning
65
that metaphor power analysis can establish political meaning for various forms of political discourse (e.g. political speeches, news media interviews and articles, television news). Metaphor power indices are calculated for large samples of political discourse and mass-media language. The method which was originally developed to gauge levels of anxiety or social stress in society is among the few quantitative (for example, Herbeck, 2004) or mixed approaches (for example, Zinken, 2004) in a qualitative research area (Ivy, 2004; Shimko, 2004; Chilton and Ilyin, 1993; Schäffner, 1993). The metaphor power (MP) index is calculated by multiplying the scores on three variables, metaphor frequency (MF), metaphor intensity (MI), and metaphor content (MC). The MP index reveals the metaphor style character of one or more pieces of political discourse. 2.1 Metaphor frequency The frequency of metaphor use (MF) provides us with cues about the metaphor power of discourse. Emotional elements pervade the public sphere and ideologies strive for emotional effects, as politicians and journalists use more metaphors. Therefore, just counting metaphors is a useful technique (Zinken, 2004; Herbeck, 2004). The MF should be calculated relative to the total amount of language or speech of the sample. Therefore, we divide the total number of metaphors (f) by the total number of words in the discourse sample (tw). As f-values then are very small and difficult to understand, we multiply this result by 100. The following formula states that MF can be understood as the number of metaphors per 100 words of discourse, or, when examining spoken language, as the number of metaphors per minute of speech (ts = total speech time in seconds): MF = 100f/tw or MF = 60f/ts 2.2 Metaphor intensity The second variable is metaphor intensity (MI) or the originality score of the metaphors. Metaphor theory states that innovative, creative, and original metaphors are more intense than dormant or dead metaphors (Tsoukas, 1991). If the reference strength to the first-order meaning of the metaphor , i.e. the literal meaning of the metaphorical source, is still strong, then we call it a strong metaphor. In contrast, if the literal meaning does not come to mind, metaphors are weak (Black, 1962; Mooy, 1976). In the metaphor power model, we calculate the metaphor intensity
66
Christ’l De Landtsheer
(MI) value of a sample of metaphorical language. High MI-values denote creative metaphor use. We give metaphors intensity scores with values ranging from (1) for weak (w) metaphors, (2) for moderate or normal (n), and (3) for creative, strong (s) metaphors. The following formula indicates how ‘MI’ stands for the sum of the weighted values of metaphor intensity divided by the total number of metaphors (T): MI = (1w + 2n + 3s)/T Levels of metaphor intensity can be calculated as follows Weak: e.g. decisions that ‘follow the wind’. Moderate: e.g. problems ‘put in the refrigerator’. Strong: e.g. 1991 Iraq War ‘ Desert Storm’. 2.3 Metaphor content The third variable introduces the content power of the metaphor (MC). Different semantic fields, identified by different lexical fields (Kittay, 1987), have different metaphor power. The British economy being the metaphor frame, several meanings can be deduced depending upon the metaphor focus content (being ‘in pain’ or ‘flourishing’) , having been ‘married’ for years to perceptions of the state of the American economy or the relative value of the American dollar. Different semantic fields or sources from which meaning is derived such as illness, nature, or family can be grouped into content categories that have less or more metaphor power (an exploration of metaphorical sources for political targets can be found in Beer and De Landtsheer, 2004: 19–21). Content categories are awarded scores on a scale ranging from 1 to 6, based on insights from psycholinguistic and social-psychological theories. This scale has been validated empirically in a number of cases that focus on political and mass-media discourse – Section 3 presents some results. Metaphors from the higher categories offer politicians the ability to evoke strong emotions and to be more manipulative. Lower-category metaphors are socially important, but they tell us a different, less distressing, story. The metaphor content dimension (MC) will be illustrated in this overview with metaphor examples from an actual Belgian case (De Landtsheer and Vertessen, 2005). The values assigned 1–6 represent lower to higher power, weaker to stronger ratings: 1. The first content category consists of metaphors that use images from the semantic field of ‘everyday-life reality’. The category includes family metaphors and popular sayings, as well as images of objects,
Collecting Political Meaning
67
actions and persons that appear in everyday-life reality. Popular metaphors (p) serve the basic function of metaphor of making the abstract tangible and comprehensible to a large audience (see also space, motion, path and object metaphors, distinguished by Zinken, 2004). Their connection with the real world being very tight, they come close to ‘rational’ or literal discourse, they are far from throwing the audience into confusion, and they do not make escape from reality possible. Popular metaphors are reassuring, they do not challenge the order of society and they offer citizens the perspective of being able to exert influence on the outside world (see, for example, Zinken, 2004 on active subject and passive subject metaphors). Alienation (see Aristotle’s definition) being at the core of the metaphor process, these metaphors are given the lowest scores (1). The president of the Belgian-Flemish extreme right wing party [Vlaams Blok/Vlaams Belang] Frank Vanhecke, uses a popular metaphor when he complains that ‘the problems between the Belgian regions were put again into the refrigerator’ (Belgian-Flemish commercial television news VTM, January 11, 2003). 2. Nature metaphors (n) belong to the second category ( score 2) because they suggest conformation and natural order, even though they always contain the possibility of change. These metaphors fit the qualification by Zinken (2004) of ‘profiling the idea of naturalness of situations and events in political life’. The Belgian-Flemish Christian democracy politician Jean-Luc Dehaene uses a nature metaphor when accusing the president of the Flemish socialist party Steve Stevaert of ‘always following the wind’ (Belgian-Flemish commercial television news, VTM, January 8, 2003). The nature metaphor category contains rational elements and the suggestion of citizen’s control over the environment, but it can also express the idea of lack of control (see, for example, Zinken, 2004, on the fauna metaphor model, with domestic animals versus wild animals, and the comparable plant, river, and body of water models). 3. Political, intellectual and technological metaphors (po) are often sophisticated constructs that are well suited for simplifying complex political processes. These metaphors enable politicians to provide a perspective for framing multi-dimensional processes. They do not produce the same amount of emotive power as the higher categories. The ‘rational’ architecture metaphors Zinken (2004), including the European house metaphor (Chilton and Ilyin, 1993; Schäffner, 1993; Musolff, 2000) clearly encourage political negotiation and changeability of society, but mechanism and machine metaphors sometimes
68
Christ’l De Landtsheer
suggest a lack of control. This ambiguity justifies that political, intellectual and technological metaphors (po) receive an intermediate score (3). they The popular Belgian-Flemish leader of the extreme right-wing party [Vlaams Blok] Filip De Winter uses such a refined metaphor in his remark that in the city of Antwerp ‘according to fair ecological practices, Green politicians have been recycled, similar to the garbage’ (public television news, VRT, May 10, 2003). 4. Disaster and violence metaphors (d) are much less neutral than the metaphor categories mentioned above. They have in common the expression of despair, depression or aggression. The images employed naturally refer to anxiety and other negative emotions, but the citizen may still be active or in control (see Zinken, 2004 on war metaphors). Their ominous character implies a stronger emotional involvement (score 4). The Belgian-Flemish Christian democracy politician Eric Van Rompuy uses a war metaphor, saying about a colleague that ‘this is a capitulation compared to everything he said before’ (public television news, VRT, January 7, 2003). 5. Sports, games and drama metaphors (sp) that appeal to many people and are manipulative receive the value (5). ‘Winning’ and ‘losing’ can, for example, be very emotional activities. These metaphors get their power from the popularity and the supposed harmlessness of the activities referred to by the images. Johan Vande Lanotte, a socialist (SP.A) member of the Belgian government, uses a sports metaphor when admitting ‘I am offensive, now and then one has to attack in order to score a goal, but it should go fast’ (commercial television news, VTM, May 7, 2003). What makes this category effective is the unreal world picture of public life (see Zinken 2004 , for example, about metaphors of sport, game, theatre, drama and religion). 6. The category that is attributed the highest metaphor power (6) is the category of body, disease, medical and death (m) metaphors. According to Gregg (2004), metaphors have more emotive power if they are closely related to the body, because ‘embodied meaning’ plays a central role in our cognitive processes. Musolff (2007) explains in his study of Hitler’s anti-Semitic imagery the functioning of body and disease metaphors playing a central role in Western philosophical traditions – and in Nazi ideology. The president of the BelgianFlemish liberal party VLD, Karel De Gucht, uses a body metaphor in his analysis that ‘one gets of course a strong reaction when one rams down the throat of the majority of the Flemish citizens something that they disapprove of ‘ [the right of immigrants to vote](Flemish commercial television news, VTM, December, 2003).
Collecting Political Meaning
69
Combining the content categories constitutes the metaphor content (MC) variable. Higher MC-values denote stronger emotional appeals. The following formula brings together all the content categories and their values, which results in the metaphor content power score: MC = (1p + 2n + 3po + 4d + 5sp + 6m)/T 2.4 Metaphor style How can we define the metaphor style of a speech, an article, a television interview, or a Web log? When can we conclude that someone is using ‘metaphor style’? The preceding paragraphs showed how to describe the metaphor style of a given text. It is also possible to express the metaphorical strength of it by calculating the metaphor power (MP) index. The MP index is a multiplication of the metaphor frequency (MF), the metaphor intensity (MI) and the metaphor content (MC) indexes: MP = MF x MI x MC: Political speech with low MP indexes focuses on the political (ideological) or on the everyday-life material (economic) content rather then on persuasive style. ‘Content style’ political actors aim at activating citizens in their role of democratic performers. Political speech with high MP indexes is increasingly distancing itself from the existing political (and material) order. In this type of speech metaphor style is extremely important. Political actors using ‘metaphor style’ try to mobilize citizens by turning them into spectators of politics, sometimes through anti-politics and the instigation of fear. The discussion in Section 3 of cases applying the MP method will show where to expect ‘content style” and ‘ metaphor style’, and when political actors may switch their style.
3
Conclusions from the cases
This Section presents an overview of the most interesting conclusions from the political semantic case studies and applications of the metaphor model. All of the studies underline the importance of style in politics and the central role of metaphor in the management of emotion on which the persuasive process relies. Based on insights from various disciplines, we can conclude that metaphors as style elements can change meaning and can influence thought. Metaphors may indeed, according to the literature (e.g., Emrich et al., 2001; Shimko, 2004; De Landtsheer, 1994), contribute to the level of emotionality in public debate, to direct attention towards ambitious leaders, or promote (deviating) opinions and worldviews.
70 Christ’l De Landtsheer
A number of other conclusions follow from empirical research with the metaphor power model in the political domain. The case studies challenge the view that in political rhetoric content dominates over style. It is shown how a metaphor (content) analysis can reveal the crucial but less accessible aspects of political and social reality, including what political groups stand for. Scholars may argue that the left or the right, socialists or conservatives, racists or anti-racists, will not speak or write very differently from each other (Van Dijk, 2004). However, the opinion that content is a more accurate predictor for ideology than style is not supported by our empirical results. No changes were found in the militant traditional right-wing extremist style to complement a number of thematic changes in the party press releases of Vlaams Blok, the Belgian-Flemish party that chose a softer image and a new name after its condemnation by the courts for racism in 2004 (De Landtsheer and Vertessen, 2006). The conclusion that there is continued extremism from our 2000–2005 analysis was supported by later events. Case studies using the political-semantic approach to metaphor and the metaphor power model show a lot of diversity. The research firstly illustrates the important role of exceptional conditions (political or military crises) that may affect political metaphor style. The effects of a number of (more) stable factors such as gender or the medium cannot be neglected either. Secondly, the research is based both upon short time samples from political campaigns (Belgian 2003 and Ukrainian 2002 elections; Euro introduction 2002) and political and military crises (Srebrenica, 1995) and upon trend studies from parliamentary rhetoric (EU 1981–1993) and long-term mass-media research (Belgium, 1831–1981). The range of cases is, thirdly, varied in such respects as languages included (e.g. English, Spanish, Dutch), countries (e.g. Belgium, Ukraine, Britain), ideological groups or political parties (e.g. political factions in the EU Parliament), formats (political speeches and interviews, news articles and comments), media technology (television, radio, newspapers), media audience (tabloid versus broadsheet newspapers), and media organizations (commercial broadcasting or public broadcasting). 3.1 Metaphors of prosperity and crisis This trend study testifies to the enormous impact of emotion in political rhetoric and opinion formation during severe economic crises. Results are based upon samples of mass-media language (439,582 words in total) that cover a hundred and fifty years of political discourse in the Belgian-Flemish community. A case study on ‘the language of prosperity and crisis’ (De Landtsheer, 1987, 1994) indicates a statistically
Collecting Political Meaning
71
significant positive correlation between indicators of economic crisis (e.g. high unemployment rates) and metaphor power in (political) media language. This general trend could be distinguished irrespective of the political preference or the ‘quality’ level of the medium. 3.2 Political metaphor in a unified Europe This study on ‘the rhetoric of a Unified Europe’ concerns the 1981–1993 metaphorical rhetoric of the European Parliament (De Landtsheer, 1998b). Debates were mapped out according to the delegates’ national, political, gender, and socio-economic programmes. The 621,012-word sample suggests that in the European Parliament, political orientations flourish – more so than national feelings and remarkably more so than linguistic awareness. The ideological orientation, one variable among the set of variables to influence the discourse, turned out to be a very prominent indicator of metaphor power in discourse. While right-wing rhetoric was more metaphorical than left-wing rhetoric, overall the extremes proved to be more metaphorical than the centre. Metaphor power in parliamentary discourse by female delegates was slightly lower overall compared to male discourse. Nationality and language, however, did not significantly affect rhetorical metaphor power. Parliamentarians’ speeches further empathized with the economic sorrows of European citizens: metaphor power in European parliamentary rhetoric as a whole positively correlated with the EU unemployment rates. This case study accordingly supports the findings from the metaphors of prosperity and crisis case. A final conclusion from the European rhetoric case study could be that the initial economic unification of the Union’s members seemed to have definitely gone beyond economic considerations and entered the political domain. Among the set of variables to influence the European discourse, ideology turned out to be the most prominent one. 3.3 Political colour of metaphor As different ideologies have different goals, they also have different strategies. The conclusions from the European Parliament case study that political groups from various ideological perspectives may differ in their rhetorical style, and that political rhetoric by extremists is more metaphorical compared to others’ political rhetoric were confirmed in two other studies. A study on Ukrainian parliamentary rhetoric (Taran, 2000) suggested a strong relationship between the use of metaphor and mythical thinking. In another Belgian study (De Landtsheer and Vertessen, 2005), data were collected from three samples of written press
72 Christ’l De Landtsheer
interviews between 2003 and 2004 with politicians (845,199 words), broadsheet papers (73,316 words), and television news interviews (255 minutes of politicians’ talk). All studies made clear that parties and ideologies make different use of metaphors. Particularly the parties at the edges of the political spectrum, represented in this case by the ecologists and by nationalist extremists, are distinctive in using more and stronger (disaster, sports and medical) metaphors. The rhetoric by the extreme-right party Vlaams Belang and by the right-wing nationalist party N-VA (which has some historical connections with the Vlaams Belang) was found to have higher metaphor power than the rhetoric by all other parties. The rhetoric of the most left-wing parties, the ecologist Groen! and the leftist-nationalist Spirit, have higher metaphor power than the parties at the political centre. However, a lot of work still has to be done. For example, we were not able to clarify the relation between ideology and metaphor use in the political centre, where the right and the left meet each other. 3.4 Political crisis language In a Dutch case study, changes in Dutch elites’ metaphor use during the Srebrenica crisis were investigated (De Landtsheer and De Vrij, 2004). In this study, the metaphor power model was tested in combination with other variables derived from existing concepts on integrative complexity from social psychology (e.g. Suedfeld and Bluck, 1988) and pragmatically ambiguous use of modals from cognitive linguistics (e.g. Dancygier and Sweetser, 2005). Combining an analysis of metaphor use with analyses of other style elements would, as we thought, strengthen the findings. The study suggested that metaphor power in politicians’ and mass media discourse is elevated during severe political crises related to military interventions. Similarly, the highly metaphorical crisis discourse was found to be less complex and it showed the empathizing use of modals (more epistemic and speech act use than content use). This crisis communication combination or CCC theory was also employed when studying the rhetoric of one radical leader. An archival study of the rhetoric of Dyab Abou Jahjah, the leader of the Arab European League (AEL) before and after his brief imprisonment during the November 2002 riots in Antwerp, suggests that the crisis situation turned the radical into an extremist politician. These studies support the CCC pattern for persuasive speech and its subsequent cognitive and attitude effects on the audience (De Landtsheer, 2007). In 2007 Jahjah and his companion were condemned to one year’s imprisonment for their provocation in the riots.
Collecting Political Meaning
73
3.5 Metaphorical election style Persuasive intensity during election campaigns resembles the intensity of political crisis discourse. This relation was revealed in a metaphor power study of ‘the Orange Revolution’ and presidential campaign rhetoric of the Ukrainian media (Krasnoboka and De Landtsheer, 2007). One Belgian study on ‘metaphorical election style’ (Vertessen and De Landtsheer, 2007) also suggests that election rhetoric in newspaper reporting and television news is more metaphorical compared to political rhetoric outside election periods. Furthermore, an international comparative study of news reporting on the introduction of the European single currency (Euro) suggests that one may consider metaphor power in political discourse as an indicator of persuasive efforts. Heavy public information campaigns are likely to have similar but more moderate indirect effects. Metaphor power in newspaper discourse on the introduction of the Euro was found to be higher in countries that intended to introduce the Euro than in countries without that intention. Media and citizens in the Euro-zone were targeted preceding the introduction of the Euro with large-scale information campaigns (De Landtsheer and Koch, 2005). 3.6 Metaphorical medium News reporting on television is more metaphorical than news in written accounts. Popular newspapers and tabloids are using more and stronger metaphors than are broadsheet newspapers. Commercial broadcasting is, except for election periods, more metaphorical than public broadcasting. These are the conclusions of a Belgian study on ‘impression management in television news’ (De Landtsheer et al., 2005). The study underlines the role that public broadcasting is supposed to play in persuading voters to participate at election times. Making use of the findings from these studies, we can perhaps identify the factors which stimulate more metaphor use: Context-factor
Stimulated to more metaphor use
Political crisis Military crisis Economic crisis Political party Gender Media technology Media audience Media organizer
media, politicians media, politicians media, politicians right wing, extremist male television popular, broadsheet commercial
74 Christ’l De Landtsheer
Elections Political crisis Military crisis Economic crisis 3.7
media, politicians media, politicians media, politicians media, politicians
Discussion
Political semantic research aims at anticipating political developments and their social implications. The prognostic value and social relevance of this research may be obvious, less evident as yet are the inquiry methods. The metaphor power method was empirically validated for the case studies described above as well as for other cases. Coders were provided with very detailed coding instructions. In some cases the data collection was performed by one to three expert coders; in other cases the coding relied upon 20 to 80 second-year undergraduates, or even 400 first-year undergraduate students. For the largest sample, intercoder reliability was rather low (Pearson r² between coders coding the same interview was 0.452 on coding metaphor content (MC) and 0.185 on coding metaphor intensity (MI)). We believe that the data are valuable, though we should at all times be careful in generalizing. More work is necessary to better validate the method and in order to explain in a profound way the interesting trends that we found in the data. It is obvious that in times of peace in democratic societies the everyday struggle between political opponents will always require the use of metaphor and other persuasive style components. One can observe special conditions in which persuasive efforts by politicians or the need to make contact with the audience are more intense. In conditions that according to Windt and Ingold (1987) stimulate the use of demagogic or audience-oriented impressive language, metaphor and emotion seem to play a special role. When the economy crashes in a democratic society or there is severe political transition, or the threat of a military intervention, the language used by politicians and political journalists (political elites) will gain in metaphor power. Election campaigns and intensive campaigning for public projects also show intensive use of metaphor, if to a lesser degree. The features of an extremely persuasive or crisisspecific style may finally appear to depend at all times upon personal or gender, party, or group characteristics as well as upon media or particular context factors. Female politicians use less metaphor power in their political rhetoric than males do. It is clear that radical (extremist) groups or parties and their leaders make special efforts to persuade. Popular and commercial media, including television, with their ‘metaphorical sound-bites’ use the same (old) propaganda technique.
Collecting Political Meaning
75
References Aristotle (2004). Rhetoric, trans. by Roberts, W. Rhys. (Mineola, NY: Dover Publications). Baranov, A. and Zinken, J. (2003). ‘Die metaphorische Struktur des öffentlichen Diskurses in Russland und Deutschland: Perestrojka- und Wende-Periode’. In B. Symanzik et al. (eds.), Metapher, Bild und Figur: Osteuropäische Sprach- und Symbolwelten (pp.93–121) (Hamburg: Verlag Dr. Kovac). Beer, F.A. and De Landtsheer, C. (2004). ‘Metaphors, Politics and World Politics’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp.5–52) (East Lansing: Michigan State University Press). Belgian-Flemish commercial television news VTM (2003). 8 and 11 January, 7 May, 7 December. Belgian-Flemish public television news, VRT (2003). 7 January, 10 May. Billig, M. (2003). ‘Political rhetoric’. In D.O. Sears, L. Huddy and R. Jervis (eds.), Oxford Handbook of Political Psychology (pp. 222–50) (New York: Oxford University Press). Black, M. (1962). Models and Metaphors (Ithaca, NY: Cornell University Press). Chilton, P.and Ilyin, M.V. (1993). ‘Metaphor in political discourse: the case of the ‘Common European House’. Discourse and Society, 4, 1: 7–31. Dancygier, B. and Sweetser, E. (2005). Mental Spaces in Grammar (Cambridge: Cambridge University Press). De Landtsheer, C. (1987). De Politieke Taal in de Vlaamse Media (Antwerp: Kluwer). De Landtsheer, C. (1994). ‘ The language of prosperity and crisis, a case study in political semantics’. Politics and the Individual, 4, 2: 63–85. De Landtsheer, C. (1998a). ‘Introduction to the study of political discourse’. In O. Feldman and C. De Landtsheer (eds.), Politically Speaking, a Worldwide Examination of Language Use in the Public Sphere (pp.1–16) (Westport, CT: Praeger). De Landtsheer, C. (1998b). ‘The political rhetoric of a unified Europe’. In O.Feldman and C. De Landtsheer ( eds.), Politically Speaking, a Worldwide Examination of Language Use in the Public Sphere (pp.129–45) (Westport, CT: Praeger). De Landtsheer, C. (2007). ‘Crisis style or radical rhetoric? The speech by Dyab Abou Jahjah, leader of the Arab European League (AEL)’. In C. Hart and D. Lukes (eds.), Cognitive Linguistics and Critical Discourse Studies (Cambridge: Cambridge Scholars Press,). De Landtsheer C., De Vries, Ph. and Vertessen, D. (2005). ‘Politiek impressiemanagement in het televisiejournaal’. In M. Hooghe, K. Deswert and S. Walgrave (eds.), Nieuws op Televisie. Televisiejournaals als Venster op de Wereld (pp.99–118) (Leuven: Acco). De Landtsheer, C., De Vrij, I. (2004).’Talking about Srebrenica, Dutch elites and Dutchbat. How metaphors change during crisis’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics. (pp. 163–89) (East Lansing: Michigan State University Press). De Landtsheer, C. and Koch, E. (2005). ‘Metaphors and the framing of the European single currency (Euro) in and out of “Euro land” ’. In R.F. Farnen et al. (eds.), Democratization, Europeanization, and Globalization Trends. Cross-National
76 Christ’l De Landtsheer Analysis of Authoritarianism, Socialisation, Communications, Youth, and Social Policy (pp. 429–46) (Frankfurt am Main: Peter Lang). De Landtsheer, C. and Vertessen, D. (2005). ‘Political colour of metaphor. A Belgian case study’. Paper given at the Annual Meeting of the International Society for Political Psychology (Toronto, 3–7 July 2005). De Landtsheer, C. and Vertessen, D. (2006). ‘Political colour of metaphor with focus on black’. Paper given at the Annual Meeting of the International Society for Political Psychology (Barcelona, 12–15 July 2006). De Sola Pool, I. (1956). ‘Variety and repetition in political language’. In H. Eulau, S. Eldersfeld and M. Janowitz (eds.), Political Behaviour, A Reader in Theory and Research (pp.217–31) (Glencoe, IL: The Free Press). Edelman, M. (1977). Political Language (New York: Academic Press). Emrich, C.G., Brower, H.H., Feldman, J.M. and Garland, H. (2001). ‘Images in words, presidential rhetoric, charisma and greatness’, Administrative Science Quarterly, 46,3: 527–57. Gaus, H. (2004). Opgang en Verval van Extreem-Rechts (Ghent: Academia Press). Gay, W.C. (1992). ‘Ricoeur on metaphor and ideology, Darshana International, 32,1: 59–70. Gibbs, R. W. Jr., Leggitt, J. S. and Turner, E. A. (2002). ‘What’s special about figurative language in emotional communication?’ In S. R. Fussell (ed.), The Verbal Communication of Emotions (pp. 125–50) (Mahwah, NJ: Lawrence Erlbaum Associates). Gregg, R.B. (2004). ‘Embodied meaning in American public discourse during the Cold War’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp. 59–73) (East Lansing: Michigan State University Press). Grey, W. (2000). ‘Metaphor and meaning’, Minerva, An Internet Journal of Philosophy, 4 , available online on http,//www.ul.ie/~philos/vol4/index.html. Hastings, A. (1998). ‘Connecting linguistic structures and social practices, a discursive approach to social policy analysis’, Journal of Social Policy, 27, 2: 191–211. Helmig, J. (2005). ‘Metaphors as agents of signification. Towards a discursive analysis of metaphors’ . Paper presented at the ECPR workshop on metaphors in political science, (Granada, Spain, 14–19 April 2005). Hellsten, I. (2002). The Politics of Metaphor. Biotechnology and Biodiversity in the Media (Tampere: Tampere University Press). Herbeck, D.A. (2004). ‘Sports metaphors and public policy: the football theme in Desert Storm discourse’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp. 121–39) (East Lansing: Michigan State University Press). Hovland C., Janis I. and Kelly H. (1953). Communication and Persuasion. (New Haven, CT: Yale University Press). Ivy, R.L. (2004). ‘Democracy, war, and decivilizing metaphors of American insecurity’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp.75–90) (East Lansing: Michigan State University Press). Kittay, E.F. (1987). Metaphor, Its Cognitive Force and Linguistic Structure (Oxford: Clarendon Press). Koeller, W. (1975). Semiotik und Metapher. Untersuchungen zur grammatischen Struktur und kommunikativen Funktion von Metaphern (Stuttgart: Metzler). Korzybsky, A. (1933). Science and Sanity, An Introduction to Non-Aristotelian Systems and General Semantics (Lancaster, PA: International Non-Aristotelian Library).
Collecting Political Meaning
77
Kövecses, Z. (2000). Metaphor and Emotion: Language, Culture, and Human Feeling (Cambridge: Cambridge University Press). Krasnoboka, N. and De Landtsheer, C. (in press). ‘Broadcasting the Orange revolution. Rhetoric of the Ukrainian media during the last presidential campaign’. In R.F. Farnen, H. Dekker, C. De Landtsheer, H. Sünker and D.B. German (eds.), Political Culture, Socialization and Education, Interdisciplinary and Cross-National Perspectives for a New Century (Frankfurt am Main: Peter Lang). Lakoff, G. and Johnson, M. (1980). Metaphors We Live By (Chicago: University of Chicago Press). Lasswell, H.D. (1952). The Comparative Study of Symbols. An Introduction (Stanford, The Hoover Institute on War, Revolution and Peace: Stanford University Press). Lasswell, H.D. (1949). ‘Style in the language of politics’. In H.D Lasswell, N. Leites et al. (eds.), Language of Politics. Studies in Quantitative Semantics (pp. 20–40) (New York: George W. Stewart Publisher Inc.). Lauerbach, G. (1999). ‘From macro to micro and back, framing, footing and genre in recent TV election night coverages’. In J. Verschueren (ed.), Language and Ideology. Selected Papers from the 6th International Pragmatics Conference (pp. 317–43) (Antwerp: International Pragmatics Association). Luke A. (1995). ‘Text and discourse analysis in education. An introduction to critical discourse analysis’, Review of Research on Education, 21: 1–48. Luke, T.W. (2004). ‘Megametaphorics. rereading globalization and virtualization as rhetorics of world politics’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp. 237–58) (East Lansing: Michigan State University Press). Montero, L. (2005). ‘Language, discourse and politics’. Paper presented at the 29th Annual Scientific Meeting of the International Society of Political Psychology (Toronto, Canada, 3–6 July ). Mooy, J. J. (1976). A Study of Metaphor. On the Nature of Metaphorical Expressions with Special Reference to Their Reference (Amsterdam: North Holland Publishing Company). Musolff, A. (2000). ‘Political imagery of Europe: A house without exit doors?’ Journal of Multilingual and Multicultural Development, 21, 3: 216–29. Musolff, A. (2007). ‘What role do metaphors play in radical prejudice? The function of anti-Semitic imagery in Hitler’s Mein Kampf’, Patterns of Prejudice, 41: 21–43. Orwell George (1949). 1984 (London: Signet Books). Richards, I.A. (1936). A Philosophy of Rhetoric (Oxford, Oxford University Press). Schäffner, C. (1993). ‘Die Europaïsche Architektur- Metaphern der Einigung Europas in der deutschen, britischen und amerikanischen Presse’. In A. Grewenig (ed.) Inszenierte Kommunikation (pp.13–30) (Opladen: Westdeutscher Verlag). Shimko, K.L. (2004). ‘The Power of metaphors and the metaphors of power: the United States in the Cold War and after’. In F. A. Beer and C. De Landtsheer (eds.), Metaphorical World Politics (pp. 199–215) (East Lansing: Michigan State University Press). Sopory, P. and Dillard, J.P. (2002). ‘The persuasive effects of metaphor, a metaanalysis’, Human Communication Research, 28, 3: 382–419. Steen, G.J. (2002). ‘Metaphor identification, a cognitive approach’, Style, 36, 3: 386–407.
78
Christ’l De Landtsheer
Suedfeld, P. and Bluck, S. (1988). ‘Changes in integrative complexity prior to surprise attacks’, Journal of Conflict Resolution, 32, 4: 626–35. Sutherland, C. (2005) ‘Nation-building through discourse theory’, Nations and Nationalism, 11, 2: 185–202. Sweetser, E. (1995). ‘Metaphor, mythology, and everyday language’, Journal of Pragmatics, 24: 585–93. Taran, S. (2000). ‘Mythical thinking, Aristotelian logic, and metaphors in the Parliament of Ukraine’. In C. De Landtsheer and O. Feldman (eds.), Beyond Public Speech and Symbols. Explorations in the Rhetoric of Politicians and the Media (pp. 120–43)(Westport, CT: Praeger). Tsoukas, H. (1991). ‘The missing link, a transformational view of metaphors in organizational science’, Academy of Management Review, 16, 3: 566–85. Van Dijk, T.A. (2004). ‘Politics, ideology and discourse’. In R. Wodak (ed.), Encyclopaedia of Language and Linguistics (2nd edn: section on ‘Language and Politics’ ) (Oxford: Elsevier). Vertessen D. and De Landtsheer C. (in press). ‘A metaphorical election style, metaphor use at election time’. In J. Pikalo and T. Carver (eds.), Politics, Language and Metaphor (London: Routledge). Windt, T. and Ingold, B. (1987). Essays in Presidential Rhetoric (Dubuque, IA: Kendall/Hunt Publishing). Wodak, R. and De Cilia, R. (2006). ‘Politics and language, overview’. In K. Brown (ed.), Encyclopedia of Language and Linguistics (2nd edn: pp. 707–19) (Oxford: Elsevier). Woodman, R. (1997). ‘Metaphor and the language of Revelation, The Journal of Bahá’í Studies, 8, 1: 1–28. Zinken, J. (2004). ‘Metaphoric practices in the German Wende discourse’, Journal of Multilingual and Multicultural Development, 25. 5–6: 1–13.
6 Metaphor and Context: A Perspective from Artificial Intelligence John Barnden
Introduction In this chapter we argue that metaphorical utterances often contain source-domain elements that are important for the overall meaning of the utterance but that should not themselves be given a parallel in (be mapped into) the target domain, either because it is difficult to do so or because doing so is not important for linking the utterance into the context supplied by the discourse. Non-parallelism goes hand-inhand with metaphor understanding processes that are strongly guided by context. The chapter discusses various ways in which understanding might be guided by context. A particularly strong form of contextguided understanding is context-driven understanding, a form of which is used in an AI approach to metaphor understanding that the author and colleagues have developed.1 On embarking on this chapter I happened to hear the following spoken comment on a TV programme (Newsnight, BBC2, UK, 3 July 2007; plausible punctuation added): (1) I don’t think strings are attached. If there are any they’re made of nylon – I can’t see them. The speaker was an African politician being interviewed about a new investment by China in mineral mining. In the first sentence, he uses stock metaphorical phraseology concerning strings being attached. We take the ‘strings’ to refer to constraints affecting the politician’s country’s future behaviour with regard to the investment or related 79
80
John Barnden
things (cf. the meaning for ‘no strings attached’ given in the Oxford Dictionary of Idioms, 2005: no special restrictions or conditions apply to an opportunity or offer). In the second and third sentences the politician then elaborates on the standard phraseology. We take him there to be saying, in source-domain terms, that he can’t ‘see’ any ‘strings’ because if they exist they’re made of nylon; and thus to be saying in target-domain terms that he is not aware of any constraints, and that if any exist they must be somehow unnoticeable. The passage raises the important issue of the degree of parallelism (structured analogy) that the understander should seek between the source-domain scenario that is painted and the target-domain scenario that is actually being portrayed. It is plausible to say that the strings have a parallel in the target domain, namely the constraints. But should the understander seek something in the target scenario that corresponds to nylon? (Our answer will be ‘No’.) Yet the nylon is clearly crucial to the target-domain meaning of the passage. There is a tendency for detailed processing models of metaphor to focus on constructing source/target parallelism as the way of understanding metaphor. Thus, psychological models of analogy (and hence to some degree, whether implicitly or explicitly, of metaphor) such as Structure-Mapping Theory (SMT) (Gentner, 1983, Gentner and Bowdle, 2001), Analogical Constraint Mapping Engine (ACME) (Holyoak, Novick and Melz, 1994; Holyoak and Thagard, 1989) and Incremental Analogy Machine (IAM) (Keane, 1988) are all models that seek to establish a strong degree of parallelism, by constructing sets of links between source and target from scratch. In AI, the models of Wilks (1978) and Fass (1997) have also had this flavour, and Martin’s (1990) model constructs links as extensions to existing analogies. By contrast, however, some models in AI – those of Narayanan (1999) and ourselves (Barnden, Helmreich et al., 1996; Barnden and Lee, 2001; Barnden 2001, 2006a,b; Barnden, Glasbey et al., 2004) and parts of Hobbs’ approach (Hobbs, 1990) – have reduced the role of parallelism in metaphor understanding. They still sometimes rely on there being some mapping links between source items and target items, but economize on these by relying on possibly-extensive source-domain inference to link non-paralleled items in the source scenario to items that do have a parallel (i.e., are mapped). Also, Conceptual Metaphor Theory (CMT) tends to emphasize parallelism, for instance in discussions of the sets of mapping links involved in a particular metaphor, and in the claim that patterns of inference in the source domain are in some way carried over to the target domain
A Perspective From AI 81
(Lakoff, 1993). When non-parallelism is attended to it is rather in the spirit that some aspects of the source domain cannot be, or as a matter of fact are not, used in the conceptual metaphor in question, the implication being that therefore they do not have a metaphorical effect. Our point will be that non-paralleled items are often crucial to the metaphorical effect. To be sure, not all theories of metaphor stress parallelism. Two main exceptions are Relevance Theory (Sperber and Wilson, 1995) and the categorization model of Glucksberg and Keysar (1990). But neither do these attend explicitly to certain deep issues raised by non-parallelism. While to some extent sketching processes that lead from utterances and other information (e.g. context and world knowledge) to metaphorical meanings, they do not explicitly engage with the issue of the extent to which there may be source elements for which target parallels are not, or should not, be sought, but which are nevertheless crucial to the construction of the metaphorical meaning. On the other hand, one special case of metaphor where non-parallelism has been a salient issue is that of metaphorical idiom. For example, Langlotz (2006) has provided a detailed and extensive treatment of (non-)parallelism in addressing issues of systematicity, transparency and motivatedness of idiom. An example of the issues is the idiom ‘spinning [one’s] wheels’, where the wheels do not themselves have parallels, but the consequent lack of forward physical movement of the car or whatever does have a parallel. The considerations he raises need to become more salient in the field of metaphor as a whole. We now come to a central point of this chapter, namely the connection of non-parallelism to the issue of how context enters into metaphor understanding. The claim is that one reason that non-parallelism has not been adequately attended to is a lack of sufficient attention to the details of how context is to help with understanding. Certainly, the idea that context is important for metaphorical meaning and understanding has often been put forward (see, for example, Gibbs, 1994; Gibbs and Tendahl, 2006; Giora, 1997; Leezenberg, 1995; Stern, 2000), but few detailed processing models have appeared of how context is actually to be used. This is not surprising because of the immense complexity of metaphor on one hand and of the notion of context on the other. Context brings in the huge research questions to do with discourse structure, discourse relations, the nature of semantics in general (Stern, 2000), and so forth. We believe that focussing on context decreases the need to assume, use, or search for parallelism, and is simultaneously friendly to views of metaphor understanding that are based on
82 John Barnden
source-domain inference and that thereby explain how non-paralleled source elements can play a crucial role. This chapter will, first, illustrate further how metaphor often involves extensive non-parallelism. It will then indicate how non-parallelism interacts with the question of how to use context to guide understanding. In the course of doing this we will examine various ways in which the metaphor understanding process might or might not be guided by context, and various directions in which processing might go (for example, entirely forwards from utterances or entirely backwards from context). We will use the word ‘parallelism’ rather than the more technical word ‘isomorphism’, as the latter implies a strict one-to-one correspondence between items in the source scenario and items in the target scenario, whereas we wish to allow for looser forms of parallelism.
1
Non-parallelism
1.1 Nylon strings To return to the nylon string example (1), the issue is the following. A source-domain scenario is described in which there are strings, attachment of them (to the African parties involved in the business investment in question, metaphorically cast as physical objects that are somehow manipulable by strings), the substance nylon, the physical property of being made of a substance, and the ability of the speaker to physically see the strings. Which of these aspects of the source scenario correspond to anything at all in the target scenario?2 Before going on we must note a complication. It could be argued that the interpretation of ‘I can’t see them’ could be separate from the interpretation of the prior parts of the passage, and be based on interpreting ‘them’ directly as referring to something in the target scenario corresponding to the physical strings, rather than to the physical strings themselves. (Actually, this raises large but largely unstudied issues about how metaphor interacts with anaphora generally.) If this is how understanding proceeds, then it is furthermore possible that the understander can appeal directly to a non-physical lexical sense of ‘see’ – e.g. a sense of abstractly being aware of something; in which case that non-physical scenario needs to be entertained in understanding ‘I can’t see them’. We will put these possibilities aside, as they would only force small amendments to our discussion, and in view of the patent coherence between the physical translucency or invisibility of nylon and the physical process of seeing.
A Perspective From AI 83
To simplify the discussion, let us assume that the (possibly absent) strings and the (possibly absent) attachment do have parallels: the strings correspond to possible future-behaviour-affecting constraints that the Chinese investors might have sought to place as a condition of the investment, and the attachment is that (counterfactual) placing and future-behaviour affecting itself. So what remains is the nylon, the being-physically-made-of, and the being-able-to-physically-see. In fact, in the following we break the last item down into being-able and the physical seeing itself. This article’s main claim about the example is that it mainly seeks to convey that: the speaker is not able to determine the existence of any extra constraints affecting the future behaviour of the African side. Here, determining-the-existence-of corresponds to (i.e. is the parallel of) seeing, and the not-being-able associated with the existence- determining in the target scenario corresponds to the not-being-able associated with the seeing in the source scenario. One can also interpret the passage as conveying that: the speaker is allowing for the possibility that there are constraints, but if so then they are of a nature such that it is intrinsically difficult for anyone to determine their existence. That is, it is not just inattention, stupidity, or whatever on the part of the speaker that he can’t ‘see’ them. How is the above difficulty in the target domain arrived at by the understander? Presumably from the difficulty of physically seeing nylon strings. And where does that difficulty come from? The translucency of nylon, combined possibly with the thinness of nylon strings. The crucial carry-over here is that of difficulty. Notice carefully that this does not of itself imply that the translucency needs to be mapped to the target scenario. Even less does it imply that being-made-of-nylon itself needs to be mapped (similarly for being thin ...) and, even less again, that being-made-of and nylon need separately to be mapped. Of course, if one asked the question of why the (possible) constraints in the target scenario are difficult to notice, one might look to the source domain and therefore seek some parallel of translucency. But certainly there was nothing in the actual context of the example as heard that could provide any specific target-domain parallel. The best that we can suggest is that some property or other of the constraints makes them
84 John Barnden
difficult to notice. But you could work that totally vacuous and useless piece of information out without any metaphorical utterance or any information being at hand; and it’s a little like saying that sleeping pills help you sleep ‘because’ they have dormative power. So, we have arrived at the suggestion that a rich understanding of the passage can be obtained on the basis merely of the following parallels: strings — future-behaviour-affecting constraints attachment — the fact of those constraints being applied physical seeing — abstract noticing (coming to determine the existence of) inability — inability difficulty — difficulty possibility, counterfactuality, etc. — possibility, counterfactuality, etc. (respectively) Going back to a point we skated over above, it may be that the difficulty of noticing the constraints is deliberate on the part of the Chinese investors. This could come from a source-scenario inference that, perhaps, the strings are deliberately made to be difficult to see. This inference must be quite uncertain, because it is only in some contexts that a nylon constitution would be aimed at invisibility. So, we may wish to add the further parallel: deliberateness — deliberateness uncertainty — uncertainty We now move to considering other examples of non-parallelism. 1.2 Conversational cat-flaps In Barnden, Glasbey et al. (2004) we extensively analysed the following example:3 (2) I tried not to run down Phil too much – I felt bad enough as it was, what with screwing his girlfriend and all. But it became unavoidable, because when Jackie expressed doubts about him, I had to nurture those doubts as if they were tiny, sickly kittens, until eventually they became sturdy, healthy grievances, with their own cat-flaps which allowed them to wander in and out of our conversation at will. (Hornby, 1995: 12; italics added)
A Perspective From AI 85
This exhibits considerable non-parallelism. We contend that it would be a mistake (a waste of time) to try to find parallels for the following source-scenario elements (and possibly others): the cat-ness of the doubts and grievances their sickliness the cat-flaps the at-will-ness of the wandering. The most obvious case here is the cat-flaps. The point about these is that they support the point (which is also explicitly stated) that the cats can wander in and out at will, this wandering corresponding to the grievances appearing intermittently in the conversation. It would seem pointless to try to work out some specific aspect of the conversation or some other part of the target scenario that corresponds to the cat-flaps. Also, the deliberateness (at-will-ness) of the cats’ wandering has no parallel with any feature of the grievances corresponding to the cats. Rather, the deliberateness of the cats’ wandering suggests that it is not the speaker and his interlocutor who are bringing the cats in and out: that is, the cats’ movements are non-deliberate on the part of the speaker and the interlocutor. This nondeliberateness maps to the non-deliberateness (on the part of those two people) with which the grievances appear in and disappear from the conversation. The main point of the sickliness (and tininess), arguably, is to buttress the metaphorical notion of ‘nurturing’ the kittens. Clearly this nurturing itself does have some parallel in the target scenario. Now, if the kittens are sickly and tiny then nurturing is especially natural and morally good. Perhaps therefore the point of the sickliness and tininess is to portray the speaker’s action in converting doubts to grievances as natural and morally good. More securely, it can be inferred in the source scenario that the kittens were helpless and were likely to die if not nurtured. From this, we might suppose that in the target scenario the doubts were likely to disappear unless the speaker acted. So, the tininess and sickliness certainly lead to source-scenario conclusions that map to the target in some way, but that does not mean that they themselves, or the steps connecting them to those conclusions, map to the target. Finally for the above list, the cat-ness is totally irrelevant other than for appealing to familiar domestic scenarios of nurturing kittens and cats using cat-flaps. So it contributes to the development of a coherent
86
John Barnden
source scenario, but there is no reason at all to suppose that some specific properties of cats (aside from the being-nurtured, the wandering, etc.) are to be mapped to the target. This of course contrasts with cases where someone might say something like ‘John [a person] is being catty’, ‘Sally [a person] is a kitten’, and so forth. So an additional issue is that cat-ness or an aspect of it does sometimes have parallels, whereas it does not in our example above. 1.3
Other examples
An example we have treated in various papers (Barnden, 2001, 2006a; Barnden, Glasbey, et al., 2004; Barnden and Lee, 2001) and implemented in our computer program for a metaphor understanding system (ATT-Meta) is the following: (3) In the far reaches of her mind, Anne knew Kyle [her husband] was having an affair, but ‘to acknowledge the betrayal would mean I’d have to take a stand. I’d never be able to go back to what I was familiar with,’ she says. Not until eight months had passed and she finally checked the phone bill did Anne confront the reality of her husband’s deception. (Cosmopolitan, 216 (3), USA ed., March 1994. Italics added.) Here we claim that the point of the ‘far reaches’ is to generate the sourcescenario conclusion that Anne’s conscious self (metaphorically conceived as a person located in the middle of Anne’s mind-space) has great difficulty in physically operating upon the idea that Kyle was having an affair (this idea being metaphorically cast as a physical object physically located in the far reaches of the mind-space). There is no need to, and it would probably be very difficult to, find some component or aspect of Anne’s mind that corresponds to the far reaches themselves. Again, we could postulate some such component, and say that the affair-thought was in some sense ‘in’ it, but what would be the point? The message of the utterance is the fact that Anne is in a mental state of having difficulty in consciously involving the affair-thought in her thinking. (The article from which the passage is taken is about denying things to oneself.) And this difficulty comes from the difficulty of, in metaphorical terms, physically operating upon the idea. The far reaches are just a tool for conveying this physical difficulty. In Barnden (2006b) we discuss the degree of parallelism in fifteen examples taken from real discourse. We briefly summarize here the conclusion about non-parallelism in one of those examples.
A Perspective From AI 87
This all means that general managers have cricks in their necks from talking down to the Community Health Councils and District Health Authorities, and up to Regions and the Department. (Goatly, 1997:162, from The Daily Telegraph) There is no need to consider or discover parallels for necks, cricks, physical suffering or head-turning: rather, the point of the passage is emotional suffering. Finally, we must note that there is an important distinction that should be made in the study of non-parallelism: the distinction between a concept or property used in the source scenario having a parallel, and a source-scenario individual that comes under that concept or property having a parallel. For instance, in the cat-flaps example, the individual cats do of course have parallels: they are the doubts/grievances. However, the concept of cat, or if you like the property of cat-ness, has no parallel. On the other hand, in the far reaches example, neither the particular far reaches in question nor the concept of far reaches needs to be taken to have a parallel.
2
The link to context
How does the question of (non-)parallelism engage with the issue of contextual aspects of metaphor understanding? We claim, based on the work on our ATT-Meta theoretical approach and implemented computer program, that one crucial strategy in metaphor understanding is to use questions or issues generated by the context of a metaphorical utterance to drive the process of understanding. This contextual-issue-driven strategy has the effect of drawing from a metaphorical utterance, or indeed from an extended passage containing several such utterances, just what is needed to illuminate those issues. This is not only a powerful way of combating the indeterminacy of metaphorical meaning (see for example Stern, 2000) and ensuring relevance, by tending to select only relevant inferences from the source scenario, but also serves to limit the use or discovery of source/target parallels to those really necessary to illuminate the contextual issues. If it turns out that, for example, the far reaches of Anne’s mind need not be given a parallel for that contextual purpose, then no parallel is even sought. We see this as boon in that (a) even when a parallel could in principle be found, it could still be a computationally expensive process to find it (see, for example, the comments on computational expense in Falkenhainer et al., 1989); and (b) often it is difficult to conceive that there could be a
88 John Barnden
parallel at all (again, the far reaches may be like this, or the nylon in the strings example, or the ‘wheels’ in the idiom ‘spinning one’s wheels’). The Anne/Kyle example, for one, shows how context can raise issues. The segment starting ‘but to acknowledge the betrayal’ and proceeding to the end of the excerpt shown raises the target-scenario issue of how able Anne was (before the eight months had passed) to think consciously with the affair-idea. Then, through the use of a familiar metaphorical mapping between physical operation on ideas conceived of as physical objects and mental operations on them, the parallel source-scenario issue is raised of how able Anne’s conscious self was to physically operate on the idea. This issue then exploits common-sense knowledge about the far reaches of physical regions to get an answer. Further detail on this process is given in Barnden (2001, 2006a) and Barnden and Lee (2001). Now consider ‘If [there are any strings] they’re made of nylon –I can’t see them’ from example (1). Here we consider the if-clause as one unit, with the seeing clause as part of its immediate context. Note the high indeterminacy of meaning that ‘The strings are made of nylon’ would have in isolation, when taken metaphorically. There could be connotations other than relative invisibility, such as ones to do with tensile strength, durability, modernity, acoustic properties (cf. some stringed musical instruments), non-biodegradability, etc. But the ‘seeing’ clause raises the contextual issue of how visible the strings are. The visibility issue then draws upon the translucency of nylon to infer intrinsic difficulty of being seen. In that example the invisibility or translucency of the strings is implicit, but there are cases of attached metaphorical strings being explicitly ‘invisible’. One attested example is as follows, with italics added to the relevant phrase: [Brad Meltzer’s] fourth novel, The Millionaires, is the story of two brothers who work for a private bank. A secret, abandoned account containing three million dollars provides them with the key to a new life – but then they discover the invisible strings that were attached ~ to the account. (http://www.ex.ac.uk /RDavies/bankfiction/related. html, accessed 11 July 2007) Thus, our nylon example may be an extension of an existing elaboration involving explicit invisibility. To the extent that the understander is familiar with this elaboration, he/she/it may be biased in the direction of inferring invisibility or the related property of translucency as
A Perspective From AI 89
opposed to other things. Nevertheless, context is still needed in order to confirm that translucency or invisibility is the relevant property, and the point that nylon and the made-of relationship themselves need not be paralleled in the target still stands. We do not claim that all metaphorical understanding is driven by contextual issues. In particular, the meanings of fixed, highly entrenched metaphorical phraseology may perhaps be directly extracted from a mental lexicon without any need to appeal to context. There are intermediate cases where the meaning can be partially derived from the utterance without appeal to context, and partially derived by such appeal. The main message of this chapter is that when and to the extent that the understanding of a metaphorical utterance is contextual-issuedriven, there is a tendency only to use/create as much parallelism as is needed for those issues. Non-parallelism can also arise for independent reasons, however. Thus, it is possible to conceive of the understander trying to find a parallel for something in the utterance but failing (and yet nevertheless extracting rich meaning from the utterance).
3 Broad forms of contextual guidance and directions of processing Contextual-issue drivenness is just one form of contextual guidance. It is possible to conceive of other forms, in particular one where understanding is driven by the nature of the metaphorical utterance itself, but the processing is nevertheless channelled in some way by context. This would, after all, just be an application of the standard technique of ‘forwards’ and ‘informed’ search in AI (Russell and Norvig, 2003). The following example illustrates this notion. Consider trying to find one’s way to a noisy party at a house located in an open plain covered in a dense fog. Every step you take is from your current position, but (barring obstacles) the step is in the direction you perceive the noise coming from. So each step you take is ‘informed’ by your perception of your nearness to achieving the desired goal. A more complex variant of the example is when there are lots of obstacles so that at every stage only a limited number of steps is possible, and at every stage you have to try out various steps from your current position; you then choose the step that seems to lead to the noise getting loudest. Details aside, the crucial aspects of the broad type of search being considered is that it is forwards from some starting point, and that it is informed by some readily calculable measure that tries to estimate the
90 John Barnden
‘nearness’ of each state reached to some goal state (or, possibly, to the nearest one of a set of alternative goal states. The nearness is typically expressed numerically in search algorithms but in principle need not be – there could be instead a way of working out qualitatively that one state is nearer a goal than another. It is important to realize, however, that an important aspect of some types of ‘forwards’ search is ‘backtracking’. A partial path one has uncovered may lead nowhere, so one needs to return to an earlier point and make a different choice from the choice(s) that one previously took at that point. The ‘forwards’ movement of the search is really a spatial metaphor for the fact that each step taken is intended to take one nearer to a goal, not for it actually succeeding in doing so. In the case of the understanding of metaphorical utterances, forwardness corresponds to trying to build up a meaning for the utterance by working from the utterance itself, until a meaning is constructed that appears to fit with context. A quick case of this is when there is a fixed, conventional metaphorical word or phrase whose metaphorical meaning is stated in the lexicon, in which case all that may be needed is to look up that sense and check that it fits context. But in other cases a semantic representation may need to be pieced together from a syntactic parse of the sentence and by also doing various steps of inference. Where it may be necessary to try more than one parse, and at every step of semantic construction or inference there may be several choices available, so that a search process of some sort is needed. The informedness of such a search would consist in having a way of estimating, quantitatively or qualitatively, how close what one had derived so far from the utterance was to fitting with context, or otherwise using the nature of context to guide what step to take at every stage. As a simple example of this ‘otherwise’, if the prevailing topic in the discourse were healthcare, then whenever a lexical choice is needed in the forwards process preference might initially be given to senses that are perceived to be within the healthcare domain. Clearly, on the other hand, a contextual-issue-driven process can metaphorically be re-described as going backwards from contextual issues. Goal-driven or backwards search is also standard in AI. The contextual issues, or rather the requirement to answer the questions implicit in them, can be thought of as the goals in a backwards search process. Recall that in the discussion in Section 2 of the Anne/Kyle example, (3), the contextual issue that is raised leads to a backwards application of the mapping between physical and mental operations on ideas, thereby setting up the issue of the degree of Anne’s conscious self being able
A Perspective From AI 91
to physically operate upon the idea. A reasoning process searching backwards within the source domain from this issue can work its way back to the fact that the idea is in the far reaches of Anne’s mind. This backwards search goes via common-sense knowledge that large distance of an object from an agent affects the agent’s ability to operate on it, and that, if an object is in the far reaches and the agent is in the central part of the region, then there is indeed large distance. Again, a backwards search process can be informed, where now the informedness consists in having a way of evaluating how near a state is to the initial state: in our case, how near to the utterance, to put it roughly. And, it should come as no surprise that the idea of combined forwards/backwards search is often entertained, and there are other less obvious possibilities. As an alternative to using the metaphor of forwardness to describe a process of searching for an utterance meaning, we can say that the process is utterance-driven. So we have utterance-driven and contextualissue-driven understanding of metaphorical utterances (or indeed any utterances) as two extreme possibilities, but it should be clear that there are many intermediate, mixed or otherwise fancier possibilities. And this is all still within the framework of traditional symbolic approaches to language understanding – statistical and connectionist approaches add further possibilities of combining utterance-based and contextual influences in rich ways. Also, we should acknowledge that there could be ways of going backwards from context without being contextualissue-driven, so in the following we will sometimes use the more general term contextually driven. A main point to draw from our discussion is that contextual guidance does not imply contextual drivenness. An informed utterance-driven process also has contextual guidance. But contextual drivenness is one especially strong form of contextual guidance.
4.
Conclusion and additional remarks
We have argued that metaphor often involves non-parallelism but that non-paralleled elements often play a crucial role in implying the metaphorical meaning in context. A focus on non-parallelism goes hand-in-hand with a strong concentration on mechanisms for explaining how context is to guide the construction of metaphorical meaning. This is partly because a strong contextual orientation leads to the understanding process only taking from the source-domain scenario what it needs to illuminate the issues that are actually relevant to
92
John Barnden
the discourse. Non-parallelism also arises from the fact that it is often difficult to see what parallels some source-scenario elements could possibly have in the first place, so that then some search needs to take place to link those elements to aspects of the source scenario that do have parallels. This search needs to be guided in some way, and clearly contextual guidance is natural. We have identified a particularly strong form of contextual guidance as being notable, namely contextual-issue-drivenness. This works backwards from contextually arising issues rather than forwards from utterances, but we have emphasized that contextual guidance in general can take other forms, including forms that are compatible with utterance-drivenness. Another ramification of the ideas in this chapter is worth mentioning briefly here. We have suggested in Barnden, Glasbey et al. (2004) that it is misguided to think that the task of metaphor interpretation is always to find some metaphorical meaning (i.e. target-domain meaning) for every metaphorical sentence or other chunk in a discourse. Instead, when a metaphorical view or set of views is used over a stretch of discourse, it can be useful to view the stretch as holistically combining to describe a source scenario, from which mappings can selectively work to create target-domain information, in a contextually driven way. The aspects of a source scenario that a mapping operates on may have been contributed to by several metaphorical chunks (e.g. sentences) in the discourse, and equally any one metaphorical sentence or other chunk may contribute only indirectly to the source aspects that the mappings work on. The chunk about cat-flaps in example (2) is a case in point; and in a variant of the discourse the chunk could even have been a separate sentence such as ‘The cats had their own cat-flaps’ or ‘The grievances had their own cat-flaps’. In that case there would still have been no need to work out a target-domain meaning for the sentence, any more than there is a need to do so for the chunk ‘with their own cat-flaps’ in the actual discourse. This stance – that metaphorical chunks need not necessarily be given target-domain meanings of their own – is somewhat related to that of Davidson (1984), that metaphors mean what their words literally mean and nothing more. But we do not make such a sweeping claim: we just allow a metaphor not to be given a target-domain meaning, in some circumstances. We suggest that, in effect, context opportunistically selects which aspects of a source-domain scenario end up being mapped or transferred to the target domain, and thereby which metaphorical sentences or parts of sentences end up with target-domain meanings of their own, leaving other related metaphorical chunks merely to contribute
A Perspective From AI 93
towards building the source scenario. But these contributions may be crucial to the meaning of the utterance.
Notes 1. This work was supported in part by grants GR/M64208 and EP/C538943/1 from the Engineering and Physical Sciences Research Council (EPSRC) of the UK. I have benefited from collaboration with Rodrigo Agerri, Sheila Glasbey, Mark Lee and Alan Wallington in the Figurative Language Research Group in my department. 2. The word ‘scenario’ is used loosely to mean just a situation (if a hypothetical one, in the case of the source scenario), and does not appeal to specialized notions of scenario such as that in Musolff (2004). 3. But the following comments to a large extent go beyond or further explicate that analysis.
References Barnden, J.A. (2001). ‘Uncertainty and conflict handling in the ATT-Meta context-based system for metaphorical reasoning’. In V. Akman, P. Bouquet, R. Thomason and R.A. Young (eds.), Proceedings of the Third International Conference on Modeling and Using Context, (pp.15–29). Lecture Notes in Artificial Intelligence, vol. 2116. (Berlin: Springer). Barnden, J.A. (2006a). ‘Artificial intelligence, figurative language and cognitive linguistics’. In G. Kristiansen, M. Achard, R. Dirven and F.J. Ruiz de Mendoza Ibáñez (eds.), Cognitive Linguistics: Current Applications and Future Perspectives, (pp.431–59) (Berlin: Mouton de Gruyter). Barnden, J.A. (2006b). ‘Consequences for language learning of an AI approach to metaphor’. In J. Salazar, M. Amengual and M. Juan (eds.), Usos Sociales del Lenguaje y Aspectos Psicolinguísticos : Perspectivas Aplicadas (pp.15–57) (Palma, Mallorca: Universitat de les Illes Baleares). Barnden, J.A., Glasbey, S.R., Lee, M.G. and Wallington, A.M. (2004). ‘Varieties and directions of interdomain influence in metaphor’. Metaphor and Symbol, 19, 1: 1–30. Barnden, J.A., Helmreich, S., Iverson, E. and Stein, G.C. (1996). ‘Artificial intelligence and metaphors of mind: within-vehicle reasoning and its benefits’. Metaphor and Symbolic Activity, 11, 2: 101–23. Barnden, J.A. and Lee, M.G. (2001). ‘Understanding open-ended usages of familiar conceptual metaphors: an approach and artificial intelligence system’. Technical Report CSRP–01–05, School of Computer Science. (Birmingham: The University of Birmingham). Barr, A. and Feigenbaum, E.A. (eds.) (1981). The Handbook of Artificial Intelligence, vol. I. (Los Altos, CA: Morgan Kaufmann). Davidson, D. (1979). ‘What metaphors mean’. In S. Sacks (ed.), On Metaphor (pp.29–45) (Chicago: University of Chicago Press). Falkenhainer, B., Forbus, K.D. and Gentner, D. (1989). ‘The Structure-Mapping Engine: algorithm and examples’. Artificial Intelligence, 41 1: 1–63. Fass, D. (1997). Processing Metaphor and Metonymy (Greenwich, CT: Ablex).
94 John Barnden Gentner, D. (1983). ‘Structure-mapping: a theoretical framework for analogy’. Cognitive Science, 7 2: 95–119. Gentner, D. and Bowdle, B.F. (2001). ‘Convention, form and figurative language processing’. Metaphor and Symbol, 16, 3&4: 223–47. Gibbs, R.W., Jr. (1994). Poetics of Mind: Figurative Thought, Language and Understanding (Cambridge: Cambridge University Press). Gibbs, R.W., Jr. and Tendahl, M. (2006). ‘Cognitive effort and effects in metaphor comprehension: relevance theory and psycholinguistics’. Mind and Language, 21, 3: 379–403. Giora, R. (1997). ‘Understanding figurative and literal language: the graded salience hypothesis’. Cognitive Linguistics, 8, 3: 183–206. Goatly, A. (1997). The Language of Metaphors (London: Routledge). Glucksberg, S. and Keysar, B. (1990). ‘Understanding metaphorical comparisons: beyond similarity’. Psychological Review, 97,1: 3–18. Grady, J.E. (1997). ‘THEORIES ARE BUILDINGS revisited’. Cognitive Linguistics, 8, 4: 267–90. Hobbs, J.R. (1990). Literature and Cognition (Stanford, CA: CSLI Press). Holyoak, K.J., Novick, L.R. and Melz, E.R. (1994). ‘Component processes in analogical transfer: mapping, pattern completion, and adaptation’. In K.J. Holyoak and J.A. Barnden (eds.), Advances in Connectionist and Neural Computation Theory, vol. 2, pp.113–80 (Norwood, NJ: Ablex Publishing Corp.). Holyoak, K.J. and Thagard, P. (1989). ‘Analogical mapping by constraint satisfaction’. Cognitive Science, 13, 3: 295–355. Hornby, N. (1995). High Fidelity (Harmondsworth: Penguin). Keane, M. (1988). Analogical Problem Solving (Chichester: Ellis Horwood). Lakoff, G. (1993). ‘The contemporary theory of metaphor’. In A. Ortony (ed.), Metaphor and Thought (pp. 202–51) (Cambridge: Cambridge University Press). Langlotz, A. (2006). Idiom Creativity: A Cognitive-Linguistic Model of IdiomRepresentation and Idiom-Variation in English (Amsterdam/Philadelphia: John Benjamins). Leezenberg, M. (1995). Contexts of Metaphor. ILLC Dissertation Series, 1995–7, Institute for Language, Logic and Computation, University of Amsterdam, The Netherlands. Martin, J.H. (1990). A Computational Model of Metaphor Interpretation (San Diego, CA: Academic Press). Musolff, A. (2004). Metaphor and Political Discourse: Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Narayanan, S. (1999). ‘Moving right along: a computational model of metaphoric reasoning about events’. Proceedings of the National Conference on Artificial Intelligence - AAAI ‘99 (pp.121–28) (AAAI Press). Russell, S. and Norvig, P. (2003). Artificial Intelligence: A Modern Approach, 2nd edn (Upper Saddle River, NJ: Pearson Education, Prentice-Hall). Sperber, D. and Wilson, D. (1995). Relevance: Communication and Cognition, 2nd edn (Oxford: Blackwell). Stern, J. (2000). Metaphor in Context. (Cambridge, MA and London, UK: Bradford Books, MIT Press). Wilks, Y. (1978). ‘Making preferences more active’. Artificial Intelligence, 11: 197–223.
Part II Metaphors in Contemporary Public Discourses: Case Studies
This page intentionally left blank
7 Metaphor and Political Communication Jonathan Charteris-Black
Introduction This chapter summarizes and develops further a view of the role of metaphor in political communication that has been introduced in some earlier research (Charteris-Black, 2004, 2005, 2007), in particular the idea that metaphors contribute to the design of a leadership style through appealing to followers to share in a particular representation or construal of social reality. Successful and charismatic leaders create metaphors onto which followers project their own meanings and in doing so find a degree of socio-psychological and emotional gratification. I will first outline the classical view of the role of metaphor in rhetorical persuasion and then analyse metaphor uses by two contemporary politicians – Fidel Castro and Tony Blair. This will lead towards a proposal for a view of metaphor in political communication in which myth and ideology are integrated with the classical role of metaphor. In a diachronic analysis of the ‘body politic’ metaphor, Musolff (2007) summarizes Susan Sontag’s view on how the rhetorical effect of this metaphor has shifted over time, i.e. ‘that there is a qualitative difference between “modern” and “classical” uses’ : whereas the latter, ‘which analogize a political disorder to an illness [ ... ] presuppose the classical medical (and political) idea of balance’, modern (i.e. post-Renaissance) Western culture ‘has seen the emergence of fatalistic diagnoses of diseases in the body politic, which culminated in totalitarian ideologies that favoured images of all-powerful “master illnesses” ‘ (Musolff, 2007: 5). My suggestion is that the classical view of rhetoric – while providing a robust account of how persuasion works through identifying the elements of rhetoric and the figures of speech
97
98 Jonathan Charteris-Black
ETHOS ESTABLISHING THE SPEAKER’S ETHICAL APPEAL
RHETORICAL APPEALS TO THE AUDIENCE
LOGOS APPEALS TO REASON BY ARGUMENTS
Figure 7.1
PATHOS APPEALS TO THE EMOTIONS
Aristotelian view of rhetoric.
which serve to verbalize persuasion – also needs to be modified, so as to accommodate explicitly the roles of ideology and of myth and thus provide a more comprehensive and explanatory account of metaphor in political communication. Power as being an end in itself arose from the political theories of Machiavelli and has become established in the notion of claims to legitimacy as constituting essential components of political rhetoric. The diagnosis of disease in the body politic is precisely such a legitimacy claim, since the elimination of a disease may be taken as something that is inherently ‘right’ and legal. It is this mythic dimension of metaphor that is so persuasive in the communication of ideologies.
Metaphor & Political Communication
99
Whereas in Ancient Greece, analysis of rhetoric – based on the notion of free individuals wilfully choosing leaders – was grounded primarily in transparent language use, modern views on persuasion integrate the persuasive resource of myth that was restricted to the domain of religion in classical thinking. Leadership in the modern period is imposed by those whose will to govern is concealed behind metaphors and symbols that rely on unconscious acceptance rather than critical reflection. It is precisely this symbolic power (Bourdieu, 1991) that accounts for the substitution of the classical idea of balance – evident in the metaphor of the ‘body politic’ – with ideologically driven metaphors of, for example, master illnesses and surgical cures. I will begin by considering the classical view of rhetoric and then outline the roles of ideology and myth in political communication and the role of metaphor in relation to these persuasive modes. Central to Aristotle’s views on rhetoric were the notions of ethos, logos and pathos. Aristotle argued that in addition to taking a stance that was morally worthy, (ethos) and providing proofs in support of argument (logos), the successful rhetorician should also be able to arouse the feeling (pathos) (Aristotle 1991, 1996). This was a significant shift from Plato’s largely negative view of the feelings as likely to interfere with clear thinking; for Aristotle, if arguments could influence feelings, then arguments could also contribute to rhetorical persuasions through arousal of the feelings and so pathos could enhance logos. I have summarized Aristotle’s view of rhetoric in Figure 7.1
1
Ideology
The Aristotelian view of rhetoric needs to be modified in order to provide a more contemporary view of rhetorical persuasion since ideology needs to be integrated with the other components of the classical model as it is essential to modern political communication. According to Seliger (1976:14), ideologies are: ‘sets of ideas by which men posit, explain and justify ends and means of organized social action, and specifically political action, irrespective of whether such action aims to preserve, amend, uproot or rebuild a given social order’. An ideology is a set of meanings through which a particular group is able to form and sustain itself; it therefore serves to create group identity by establishing and reinforcing shared meanings within the group and by communicating this group identity to others as an act of selflegitimization. An ideology claims that what is in the interest of the group is in the interests of all; it is a representation of what is truthful
100 Jonathan Charteris-Black
for some as being truthful for all. These sets of meanings can relate either to politics or to religion, or to any form of social activity. I propose that metaphor is one of a number of linguistic, cognitive and symbolic resources employed by political leaders for communicating ideology. By establishing a shared view – by arbitrary decisions about what is right and wrong, good and bad – a group engages in a process of selflegitimization through which it aspires to power. Language is evidently central in the process of ideology formation because it is the verbalization of sets of ideas that constitutes ideological representation of the world and the basis for acting in it. Ideological statements are textual in nature – Das Capital, the Koran, The Declaration of Independence or the Ten Commandments – and have the speech act force of committing the members of a group to certain forms of action. Ideologies are therefore formal statements on which subsequent legitimacy claims are based and ideological language is performative rather than representational; therefore, ideology needs to be integrated into an explanatory account of political rhetoric.
2
Myth
My view on myth takes its inspiration from Cassirer (1946), Barthes (1993) and Flood (1996) and is explained in more detail in CharterisBlack (2005: 22–4). Here I will summarize some of the central elements of this point of view. Myths are explanatory narratives for anything that requires an explanation and therefore purport to be a set of truths. Explanatory needs vary according to specific psychological and social conditions. At various times there may be the need for explanations of metaphysical phenomena such as the origins of the universe, of male and female or other bipolarities. At times of social turmoil explanations may be required for the causes of social evils such as crime; economic problems may also demand explanations. In times of epidemics or mysterious deaths, mythic explanations may include the identification of scapegoats – hence the identification of witches and the possessors of demonic powers. In more rational times, explanations may be needed for the origins of the matter of which the universe is comprised, and these may still retain traces of mythic irrationality. Myth engages the hearer through a narrative that embodies a set of beliefs expressing aspects of the unconscious. It provides a narrative-based representation of intangible but evocative experiences that are unconsciously linked to emotions such as sadness, happiness and fear. Its discourse function is to explain and it is a two-sided weapon because it can be used for
Metaphor & Political Communication
101
positive or negative evaluation. Myths rely heavily on metaphors and symbols. Mythic thinking has been contrasted with Aristotelian logic (Taran, 2000) since it permits contradictions and its narrative operates in cyclical rather than linear time. The cyclical notion of time in mythic thinking persuades followers of the inevitability of events; for example, Martin Luther King employed metaphor to develop the myth that he was a Moses who would, by analogy, lead African Americans out of the oppressions of Egypt. Myth encourages icons of national unity such as Mother Russia, Uncle Sam, or John Bull; symbols such as these gain rhetorical power by defying analysis. Taran proposes that mythic thinking is characteristic of the discourses of both the extreme right and the extreme left. Evidently, then, the persuasive power of myth was not an element that Aristotle would happily have integrated into his model of rhetoric because its substitution of logic by narrative gives it the potential for motivating irrational and potentially destructive behaviour. In the classical view, acts of wanton destruction were undertaken by gods in the pursuit of their own ends whereas in the modern period it is humans who undertake wanton acts of destruction in the pursuit of power. However, through myth such destructive acts may be represented as pure and indeed natural: ‘Myth does not deny things, on the contrary, its function is to talk about them; simply, it purifies them, it makes them innocent, it gives them a natural and eternal justification, it gives them a clarity which is not that of an explanation but that of a statement of fact’ (Barthes 1993: 143). Awareness of the subliminal level of political discourse is not restricted to language alone. The subliminal level – by which I mean positive or negative evaluations arising from unconscious associations – may be communicated by other semiotic means; these could include photographs, clothes, political cartoons, political posters and short film excerpts such as those used in party political broadcasts. Barthes (1993: 126) proposes that ‘The mythical signification [....] is never arbitrary; it is always in part motivated, and unavoidably contains some analogy’. Through conceptual analysis of the nature of the analogy on which political myth is based we can identify what motivates myths and exactly how ‘history’ is transformed into ‘nature’. As Barthes (1993: 124) continues: ‘A myth is a type of speech which is defined by its intention ... .. much more than by its literal sense [....].’ This view of language evokes Austin’s and Searle’s analyses of speech acts through identifying speakers’ intentions; I will explore the notion of intention further in the discussion of ethos in the following section.
102
Jonathan Charteris-Black
Ideology and myth differ in the extent to which appeal is made to conscious cognition or to unconscious association. As with reasoned argument (or logos), ideology appeals through a consciously formed set of beliefs, attitudes and values while, as with pathos, myth appeals through only partially conscious sets of beliefs, attitudes and feelings. I propose that metaphor shares with ideology and myth the discourse function of persuasion and indeed connects them to provide an expressive resource for communicating both the cognitively oriented dimension of ideology as well as the emotionally oriented dimension of myth. It is the powerful effect of the mythic dimension that is so often persuasive in the formation and communication of ideology: ‘Semiology has taught us that myth has the task of giving an historical intention a natural justification, and making contingency appear eternal. [....] If our society is objectively the privileged field of mythic significations, it is because formally myth is the most appropriate instrument for the ideological inversion which defines this society [....]’ (Barthes 1993:142).
IDEOLOGY Conscious set of ideas, beliefs and values
COGNITION
METAPHOR
EMOTION
POLITICAL MYTH
Figure 7.2 Metaphor mediating myth and ideology.
Metaphor & Political Communication
103
My argument here is that metaphor has the potential to integrate conscious ideology with less conscious myth and is summarized in Figure 7.2. Metaphor is a persuasive aspect of discourse because it mediates between conscious and unconscious means of persuasion – between cognition and emotion. It is therefore a central strategy for political legitimization. This chapter will illustrate a contemporary view on rhetoric with reference to some specific uses of metaphor by Fidel Castro and Tony Blair. I will now consider each of the primary functions of the Aristotelian view of rhetoric and what modifications need to be made to this classical model to integrate ideology and myth.
3
Establishing the politician’s ethos
Metaphors can heighten a politician’s ethical qualities by self-representation as what I will refer to as a leader who has the right intentions. Intentionality is central in measuring an individual’s ethical value. Self-representation by a politician as a moral arbiter provides the basis for representing followers as insiders, who share the ethical virtues of the leader, and opponents as outsiders. This view of metaphor fits well with a recent theory of political communication developed by Chilton (2004) that proposes that the basis for evaluation is spatial proximity to the speaker because the Self is the origin of what is true and what is right: ‘the general idea is that Self is not only here and now, but also the origin of the epistemic true and the deontic right’ (Chilton, 2004:59). It is the deontic sense of being righteous that is established through having right intentions. An important ethical dimension of metaphor is that it can be employed either as a form of self-evaluation of the speaker or as a form of evaluation of policies, political opponents, or groups in society; I will illustrate this with reference to Fidel Castro: To understand this problem of America, a book which explains the truculence of U.S. policy in our continent must be read. It is called “The Fable of the Shark and the Sardine”. The shark is the Yankee empire; the sardines the weak American nations. (Castro, 30 August 1960) Castro’s preferred animal metaphors are the shark and the sardine to refer to the USA as a symbol of greed and rapacity with Cuba and other small countries as symbols of moral integrity. There are several
104
Jonathan Charteris-Black
characteristics of this metaphor that are worthy of note and which are discussed in Charteris-Black (2007); deep-sea fishing is Castro’s own favourite pastime and has attracted others, such as Ernest Hemingway, to Cuba. More importantly, the choice of a literary genre such as fables that are based on the conceptual metaphor PEOPLE ARE ANIMALS is significant. Fables have a didactic moral purpose in which animal attributes signify human moral characteristics and interpreting the metaphor relies on working out correspondences between animal and human attributes and behaviours. It is this didactic function that is developed in a political context. Other primary attributes of the shark – size, greed, rapacity etc. are also part of the ethical implications of metaphor since bigness of size is associated with immoral behaviours such as being greedy and rapacious. Such juxtaposition of opposites is common to the genre of the fable because contrasts between appearance and reality are fundamental to this genre – for example, slow animals such as tortoises have speed of thought and small ones are typically intelligent because they have to think harder to outwit their opponent. The entailments of metaphors that originate in such fable-based thinking heighten ethical qualities by representing the leader as a judge of ethical issues, as someone who has the right intentions towards followers. The fable-teller allies himself to knowledgeable figures – parents or teachers – to whom respect is due. Therefore self-representation as a fable-teller by a leader provides the basis for self-representation as a moral arbiter and as a ‘natural’ leader who is expected to provide an ethical perspective on social issues.
4
Heightening the pathos
Increasing the emotional impact is a very vital role for metaphor in a range of political contexts and is part of what I will refer to as sounding right. These include the need to sustain morale during times of national crisis, and the need to inspire in times of apathy; emotions of fear are aroused in times of invasion by the threat of an unknown Other, because this is how humans respond when protecting their group from danger. Fear of the unknown Other is often communicated by metaphors and symbols representing the Other by personification; this figure provides a potentially emotive and cognitively accessible framework for the evaluation of abstract political ideologies because it relies on pre-existent, culturally rooted stereotypes to communicate emotionally potent evaluations. Personifications have long provided propaganda material for the political cartoonist; Victorian editions of the London
Metaphor & Political Communication
105
Gazette symbolize the nations of Europe by caricatures of actual leaders (e.g. Napoleon) or mythical ones (e.g. John Bull) or by images of animals – the Russian bear, the French cock etc. Symbolic figures such as the Wandering Jew, the marauding Turk have a long history in expressing emotive responses to particular racial and ethnic groups. It is these that are drawn on in controversial contemporary visual representations of ideological reality – such as those of suicide bombers. Personifications provide a concrete and accessible framework for the evaluation of abstract political ideologies. Negative evaluations may be communicated by depersonifications (Charteris-Black 2004: 168) that represent a political opponent as an animal, a parasite, or a thing. Castro’s metaphors ‘the US is a greedy shark’ and ‘Cuba is a brave fish’ evoke feelings of heroism and courage from a combat between unequal opponents: If this country, vis-a-vis imperialism – which is a wild beast, a barracuda, a shark vulture with all its deft tricks, if this small country had ever shown fear and hesitation before the imperialists they would have devoured us. (Castro, 20 May 1970) A good example of how an appeal to pathos can be combined with an appeal to ethos is when Castro breathed life back into his favourite metaphor by rejecting the view of Cuba as a powerless sardine: If someone was able to write that bit about the shark swallowing the sardine, and a book with that title even appeared in the initial years of the revolution by an author who at that time had certain decorously progressive ideas, today you cannot talk about the shark and the sardine. Today you can talk about the shark and the fireball, and ask if the shark could swallow the fireball. [applause] Today you can talk about the shark and steel, and ask if the shark could swallow that gigantic ball of steel which is the Cuban revolution today. [applause, chanting] (Castro, 2 November 1991) The audience response here indicates the success of the intertextual strategy of elaborating a metaphor that has already been established. An important characteristic of the use of metaphors in that it is up to followers to interpret them. A joint act of meaning creation is effective in political argument because it stimulates the hearer to share the leader’s inferences and therefore to participate in his or her discourse. Shared beliefs and implied common understanding are at the heart of pathos, interpretation is evoked rather than stated.
106
5
Jonathan Charteris-Black
Creating humour
Humour is a form of appeal the audience’s emotions and is also part of what I have referred to as ‘sounding right’. Churchill frequently employed metaphor to amusing effect: An appeaser is one who feeds a crocodile – hoping that it will eat him last. Trying to maintain good relations with a Communist is like wooing a crocodile. We are waiting for the long-promised invasion. So are the fishes. (Enright, 2001:42) In the 1980s humour was employed by the political right in their heyday. Margaret Thatcher’s humour successfully combined irony with animal metaphors: Today, instead of the voice of compassion, the croak of the Quango is heard in the land. (Margaret Thatcher, Conservative Party Conference, October 1978) So it’s back to square one for the Socialists. The Labour Leopard can’t change its spots – even if is sometimes thinks wistfully of a blue rinse. (Margaret Thatcher, Conservative Party Conference, October 1988) Thatcher’s close political ally Ronald Reagan also employed humour to good effect: My friend, some years ago, the Federal Government declared war on poverty, and poverty won. (Reagan, 25 January 1988) Humour depends on context in terms of culture but also in terms of audience, as Virtanen and Halmari (2005) highlight: ‘If the goal of a persuasive text is to strengthen the beliefs of those audience members who already believe in the goals of the speaker/author, the most effective linguistic forms for that goal will be different from the most effective linguistic forms if the goal is to persuade a fiercely opposing audience. And in many cases, the persuaders face multiple audiences – those who already share their views and those who do not’ (Virtanen and Halmari, 2005: 9). Humour is a particularly effective strategy when dealing with such multiple audiences. This is because it can both strengthen the support of followers and undermine the opposition of opponents by uniting them through an empathetic response.
Metaphor & Political Communication
107
Laughter is an affective response that unites and is therefore highly conducive to political persuasion. A contemporary rhetorical deficiency has been a decline in the ability of politicians to be humorous, perhaps because of the risks entailed by humour. Although humour can be unifying, it is only effective if it appeals to the values of the speech community to which it is addressed; however, given the increasingly multi-layered nature of political audiences in Western democracies it is becoming more challenging to employ humour effectively because the politician is dealing with disparate discourse communities, each with their own norms and styles of communication – of which humour may not be one. The style of the intercultural negotiator is more often earnest than humorous, and there has been a shift over to ideological or to mythical thinking at the expense of appeals to pathos through humour. Fidel Castro has often employed humour in his design of leadership style; the informality of humour serves to reduce the rhetorical distance between leader and followers by communicating that they share the same values; here he makes fun of ideological stereotypes: I’ll go to hell. The heat will be unbearable. But the pain won’t be as bad as expecting too much from heaven, which always breaks its promises. When I arrive in hell, I’ll meet Marx, Engels and Lenin. And you too, by the way, because capitalists also go to hell, especially when they enjoy life too much. (Castro, 27 October 1994) Castro combines humour with animal metaphors as in the following: It is difficult to speak of social revolutions when we do not even have independence. What happens? Cuba, the small country and the elongated island shaped like an alligator, makes its revolution and then comes the shark, the ferocious tiger, or the elephant that wants to trample on it, and Cuba sits there alone. (Castro, 18 July 1985) Stylistically, there was a humorous element in evoking a child’s fable to communicate a political argument. Such use of humour is a very important stylistic shift that makes the leader at one with his audience, since laughter is a shared social experience and is the right sound for creating an emotional climate that encourages acceptance of the right intention. Humour is a form of style shifting that counterbalances the didacticism that often characterizes other modes of revolutionary rhetoric. Humour creates interpersonal meaning and shows that underneath the high-minded revolutionary there is a warm human being who sounds right.
108 Jonathan Charteris-Black
6 Communicating and explaining political policies (logos) Logos is a central component of persuasion in political communication because it shows that the leader has the power of what I will refer to as thinking right. Metaphors are very effective in the communication of policy because they provide cognitively accessible ways of communicating political policy. They provide proofs or warrants that support arguments and have particular entailments. Metaphors may be exploited or manipulated or even reversed in order to communicate a particular political argument. They can also be analysed in the way that Musolff (2004) proposes: ‘Linguistic analysis [ ... ] is concerned with reconstructing the sets of presuppositions that are contained in scenarios, and can thus expose the analogical warrants used to legitimise specific conclusions’ (Musolff (2004: 175). The metaphor of communism as a disease communicated a particular political argument – i.e. the speed and inevitability of the spread of communism that entailed the need to take remedial action. The metaphor of Domino theory communicated the political argument that if one country fell to communism then so would its neighbour and entailed the need to stop this chain effect. Metaphors can be used to frame a particular political argument and, like myths, in the very selection of the terms through the scenarios that are chosen they suggest particular conclusions. For example, in the US Congress there was a rhetorical analogy between Saddam Hussein and Hitler which supported the argument that a war on Iraq would be as justified as the Second World War; however, opponents of the war supported their argument with an analogy between the Iraq War and the Vietnam War. The recent political activity of war veterans for the Democrats in the United States has supported the latter analogy. In the United Kingdom, Margaret Thatcher reversed the Iron Lady metaphor from one that communicated inflexibility and heartlessness to a symbol of strength and self-conviction. Tony Blair’s metaphor ‘I have no reverse gear’ was subsequently thrown back at him when he shifted his decision as to whether to allow a referendum on Europe. Metaphors drawing on logos contribute therefore to the historical evolution of political arguments. Conceptual aspects of a source domain can be exploited to develop a political argument; in the case of Castro’s shark and sardine metaphor we usually think of sharks as single operators while sardines typically
Metaphor & Political Communication
109
move in shoals. Castro uses the metaphor to appeal to his Latin America allies: The powerful country had violated the law against economic aggression; but when the time came to condemn the shark, the sardines met and condemned the other sardine. But this sardine was no longer a sardine. (Castro, 23 April 1961) Here there is a political argument that smaller countries gain political power by forming alliances against larger ones. If we accept the conceptual representation COUNTRIES ARE ANIMALS, then the entailment arising from interpretation of the metaphor is that the small countries are the republics of Central America that constitute Cuba’s natural allies and with whom Castro has sought to form alliances. Sometimes metaphor entailments are made explicit: In a brief meeting with representatives of the Western press during a reception offered for his visiting colleague from Trinidad-Tobago, Eric Williams, the Cuban prime minister described the TIAR [‘InterAmerican Treaty of Reciprocal Assistance’] as a shameful and indecent pact between the shark and the sardines. By means of that treaty, according to Castro, the shark is attempting to keep the sardines unarmed in order to train armies and maintain the dominance of transnational enterprises. This led reporters to identify the shark as being the United States and the sardines the countries south of the Rio Grande. (Agence France-Presse, 21 June 1975) Here the basis for analogy explains the warrants of the argument. New metaphors can lead us to fresh perspectives on political issues because they explain political policy and communicate political arguments, but they also contain a hidden bias because they make certain assumptions that are not clearly formulated at the time when the metaphor is first used. Metaphors, like myths, do not spell out what is taken for granted as natural. They provide a cognitive filter that colours the covert political intentions of the politician; metaphors therefore only reflect ‘right’ thinking from the perspective of the particular social group who benefits from their arguments.
7
Metaphor in rhetorical interaction
While, for sake the of analysis, different rhetorical purposes of metaphor have been identified separately, in practice the effectiveness of
110
Jonathan Charteris-Black
metaphor in political communication is that it combines interacting rhetorical appeals. We have seen, for example, how Castro appealed to the emotions by employing animal metaphors for humorous purposes, but they also contributed to establishing his ethos through their didactic role as well as communicating specific political arguments. When repeated systematically over time the use the rapacious shark and sardines metaphor, as well as another of his preferred myths – the story of David and Goliath – serves to develop political myths resistant to analysis. I would like to illustrate a similar interaction of rhetorical purposes by Tony Blair’s use of a particular metaphor – that of the ‘rogue state’. Figure 7.3 shows the frequency of this term in Hansard ( the record of parliamentary debates) over the last 16 years. We can see that the use of the term ‘rogue state’ shown in Figure 7.3 provides a good indicator of New Labour foreign policy, with a rise taking place during the war on the Taliban in Afghanistan and then reaching a peak with the Second Gulf War in Iraq. It was used only once in relation to Iraq in the period 1989–1992 but 68 times in relation to Iraq in the period 2002–2004. Of all uses in the period 1988 until the present, 21% were in 2003. The notion of a rogue state is therefore one that
400 350 300 250 200 150 100 50 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year Figure 7. 3 Occurrence of ‘rogue state’ in Hansard debates 1995–2005.
Metaphor & Political Communication
111
is employed for the general rhetorical purpose of communicating New Labour foreign policy. A typical use might be the following: After 11 September, it was time to take an active, as opposed to reactive, position on the whole question of weapons of mass destruction. We had to close down the capability of the rogue states – usually highly repressive and unstable – to develop such weapons and the commercial networks ... the country where we had to take a stand was Iraq. (Tony Blair, Commons Hansard, 14 July 2004) Here we see how the use of ‘rogue state’ contributes to the effect of ‘sounding right’. In what sense then is ‘rogue state’ a metaphor? The first sense of the word ‘rogue’ is ‘a dishonest or unprincipled man’ (OED 2nd edition, 2003); so when applied to a state it is a well established figure of speech – a personification. It evokes emotions of the need for protection. These emotions are further aroused by a secondary meaning ‘an elephant or other large animal living apart from the herd and having savage or destructive tendencies’. Evidently this sense is also transferred to a political context to create a naturalized myth resistant to analysis. The fact that a non-literal sense is being employed is evident from the fact that a Member of Parliament – seeking to denaturalize the myth – requested a literal definition: Mr. Gummer: To ask the Prime Minister how he defines rogue states. The Prime Minister: The phrase “rogue state” is generally used to describe those believed to be engaged in the proliferation of weapons of mass destruction and their means of delivery, or those which support terrorism and offer succour to those who commit terrorist acts. (Tony Blair, Commons Written Answers, 19 January 2004) This is an instance of providing a particular context for the ‘rogue state’ metaphor: it is not just any state that has destructive tendencies but one that is claimed to be engaged with weapons of mass destruction. The incantation of this fixed phrase contributed to the development of a myth in the political debates leading up to the invasion of Iraq. The ‘rogue state’ metaphor became synonymous with the Prime Minister’s argument that military action should be undertaken against the specific state that was identified as developing ‘Weapons of Mass Destruction’; the scenario they evoked implied precisely this, as is shown clearly in the following exchange: Mr. James Gray (North Wiltshire): [ ... ] but does he agree that, lying behind that debate, there would have to be the very fundamental
112
Jonathan Charteris-Black
principle that if a rogue state threatens the peace and security of this nation, it would be a gross dereliction of duty if we did not take military action against it? The Prime Minister: Of course we must make sure that we are protected against any such state. I hope that this Government’s record on issues such as Kosovo and Afghanistan and, indeed, in relation to Iraq shows that we are prepared to take that action. (Commons, Hansard, 10 April 2002) Here we see that the ‘rogue state’ metaphor is shifting from its role in arousing pathos by sounding right to its role in logos – thinking right – as it provides a warrant for the argument for military intervention in Iraq. Why was the metaphor chosen? Based on the idea of ‘a rogue elephant’ it argues – drawing on emotions concerned with protection of the group – that such a state is dangerous and this entails a need to eliminate it as a source of danger. This destruction of something that is naturalized as ‘threatening’ demonstrates that the agent does indeed have the right intentions. The state of being a rogue elephant is not one that is considered reversible. It also entails that whoever eliminates the rogue state/elephant is not an aggressor because he is doing something that is for the general benefit. It therefore represents what might in other circumstances be interpreted as an aggressive action as defensive because of the right intentions of the agent – it therefore holds an appeal to ethos as well as to logos and pathos. Moreover, its constant reiteration, as we have seen, gives it the status of a political myth – an explanatory account of good and evil. Barthes proposed that myth is depoliticized speech – as he puts it: ‘A conjuring trick has taken place; it has turned reality inside out, it has emptied it of history and has filled it with nature, it has removed from things their human meaning so as to make them signify a human insignificance’ (Barthes 1993: 142–3). The adoption of the concept of ‘rogue state’ has precisely the role in political communication of emptying it of history, how a nation came to be regarded as a rogue state – and it removes the human implications of eliminating the rogue.
8
Conclusion
In this chapter I have argued for the continuing value of the classical Aristotelian view of rhetoric and I have shown how metaphor can be employed to create appeals based on logos, pathos and ethos; but I have also argued that the emergence of specific ideologies in the modern
Metaphor & Political Communication
113
period requires that a model for metaphor in political communication should incorporate an ideological and a mythic dimension. I have illustrated the interaction of myth and ideology with pathos, ethos and logos through analysis of Castro’s animal metaphors – specifically ‘the shark and the sardine’ – and the use of a ‘rogue state’ metaphor by Tony Blair that draws on the natural images of the rogue and the rogue elephant. This model of metaphor in political communication is summarized in Figure 7.4. Needless to say there may be elements of political communication that are overlooked in this model; the role of metaphor is not exclusively employed in each of these rhetorical roles but has the potential to be so utilized. Nor is Figure 7.4 intended to imply that different metaphors perform these appeals separately: as I have illustrated in the
METAPHORS THAT ESTABLISH ETHICAL INTEGRITY
METAPHORS THAT COMMUNICATE POLITICAL ARGUMENTS
HAVING THE RIGHT INTENTIONS
THINKING RIGHT
FORMING LEGITIMACY
METAPHORS THAT HEIGHTEN EMOTIONAL IMPACT
METAPHORS THAT COMMUNICATE IDEOLOGY AND POLITICAL MYTH
SOUNDING RIGHT (e.g. BY HUMOUR)
TELLING THE TRUTH AND TELLING A SET OF TRUTHS
Figure 7.4
A contemporary view of metaphor and political communication
114
Jonathan Charteris-Black
analysis, a single metaphor may appeal to ethos, pathos and logos while simultaneously creating a myth and communicating an ideology.
References Aristotle (1991). The Art of Rhetoric, ed. by H. C. Lawson-Tancred (London: Penguin). Aristotle (1996). Poetics, ed. by M. Heath (London: Penguin). Barthes, R. (1993). Mythologies. Selected and trans. from Mythologies (1957) by A. Lavers (London: Vintage). Bourdieu, P. (1991). Language and Symbolic Power, ed. by J.B. Thompson , trans. by G. Raymond and M. Adamson (Oxford: Polity Press). Cassirer, E. (1946). The Myth of the State (New Haven and London: Yale University Press). Castro, F. ‘Castro Speech Database’. Online. Available at: http://www1.lanic. utexas.edu/la/cb/cuba/castro.html (Accessed 19 January 2005) Charteris-Black, J. (2004). Corpus Approaches to Critical Metaphor Analysis (Basingstoke : Palgrave Macmillan). Charteris-Black, J. (2005). Politicians and Rhetoric: The Persuasive Power of Metaphor (Basingstoke : Palgrave Macmillan). Charteris-Black, J. (2007). The Communication of Leadership: The Design of Leadership Style. (London : Routledge). Chilton, P. (2004). Analysing Political Discourse: Theory and Practice (London: Routledge). Churchill, W.C., UK Quotations. Online, available at: http://www.ukquotations. co.uk/Winston-Churchill.htm (Accessed 28 May 2007) Transportation Home page. Online, available at: http://www.craven.co.uk/ transportation_home_page.htm (Accessed 28 May 2007) The Churchill Society London. Online, available at: http://www.churchillsociety-london.org.uk/LaFrance.html (Accessed 28 May 2007). Enright, D. (2001). The Wicked Wit of Winston Churchill (London: Michael Omara Books Ltd). Flood, C. G. (1996). Political Myth: A Theoretical Introduction (London and New York: Garland). Hansard. Online, advanced search available at: http://www.publications. parliament.uk/cgibin/semaphoreserver?FILE=search&SAVEDB=semsimple (Accessed 28 May 2007). Musolff, A. (2004). Metaphor and Political Discourse: Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Musolff, A. (2007). ‘Is there such a thing as discourse history? The case of metaphors’. In C. Hart and D. Lukès (eds.), Cognitive Linguistics in Critical Discourse Studies: Application and Theory. (pp. 1–26) (Cambridge: Cambridge Scholars Press). OED/ Oxford English Dictionary Online, available at: http://au.oup.com/content/ General.asp?ContentID=475&MasterID=48 (Accessed 28 May 2007). Reagan, R. (1988). ‘State of the Union Address, January 25th 1988’. Online, available at: http://reagan2020.us/speeches/state_of_the_union_1988.asp (Accessed 28 May 2007).
Metaphor & Political Communication
115
Seliger, M. (1976). Ideology and Politics (London. Allen & Unwin). Sontag, S. (1979). Illness as Metaphor (London: Allen Lane). Taran, S. (2000).’Mythical thinking, Aristotelian logic, and metaphors in the parliament of Ukraine’. In C. D. Landtsheer and O. Feldman (eds.), Beyond Public Speech and Symbols: Explorations in the Rhetoric of Politicians and the Media. (pp. 139–52) ( Westport and London: Praeger). Thatcher, M. Online, available at: http://www.brainyquote.com/quotes/ authors/m/margaret_thatcher.html (Accessed 28 May 2007). Virtanen, T. and Halmari H. (2005). Persuasion across Cultures: A Linguistic Approach (Amsterdam and Philadelphia: John Benjamins).
8 Missions and Empires: Religious and Political Metaphors in Corporate Discourse Veronika Koller
Introduction ‘For these men, the economy was their religion and their party, the company was their church and their parliament.’ (Spinnen and Posner, 2005:178, trans. VK) Proposing a paradigm shift from religion to politics to business as the defining power in post-industrial societies, this chapter looks at how residues of the former two surface in corporate discourse as metaphoric expressions (e.g. corporate mission, empire-building). Market factors such as global competition and saturated demand are forcing companies to differentiate themselves by constructing corporate brand personalities. Simultaneously, multi-national corporations have amassed economic and political power that in many cases already exceeds that of nation states, let alone faith-based communities. This chapter argues that the producers of corporate discourse appropriate previous paradigms by means of metaphor both to construct their brand personality and to legitimate their increasingly hegemonic role by drawing on established, if historical, sources of power.1
1
Shifting paradigms
The quotation above , while specifically referring to the disillusionment and depoliticization of Germans who had first supported fascism and 116
Metaphors in Corporate Discourse
117
then survived Word War II, nevertheless addresses a wider phenomenon: how the market economy gained in importance to become the defining paradigm in contemporary Western society, replacing politics and religion (Koch, 2001). A paradigm can be called defining when it provides the conceptual schemas to make sense of social and individual life in a given society at a given point in time. Obviously, the notion that sees religion, politics and the economy as successive defining paradigms, with the latter ascending during the second half of the twentieth century and particularly after 1989, is over-simplified as it stands. Politics did not simply ‘take over’ from religion in the eighteenth century, to mark the beginning of a ‘disenchanted’ (Weber, 1920 [1930]) post-Enlightenment view of the world. Likewise, while the market economy has definitely gained in importance and scope after the end of the Cold War, its concomitant ideology has been in the making for a much longer time. Also, although institutional religion and the state have been sidelined as providers of meaning and community, they continue to be available as points of reference and comparison, not to mention the upsurge in (fundamentalist) religion in regions as superficially distinct as the United States and large parts of the Islamic world. Finally, to complicate matters further, religious fundamentalism is often fuelled by political and economic factors. While the transition from religion to politics to the economy is far from smooth and clear-cut, then, the claim has still been made that in Western societies, a broad historical trend can be observed which has institutional religion lose its meaning-making power for social groups and individuals in favour of the grand political ideologies of nationalism, fascism and socialism in the nineteenth and twentieth centuries. These in turn have been eclipsed by the increasingly central paradigm of the market economy, which, it can be argued, currently provides models for social identities, relations and communities, as well as for beliefs, values and goals. In doing so, the economy has become endowed with an ideology, i.e. the knowledge structures that define a group’s membership, its tasks and activities, values, goals and attributes (van Dijk, 2005: 87). Table 8.1, a summary of the argument made by Koch (2001), illustrates the parallels between religion, politics and the economy as defining paradigms. Even though the importance of institutional religion and the state may have waned, they still provide cognitive raw material, in that social groups and individuals retain knowledge structures about institutional religion and the state which inform their ideas of how social relations are or should be organized, what behaviour by what individual or
118
Veronika Koller
Table 8.1 Comparison of defining paradigms (based on Koch, 2001) Religion
Politics
Economy
novel
building
altarpiece, church building, mass cathedral
parliament
community
congregation
nation
ethics
Christian morality
bourgeois ethics
generalized philosophical code genesis idea of human being institution metaphorical relationships
God-given reality
objectivity/reason
advertizing, film, pop music shopping mall, corporate headquarters brand, corporation economic independence profit maximization
creation God’s creation, believer church Church as mother, Jesus as bridegroom of the soul, God as father God dogmatism clergy theology
aesthetics citizen
art form
metaphysics power abuse representatives theoretical foundation
ultimate goal
heaven
nation state state as father, citizens as brothers (fraternité) nation state imperialism politicians philosophy (German idealism)
harmonious society
branding consumer, employee corporation (corporate) brand as parent, brand as lover and friend corporation neoliberalism executives business administration, especially marketing (psychology, anthropology etc.) wealth
corporate actor is appropriate in what context, and how groups and individuals can be categorized. Crucially, such knowledge structures can serve as the material for metaphor. In the case of religion, politics and the economy, we are dealing with a rather complex situation in which all three can serve as either source or target domain for the others, as suggested by anecdotal evidence and demonstrated in empirical studies (Charteris-Black, 2004: 79, 103–4; Mautner, forthcoming).
Metaphors in Corporate Discourse
119
1
Religion
Politics 3 4
5 2
6
Economy
Figure 8.1 Metaphoric relations between religion, politics and the economy.
In Figure 8.1, arrows indicate mappings from source to target. Thus, we get: 1. Religion mapped onto politics: ‘But what happened to others who could allow Labour to claim to be a broad church?’ (Seddon, 2005). 2. Religion mapped onto the economy: ‘mission statement’, ‘NYK Lines business credo’.2 3. Politics mapped onto religion: ‘the global constituency of the Anglican church’ (‘New archbishop enthroned’ news item on BBC website, 2003 [news.bbc.co.uk/2/hi/uk_news/2803139.stm; accessed 7 July 2008]). 4. Politics mapped onto the economy: ‘corporate citizen’, ‘corporate governance’, ‘Magna Corporate Constitution’. 3 5. Economy mapped onto religion: ‘We all tend to think the papal office is like the role of a chief executive officer of an international organization’ (BBC Radio 4, Today programme, 28 Feb 2005).4 6. Economy mapped onto politics: ‘The Conservative Party doesn’t win elections for precisely the same reasons that Marks & Spencer doesn’t dominate the high street anymore’ (Lord Tebbit, former Conservative party chairman, BBC Radio 4, Today programme, 16 May 2005).5
120 Veronika Koller
‘Like every CEO of a major organization, I am responsible for what happens at the Department of Justice’ (Alberto Gonzalez, US Attorney General, BBC Radio 4, Today programme, 22 March 2007). Rather than addressing items 5 and 6, i.e. the colonization of sectors other than the economy by economic and corporate discourse (Fairclough, 2003: 103), the focus in this chapter will be on items 2 and 4, i.e. on the appropriation of religion and politics as metaphor source domains in corporate discourse.
2 Cognitive Critical Discourse Analysis and the role of metaphor The approach taken in this chapter to the study of metaphor in discourse relies on an integrated critical cognitive analysis of discourse. In particular, it focuses on the role that metaphor plays in linking cognition with language use as social practice (see Musolff, 2004; O’Halloran, 2004). Alternatively referred to as ‘cognitive text linguistics’ (de Beaugrande, 2004), ‘cognitive sociolinguistics’ (Kristiansen and Dirven, forthcoming), ‘cognitive pragmatics’ (Nuyts, 2004) or ‘sociocognitive discourse studies’ (van Dijk, 2006), research into the socio-cognitive aspects of discourse investigates how humans use their mental faculties and semiotic systems (textual metafunction of discourse) to facilitate their interactions (interpersonal metafunction) and make sense of the world (ideational metafunction). Conceptual metaphors are here seen as derived from embodied primary metaphors (Grady, 1997). These are usually of a spatial nature and are theorized to form in early childhood through the experience of the child moving his or her body through space and interacting with objects and other people. An example would be the conceptual metaphor POWERFUL IS UP , which originates from the interaction with taller persons, such as the child’s caregivers, who are also experienced as more powerful. In a process of conflation, the physical and the psychological experiences converge, giving rise to the typically concrete source domain and typically abstract target domain of conceptual metaphor. It is worth noting that the theory of embodiment has its (often unacknowledged) origins in the notion of psychophysical parallelism, a central tenet of gestalt theory as developed by Max Wertheimer (1920). At the surface level of language, the above-mentioned embodied metaphor is realized in metaphoric expressions relating to the idea that governing is being
Metaphors in Corporate Discourse
121
elevated, e.g. ‘Cook was poised for high office in Brown cabinet’ (Nelson and Macdonnell, 2005) or the phrase ‘to rise through the ranks’. According to the theory presented here, the pool of conceptual, ultimately primary, metaphors is enlarged through combinations and semantic re-accentuation. Thus, GOVERNING IS BEING ELEVATED can combine with BUSINESS IS WAR – itself a culture-specific derivation of the spatial ideological model that sees OPPOSITION AS PHYSICAL AGGRESSION – to yield SENIOR MANAGEMENT ARE GENERALS, expressed e.g. in ‘Enron Corp was collapsing and top brass pulled out ... money before it went bankrupt’ (Cropper, 2004). However, recombination and re-accentuation of metaphoric models in discourse is limited by personal and social cognition, i.e. the mental models in individual episodic memory and shared social representations that structure social identities, roles and relationships. Metaphor is an ideal entry into investigating these and other links between cognition and discourse as its double nature as conceptual model and linguistic expression locates it at the interface between the two. Discourse is here defined as language use as social practice, operating at the levels of text, textually mediated interaction and socio-economic context (Fairclough, 1995). A cognitive critical analysis of discourse posits that mental models and representations, including those structured metaphorically, are reflected in texts. In its ideational metafunction, metaphor helps construct particular aspects of reality and reproduce (or subvert) dominant schemas. The models of reality held and reinforced by groups with most power and discourse access then become quantitatively and qualitatively salient. In a cyclical fashion, discourse recipients are therefore more likely to encounter such hegemonic mental models underlying discourse and, given similar conditions of reception, such repeated exposure is likely to anchor and reinforce such models even further. To concretize the above theory, the following question informs the analysis: are metaphoric expressions that draw on the source domains of religion and politics systematically realized in corporate discourse? If we accept that metaphor is, at least to some extent, intertextually constituted (Eubanks, 2005) then its use and function in discourse needs to be ascertained by looking at a large number of texts produced by various members of a discourse community. The above research question therefore necessitates quantitative analysis as a first step (for a qualitative investigation of what form and function the respective metaphoric expressions take in concrete texts, see Koller forthcoming). The next section will elaborate on the corpus of mission statements used as data and on the methods of analysing it.
122 Veronika Koller
3 A quantitative approach to metaphor in discourse This study works with mission statements as one genre of corporate discourse (see Swales and Rogers, 1995; Isaksson, 2005). More specifically, it investigates the mission statements of the top and bottom fifty largest companies worldwide, as listed in the 2003 Fortune magazine Global 500, an annual ranking that enjoys considerable importance in the business world. As a purpose-built corpus, the collection comprises 29,043 words. This corpus was analysed with the help of the semantic tagger USAS, which, together with a keyword tool and a part-of-speech tagger, forms part of the web-based software suite Wmatrix (Rayson et al., 2004; Hardie et al., 2007), originally developed for the analysis of lexical, grammatical and non-metaphoric semantic patterns in text corpora.6 Utilizing semantic annotation software to analyse the use of metaphor in discourse represents a new direction in computer-assisted research into metaphor in authentic, naturally occurring text. A number of researchers have started to use electronic corpora for the investigation of metaphoric patterns (Charteris-Black, 2004; Koller, 2004; Musolff, 2004; Deignan, 2005; Semino, 2005; Stefanowitsch, 2005; Skorczynska and Deignan, 2006); however, these studies, although differing with regard to genre, research focus and micro-methodology, all employ a lexical approach based on concordances of lexical items and word fields. This helps identify linguistic patterns, such as part-of-speech patterns in metaphoric expressions in particular discourses, but restricts results to finding further instances of previously identified expressions. Given that the identification of metaphorical expressions in texts has not been successfully automated, despite some promising attempts (e.g. Barnden et al., 2003; Mason, 2004), the method used in this study represents a half-automated approach. This is employed to search for expressions belonging to semantic fields that are likely to correspond to metaphor source domains. Hence, it can help identify potential metaphorical expressions in large data sets without being restricted to a pre-established set of lexical items. To elaborate, the USAS tagger operates with pre-defined semantic domains that are stored in a manually compiled and edited lexicon of single grammatical words and multi-word units. These domains are assigned to the tokens found in a text collection, generating lists of types tagged for a particular domain, which are then compared to an external reference corpus. This can be the British National Corpus (BNC), a part thereof or a purpose-built corpus uploaded into Wmatrix.
Metaphors in Corporate Discourse
123
Figure 8.2 Key concepts in the mission statement corpus.
Such a comparison allows for ascertaining the statistical significance in terms of log-likelihood (LL) for a given semantic domain, the cut-off point being 6.63 for 99% significance. Figure 8.2 is a screen shot for key concepts, i.e. significant semantic domains, in the mission statement corpus as compared to the BNC Sampler Written (968,267 words). It is important to note that a word can have more than one semantic tag, with the tags ranked in order of decreasing likelihood in general English, as based on corpus evidence. At the time of writing (May 2007), the lexicon comprises 55,009 single grammatical words, while the list of multi-word units includes 18,922 items. This allows 92% accuracy of semantic annotation, alongside 97% to 98% accuracy of part-of-speech tagging (for details see Piao et al., 2005). For metaphor research, this means that metaphorically used words in the corpus can be assigned either their source or their target domain as first tag. For instance, the conventional metaphoric expression ‘brand name’ will be allocated ‘business (selling)’ and hence the target domain as its first tag. Indeed, the hypothesis that justifies using a semantic corpus tool for metaphor analysis which was not designed for investigating metaphor is precisely this: semantic annotation software like the USAS tagger
124 Veronika Koller
can help identify metaphor in discourse in that (a) semantic domains roughly correspond to conceptual domains drawn on in metaphor formation, and (b) more conventional metaphoric expressions are likely to be assigned their target domain as first tag and their source domain as a second or further tag, or not at all. Novel expressions by contrast would be assigned the source domain as their first, or even only, tag. In order to adapt the software for the identification and analysis of metaphor in discourse, a metaphor version of Wmatrix incorporates two extra features, namely domain-push and broad-sweep functions. Normally, a search for a particular domain tag in Wmatrix will only yield results if that tag occurs first in the list of domain tags allocated to a word. However, the domain-push function prioritizes particular domains that are searched for, so that the respective tag, even if normally a second or further tag, will be listed first. This makes it easier to filter out its occurrences, saving the researcher from having to trawl through whole tag strings. However, if the aim is to get a comprehensive picture about where in the list a relevant tag occurs – as it is in this study – the broad-sweep search lists all types in the corpus which have that tag anywhere in the list of tags. Under the second function then, results will not only be single domain tags occurring in, or being pushed to, first position, but rather, the broad-sweep search yields results of whole tag strings, including the tag that is searched for. In pilot studies such as the one presented below, it has proved more fruitful to search for source-domain rather than target-domain tags. In its current, adapted version, the USAS tool still leaves open some questions, most notably about the relations between metaphorically used words and part-of-speech, as well as about polysemy. As for the latter, nouns like ‘mass’ or ‘capital’ are allocated various domains; in the case of ‘mass’ these are, in this order, ‘quantities: many/much’, ‘weight: heavy’, ‘religion and the supernatural’, ‘belonging to a group’ and ‘disease’, while ‘capital’ is tagged as ‘money and pay’, ‘places/ government’ and ‘paper documents and writing’. The different meanings have to be prised apart manually, by looking at the respective concordance lines. Further, it is not yet clear how semantic tagging can account for multiple metaphoricity due to semantic shifts, e.g. in the case of ‘corporation’, which uses a BODY source domain to conceptualize a political entity that is then in turn used to conceptualize a business organization. In this context, it should be noted that a pilot study carried out on a corpus of 42,000 words of articles on marketing taken from Business Week magazine showed that having the source domain as secondary tag not only indicates conventionality
Metaphors in Corporate Discourse
125
of a metaphoric expression, but also polysemy or a comparatively loose semantic link (e.g. ‘blood’ being assigned a secondary tag ‘war’). However, a corpus analysis using USAS also has certain advantages over concordance searches based on word fields: Broad-sweep searches of all tags capture a much wider range of potentially metaphoric uses, even if they do so at the cost of more ‘noise’. Also, the software allows searches for portmanteau tags, which automatically include various subcategories. Because any analysis relies on the underlying lexicon, however, the quantitative analysis will have to be complemented by a qualitative investigation that looks at the form and function of metaphoric expressions in concrete sample texts that show a high number of relevant metaphoric occurrences. For reasons of space, the following section will be restricted to results from the quantitative analysis of the mission statement corpus.
4 Evidence for religious and political metaphors in corporate discourse Due to the fact that the USAS semantic tagger can allocate first and secondary semantic domains to types in a corpus, and because these are theorized to roughly correspond to conceptual domains, it is possible to search for instances of metaphoric usage starting either from the target or the source domain. The following analysis will briefly show how starting from the target domain is less than fruitful, before going on to present the results from the other search direction, starting with source-domain tags. Searching for the key semantic domains in the mission statement corpus as compared to the BNC Sampler Written yields the following top ten domains and their five most prominent single grammatical words (see also Figure 8.2): ●
●
●
●
●
Business: Generally (LL 1223.18): ‘business’ (156 occurrences), ‘company’ (134), ‘companies’ (52), ‘businesses’ (43), ‘corporation’ (21) Business: Selling (LL 1163.63): ‘customers’ (187), ‘customer’ (64), ‘markets’ (38), ‘market’ (31), ‘clients’ (26) Helping (LL 538.99): ‘services’ (103), ‘service’ (43), ‘help’ (27), ‘benefit’ (25), ‘serve’ (25) Belonging to a group (LL 474.40): ‘group’ (95), ‘corporate’ (64), ‘society’ (56), ‘communities’ (32), ‘members’ (20) Important (LL 294.28): ‘values’ (81), ‘value’ (75), ‘key’ (49), ‘major’ (16), ‘important’ (15)
126 ●
●
●
●
●
Veronika Koller
Green issues (LL 236.91): ‘environment’ (42), ‘environmental’ (21), ‘environments’ (5), ‘nature’ (4), ‘environmentally’ (4) Interested/excited/energetic (LL 234.20): ‘energy’ (43),7 ‘commitment’ (37), ‘commitments’ (18), ‘actively’ (8), ‘passion’ (8) Money and pay (LL 201.78): ‘shareholders’ (40), ‘stakeholders’ (24), ‘assets’ (18), ‘investment’ (14), ‘banking’ (13) Science and technology in general (LL 197.66): ‘technology’ (40), ‘technologies’ (28), ‘GM’ [General Motors] (21), ‘technological’ (8), ‘chemistry’ (5)8 Evaluation: Good/bad (LL 178.50): ‘quality’ (50), ‘standards’ (22), ‘standard’ (7), ‘qualities’ (3), ‘evaluating’ (2).
The above list shows that semantic domains become, for the most part, significant because of a few highly frequent types. It also gives a flavour of the central topics, attitudes and self-image projected through corporate discourse in general, and mission statements in particular. However, it does not yet provide any pointers regarding the use of religious and political metaphors (with the possible exception of ‘passion’, see below). It should also be noted that neither domain comes up as significant in the list of key domains as a whole. This is not surprising, however, as any metaphor source domain would have to be utilized on a massive scale in order to come up as significant when compared to a larger corpus. This is rarely the case for complex derived metaphors. On the whole, analysing target-domain tags proves time-consuming, involving extensive manual reworking of concordance lines while yielding very few valuable results. Reversing direction, then, means looking at the semantic domains ‘religion and the supernatural’ and ‘government and politics’, which, given the genre and topic of the corpus, are likely to be source domains. The lists for types with a relevant first tag read as follows: spirit (21); spirits (3); wizard, religion, soul (2 each); nt,9 religious, elf, new_age, spiritually, shrine, sacred (1 each) country (7); government (6); governance, revenues, nation (5 each); states, nationality (4 each); governments (3); official, revenue, constituents, civil, nations, municipal, public_sector (2 each); state, non-government, deregulation, bureaucratic/bureaucracy, officials, authority, council, political/politics, candidates (1 each). Even a cursory glance at the ‘religion and the supernatural’ list shows that automated semantic tagging is not without its pitfalls. Thus, a
Metaphors in Corporate Discourse
127
look at the respective concordance lines corroborates that ‘wizard’ is obviously a technical term denoting an online help program, while ‘elf’ is part of the proper name of energy provider Elf Acquitaine. With these clarifications in place, the extended concordances exemplify the following metaphoric usages: 1. ‘They are an intrinsic part of our nature and key to fulfilling our vision for HP: to be “a winning e-company with a shining soul.” ’ 2. ‘Our customers and patients are our most important constituents. Our vision for the New Schering-Plough clearly sets forth what we will do as a global team.’ 3. ‘The people who work in our corporate areas ... spend their time identifying and sharing best practices across the company, performing necessary corporate governance roles, ensuring rigor in key decisions and assessing financial risks.’ While example (3) includes the collocation ‘corporate governance’ as a technical term, the co-text for (1) and (2) shows how one source domain can be elaborated through clustering (‘vision’ and ‘soul’) and how expressions from two different source domains can be combined (‘constituents’ and ‘vision’). Again, the above lists can be extended to capture words and multi-word units that have been allocated one of the two source semantic domains as secondary or further tags. The broad-sweep search for the secondary tag ‘religion and the supernatural’ yields the following additional items:10 services (103); mission (45); service (43); deliver (35); vision (33); highest (26); believe (18); high (15); delivering (11); offices, presence, passion (8 each); higher, communicate, beliefs (6 each); chapters, belief (5 each); delivers (4); office, delivery, host, creed (3 each); fortune, possess, faith, save, delivered, font, call/s, chapter, believes (2 each); converted, acts, cell/s, celebrate, living, saved/saves/saving, communicate/s, calls, vow, mediums, meeting, father, believed (1 each).
●
Singling out ‘faith’, ‘creed’, ‘belief*/believ*’, ‘vision’, ‘passion’ and ‘mission’, the following picture emerges from the concordances: ●
non-metaphoric items: ‘belief*/believ*’ realizes only its first tag ‘thought, belief’;
128 Veronika Koller ●
●
● ●
items used as hyperbole, not necessarily metaphoric: ‘passion’, tagged in this order as ‘interested/excited/energetic’, ‘relationship: intimacy and sex’, ‘religion and the supernatural’ and ‘sad’ realizes only its first tag; terms arguably used metaphorically: ‘faith’, as in ‘to have faith in the company’; metaphoric technical terms: ‘vision’ and ‘mission’;11 novel metaphoric terms: ‘creed’, as in the coinage ‘corporate creed’ (which in phonetic terms seems perilously close to the more established collocation ‘corporate greed’).
Analysis of the types that have been annotated with the portmanteau tag ‘government and politics’ as secondary or further semantic domain provide similar, if less differentiated, results. The list is as follows:12 service (43); corporation (21); benefit (16); benefits (14); private, capital (8 each); care, center (6 each); return, centers, fiscal, corporations (5 each); demonstrated, returns, stand, divisions (4 each); demonstrates/ ing, regulations, officer, executive/s, approved (3 each); demonstrate, agencies, run, stands, green, parties, duties, welfare (2 each); commission, tap, assessment, budget, agency, running, centre/s, standing, blue, registration, division, approval, establishment (1 each).
●
The following words were identified as possible metaphoric expressions: ‘officer’, ‘corporation’, ‘parties’ and ‘welfare’. After analysing the respective concordances these can be seen to pattern as follows: ●
● ●
●
●
non-metaphoric items: ‘parties’ realizes its first tag ‘participation’, its third-ranked cross-classification as ‘participation/people’ and its fourth tag ‘belonging to a group’ (e.g. ‘third parties’), but not its second tag ‘politics’; items used as hyperbole, not necessarily metaphorically: none; terms arguably used metaphorically: ‘welfare’, as in ‘the welfare of all or constituents’ (a political metaphor elaborated through clustering); metaphoric technical terms: ‘officer’ as part of ‘chief executive officer’, which shades over into the adjacent domain of warfare, its first tag; ‘corporation’ (see above); novel metaphoric terms: none.
It is noteworthy that in contrast to metaphoric expressions drawing on religion, those traceable to politics are neither used for hyperbole nor
Metaphors in Corporate Discourse
129
for creative, novel metaphoric coinages. The former may be due to the greater emotional potential of religion, as reflected in multiple semantic domains of emotion and religion for some items. The lack of creative usage, on the other hand, is perhaps best explained by the relative conceptual proximity of politics and business: Historically, political entities arguably developed from religious institutions while business organizations were modelled on political entities.13 This relative conceptual proximity between politics and business reduces the tension in any metaphor combining the two domains and may thus prevent politics being used as a source domain in other than formulaic terms. To sum up, the semantic annotation software used in this study allocates multiple tags to types in a corpus, the assumption being that more conventional metaphoric expressions will receive their target domain as a first tag. This was corroborated by technical terms such as ‘corporation’, ‘mission’ and ‘vision’, all of which were seen to have the source domain as a second or further tag. Conversely, relatively creative metaphoric expressions like ‘soul’ and ‘constituents’ showed the source domain as first tag. This pattern is modified, however, by a smaller number of types which do not confirm the original assumption; thus, ‘creed’ has a first tag ‘thought, belief’ although its metaphoric use in corporate discourse can be regarded as relatively creative. On the other hand, ‘governance’, as part of the well-established collocation ‘corporate governance’ has the source domain of ‘government’ as its only tag. Here, semantic annotation reaches its limits; if the persons originally assembling and further supplementing the manually compiled lexicon underlying the tagger do not belong to a discourse community in which ‘governance’ is an entrenched metaphoric expression, they are likely to tag the word with its source rather than target domain. It was further shown that a search starting from target-domain tags is time-intensive but yields little of interest. Starting with source-domain tags and extending them by broad-sweep searches seems more promising. Admittedly, this method still requires manual reworking to filter out erroneous codings such as ‘elf’ (see above) and other ‘noise’. Also, it does not capture all metaphoric instances; ‘credo’, for instance, was used as a novel metaphoric expression in the corpus (‘NYK Line business credo’) but ended up as ‘unmatched’ because the type was not part of the lexicon. On the whole, however, concordancing broad-sweep searches of source-domain tagged types not only has the potential to identify more metaphoric expressions than a corpus analysis based on word fields but also helps establish patterns of metaphoric usage. Further studies could use the Wmatrix tool to investigate the relations between metaphoric
130 Veronika Koller
expressions and parts-of-speech, or to identify metonymic usages as highlighted by secondary tags (e.g. ‘mission’ being cross-classified as ‘religion and the supernatural’/’architecture, houses and buildings’, or ‘soul’ being allocated a third tag ‘people’). The final section will go beyond such methodological implications and discuss the results in relation to a critical cognitive approach to metaphor in discourse.
5 Corporations as religious and political institutions within the defining paradigm The above analysis has established that metaphoric expressions of religion and politics are used in one genre in corporate discourse, namely mission statements, and form patterns of usage there. Instances are predominantly formulaic, with relevant expressions occurring mostly in technical terms. This is unsurprising given that formulaic language is a feature of most corporate genres. Still, more creative, novel uses can also be ascertained. Further analysis would help to bolster up these findings, not only through in-depth qualitative text analysis, but also within the quantitative paradigm: given that different symbolic capital accrues to different sectors in different cultures, it would be interesting to compare cultural contexts in the form of corpus data for, say, US and Japanese firms. Further, the above results could usefully be compared against other salient, even significant source domains, e.g. ‘organism’. What do the above preliminary results signify, however? The use of historically defining paradigms as source domains for metaphor in corporate discourse reflects a symbolic universe or ‘a matrix of all social and individual meanings that determines the significance of all group and individual experiences’ (Fox and Fox, 2004: 6). The economy can be seen as the contemporary defining paradigm, and its discourse transports mental models from previous paradigms, the values of which still reverberate. Both religion and politics provide adequate knowledge structures in this context, representing as they do ‘collectively held vision[s] of a social order in which the authority of fundamental social values is never questioned’ (Fox and Fox, 2004: 137). Using such models as source domains to conceptualize the economy not only legitimizes business organizations as instantiations of the defining paradigm by drawing on and appropriating previous sources of power, but in addition, such metaphorizations provide added emotional value for employees by conceptually and discursively structuring the corporation as a community, thereby fostering stakeholder loyalty. Ultimately, any
Metaphors in Corporate Discourse
131
communicative purpose of corporate discourse, its genres and features, is seen to be ancillary to the material aim of corporations: to increase productivity and thereby profits. It would obviously be absurd to claim that every use of technical terms such as ‘mission statement’ or ‘corporate governance’ is a conscious ideological move on the part of text producers, meant to anchor and reinforce metaphoric models of the economy and one of its instantiations, the corporation, as the site of legitimate power. The question remains, however, how such metaphoric terms became entrenched in corporate discourse in the first place, and who is actually driving the reconceptualization of the economy as the new politics and the new religion. To answer these queries, further studies will have to look into the history and conditions of text production and distribution in a corporate context. This investigation has used quantitative methods to show that metaphors of religion and politics are used systematically in corporate discourse. By doing so, it has hopefully laid the foundation for future research into the cognitive, discursive and ultimately material phenomenon of corporate missions and business empires.
Notes 1. I would like to thank Andrew Hardie and Elena Semino for their helpful comments on an earlier version of this chapter. 2. http://www.nykline.co.jp/english/profile/philosophy/charter.htm, accessed 21 April 2007. 3. http://w w w.magna.com/magna/en/responsibilit y/constitution/pdf/ Corporate_constitution.pdf, accessed 19 April 2007. For an analysis of that text, see Koller (forthcoming). 4. The simile is here included as a surface realization of the conceptual metaphor RELIGIOUS INSTITUTIONS ARE COMPANIES. 5. The Tories improved in the polls in the second quarter of 2006, after the nomination of David Cameron as the new party leader, and indeed had their best results in a local English election since 1992 soon after (‘Labour suffers’, 2003). Interestingly enough, Marks & Spencer’s profits picked up by 35% around the same time, after a 26% slump the year before (Marks & Spencer, 2006). 6. http://www.comp.lancs.ac.uk/ucrel/wmatrix 7. The concordance lines show that of the 43 occurrences of ‘energy’, 38 relate to the energy industry, i.e. to the item’s secondary tag ‘science and technology in general’. 8. ‘Chemistry’ has a secondary tag ‘personal relationships: general’, which, however, is not instantiated in the corpus. 9. The question of what the New Testament (‘nt’) does in a collection of mission statements is clarified again by concordancing, which shows it to be down to an extra blank and no apostrophe in the contracted verb
132 Veronika Koller
10.
11.
12.
13.
form ‘don’t’. This should provide fresh ammunition for the defenders of traditional punctuation. Multi-word units comprising ‘high/er/est’, ‘service/s’ and ‘call’ (e.g. ‘high_quality’, ‘service_company’ or ‘call_centres’) were excluded from the list as they are obviously not candidates for relevant metaphoric expressions. Incidentally, the first tag for ‘vision’ (‘mental object: conceptual object’) encapsulates a target domain, while its remaining three tags (‘sensory sight’, ‘wanted’, ‘religion and the supernatural’) represent a cline from concrete to abstract that is reminiscent of prototypical cross-domain mappings from abstract to concrete. Further, the last of the four tags for ‘mission’ represents a metonymic sense, in that it cross-classifies the word as ‘religion and the supernatural’/’architecture, houses and buildings’. Multi-word units comprising ‘stand*’, ‘care’, ‘services’ and ‘division’ (e.g. ‘stands_out’, ‘care_industry’, ‘service_station’ or ‘division_company’) were excluded from the list as they are obviously not candidates for relevant metaphoric expressions. Noun compounds such as ‘capital_market’ were also disregarded. ‘Capital’ in the monetary sense is a metonymic expression, relating – as indeed does the ‘capital’ as a country’s main city – to the Latin caput (head [here: of cattle]) to transfer the sense of value from livestock to the currency paid for it. It is obvious, however, that such compounds are not realizations of political metaphors. In truth, the situation is of course not as clear-cut; for instance, medieval convents functioned as religious communities but also as political entities and economic units.
References Barnden, J.A., Glasbey, S.R., Lee, M.G. and Wallington, A.M. (2003).’Domaintranscending mappings in a system for metaphorical reasoning’, Conference Companion to the 10th Conference of the European Chapter of the Association for Computational Linguistics (pp. 57–61). Available at http://www.cs.bham. ac.uk/~jab/ATT-Meta/Papers/eacl03.pdf. (Accessed 9 May 2007). de Beaugrande, R. (2004).’Language, discourse, and cognition: retrospects and prospects’, in T. Virtanen (ed.), Approaches to Cognition through Text and Discourse (pp.17–31). Trends in Linguistics. Studies and Monographs 147 (Berlin: de Gruyter). BBC/British broadcasting Corporation (2003). ‘New archbishop enthroned’. Available at http://news.bbc.co.uk/1/hi/uk/2803139.stm. (Accessed 21 April 2007). BBC / British Broadcasting Corporation (2006). ‘Labour suffers local poll losses’ Available at http://news.bbc.co.uk/1/hi/uk_politics/4969812.stm. (Accessed 3 March 2007). Charteris-Black, J. (2004) Corpus Approaches to Critical Metaphor Analysis (Basingstoke: Palgrave Macmillan). Cropper, C.M. (2004).’Executive comp: on a shorter leash’, Business Week, 25 October. Available at: http://www.businessweek.com/magazine/content/ 04_43/c3905070_mz013.htm. (Accessed 20 April 2007).
Metaphors in Corporate Discourse
133
Deignan, A. (2005). Metaphor and Corpus Linguistics (Amsterdam: Benjamins). Eubanks, P. (2005). ‘Globalization, corporate rule, and blended worlds’, Metaphor and Symbol, 20, 3: 173–97. Fairclough, N. (1995). Critical Discourse Analysis (London: Longman). Fairclough, N. (2003). Analysing Discourse: Textual Analysis for Social Research (London: Routledge). Fox, R. and Fox, J. (2004). Organizational Discourse: a Language-Ideology-Power Perspective (Westport, CT: Praeger). Grady, J. (1997). ‘Foundations of meaning: primary metaphors and primary scenes’, Doctoral dissertation, University of California at Berkeley. Hardie, A., Koller, V., Rayson, P. and Semino, E. (2007). ‘Exploiting a semantic annotation tool for metaphor analysis’, Proceedings of the Corpus Linguistics 2007 Conference, 27–30 July, Birmingham, UK. Isaksson, M. (2005). ‘Ethos and pathos representations in mission statements: identifying virtues and emotions in an emerging business genre’. In A. Trosborg and P.E. Flyvholm Jørgersen (eds.), Business Discourse: Texts and Contexts (pp.111–38) (Bern: Peter Lang). Koch, J. (2001). Megaphilosphie: Das Freiheitsversprechen der Ökonomie (Göttingen: Steidl). Koller, V. (2004). Metaphor and Gender in Business Media Discourse: A Critical Cognitive Study (Basingstoke: Palgrave Macmillan). Koller, V. (forthcoming). ‘The integration of other social domains into corporate discourse: the case of political metaphors’. In H. Kelly-Holmes and G. Mautner (eds.), Language and the Market (Basingstoke: Palgrave Macmillan). Kristiansen, G. and Dirven, R. (eds.) (forthcoming). Cognitive Sociolinguistics: Language Variation, Cultural Models, Social Systems (Berlin: de Gruyter). Marks & Spencer (2006). ‘Five year group financial summary’. Available at http:// www2.marksandspencer.com/thecompany/investorrelations/downloads/2006/ five_year_summary.pdf. (Accessed 3 March 2007). Mason, Z. (2004). ‘CorMet: a computational, corpus-based conventional metaphor extraction system’, Computational Linguistics, 30, 1: 23–44. Mautner, G. (forthcoming). ‘The spread of corporate discourse to other social domains’. In H. Kelly-Holmes and G. Mautner (eds.), Language and the Market (Basingstoke: Palgrave). Musolff, A. (2004). Metaphor and Political Discourse (Basingstoke: Palgrave). Nelson, F. and Macdonnell, H. (2005). ‘Cook was poised for high office in Brown cabinet’, The Scotsman, 8 August. Available at http://thescotsman.scotsman. com/index.cfm?id=1743442005. (Accessed 20 April 2007). Nuyts, J. (2004). ‘The cognitive-pragmatic approach’, Intercultural Pragmatics, 1, 1: 135–49. O’Halloran, K. (2004). Critical Discourse Analysis and Language Cognition (Edinburgh: Edinburgh University Press). Piao, S.L., Archer, D., Mudraya, O., Rayson, P., Garside, R., McEnery, T. and Wilson, A. (2005). ‘A large semantic lexicon for corpus annotation’, Proceedings of the Corpus Linguistics 2005 Conference, July 14–17, Birmingham, UK. Available at http://www.comp.lancs.ac.uk/computing/users/paul/publications/cl2005_ estlex.pdf. (Accessed 9 May 2007). Rayson, P., Archer, D., Piao, S.L. and McEnery, T. (2004). ‘The UCREL semantic analysis system’, Proceedings of the Workshop on Beyond Named Entity Recognition
134 Veronika Koller Semantic Labelling for NLP Tasks in association with LREC 2004, Lisbon, Portugal, pp. 7–12. Seddon, M. (2005) .’How I was kippered by my party’, Guardian 16 March. Available at http://www.guardian.co.uk/comment/story/0,,1438496,00.html. (Accessed 3 March 2007). Semino, E. (2005). ‘The metaphorical construction of complex domains: the case of speech activity in English’, Metaphor and Symbol, 20, 1: 35–69. Skorczynska, H. and Deignan, A. (2006). ‘Readership and purpose in the choice of economics metaphors’, Metaphor and Symbol, 21, 2: 87–104. Spinnen, B. and Posner , E. (2005). KlarsichtHuellen: Ein Dialog über Sprache in der modernen Wirtschaft (Munich/Vienna: Hanser). Stefanowitsch, A. (2005). ‘The function of metaphor: developing a corpus-based perspective’. International Journal of Corpus Linguistics, 10, 2: 161–98. Swales, J. and Rogers, P.S. (1995). ‘Discourse and the projection of corporate culture: the Mission Statement’, Discourse & Society, 6, 2: 223–42. van Dijk, T.A. (2005). ‘Contextual knowledge management in discourse production: a CDA perspective’. In R. Wodak and P.A. Chilton (eds.), A New Agenda in (Critical) Discourse Analysis (pp. 71–100) (Amsterdam: Benjamins). van Dijk, T.A. (2006). ‘Discourse, context and cognition’, special issue of Discourse Studies, 8, 1: 159–77. Weber, M. (1930). The Protestant Ethic and the Spirit of Capitalism (first published 1920), trans. by T. Parsons (London: Unwin University Books). Wertheimer, M. (1920). Über Schlussprozesse im produktiven Denken (Berlin: de Gruyter).
9 How Business Press Headlines Get Their Message Across: A Different Perspective on Metaphor Michael White and Honesto Herrera
Introduction Form, function and actual practice in the use of headlines, on the one hand, and, on the other, growing research in the field warrant a consideration of press headlines as a discourse type in its own right. The chapter sets out to chart and analyse the use of metaphor in the sub-field of business press headlines with the added aim of exposing the most productive patterns within the field and how these foster communication. From the reader’s point of view, headlines reconcile the apparently contradictory in that they are recognized to be singularly complex and yet, at the same time, they exercise an equally singular appeal to that reader’s attention. On the basis of strongly supported empirical evidence, we propose to isolate the most productive metaphor patterns or mechanisms involved in business press headlines and then to go on and analyse such patterns and show how they serve to reconcile the above perplexing claim.1 Headlines are the shop window display of newspapers. As such they have a crucial role in gaining the reader’s attention and in convincing him or her to consume the news item available. Given newspaper history and sales figures, it would seem axiomatic that headlines are successful in those pursuits. A major issue for the linguist is to uncover the textual features specific to headlines which work towards such success. The present chapter focuses on what could be one significant contributor, namely, the role played by metaphor in constructing business press headlines. On the basis of empirical data and inspired by the insights of metaphor studies, our chapter analyses a single but particularly productive metaphor pattern, which has received little attention within the cognitive paradigm, aiming to uncover how it may be generating 135
136 Michael White and Honesto Herrera
or enhancing communicative potential. Our approach is also framed in cross-linguistic terms, as our evidence is drawn from both British and Spanish press sources, allowing us to consider the pattern in comparison and contrast, and in the process to highlight possible cultural specificities.
1
A different perspective on metaphor use
Mainstream research of an empirically driven or empirically informed type in field-specific discourses has tended to follow a methodology based on identifying source domains and then running electronic checks of the more salient lexical items of those domains. This procedure has led to strong, empirically supported findings in many fields and these findings have not only satisfied applied agendas but have also involved theoretical developments. Musolff (2004), for instance, has empirically established metaphor patterns in political discourse while Charteris-Black (2004, 2007) has shown reliable tendencies for metaphor use in fields such as business or the discourse on leadership. On the other hand, theoretic consequences can be drawn from Deignan’s (2005) evidence revealing how empirically driven metaphor research can show up linguistic claims based on introspection to be alien to real language use. Nevertheless, despite the fact that such studies are placing research on quantitative underpinnings to extents undreamt of in the not too distant past, at the same time, all these authors are aware of intrinsic limitations in machine searching, such as the fact that the search headwords will necessarily condition the results that are going to be obtained. That said, however, the major source domains empirically shown by the above approach to be prevalent in business discourse – WAR, COMPETITIVE GAMES, LIVING ORGANISM, MARRIAGE, METEOROLOGY, NATURAL DISASTERS or the whole VERTICALITY framework associable with the MORE IS UP metaphor – are more than likely to surface in any extensive study of business discourse. Our study sets out from a different perspective since, in the type of headline we are going to focus on, motivation springs from within the topic or subject matter under discussion in each individual article, rather than from those mainstream domains mentioned above. This process is so frequent that we can consider it to form a pattern and the most salient feature of that pattern is that it is characterized by the condition whereby lexis proper to the semantic field of the subject matter under discussion is deployed figuratively. The point is that when any subject whatsoever is under discussion, it is axiomatic that lexis from
Business Headlines & Metaphor
137
that field will appear under a normal denotative guise. For example, if we are talking about cars, a whole array of lexis from that field, from the most immediately concrete to the more abstract or peripheral, will arise: e.g., engines, wheels, gears, oil, petrol, travelling, speed, tuning, etc. What we find in newspaper headlines is that such lexis does appear, but it does so with a meaning shift, operating in a figurative sense. We are calling this type of practice interface because we feel that there is a leakage from mainstream denotative to figurative senses.2 The ease in understanding this kind of discourse type should not blind us as to the relatively complex mechanisms involved. Arising in contexts thematically natural to it, the type of lexis used in this way raises literal expectations and these provide, as it were, a push-button interface system of access to a figurative dimension. Put in other words: the denotative lexis of the particular subject matter under discussion provides the linguistic metaphor that endows the headline with figurative meaning. Hence, rather than an overarching source domain, each particular subject matter being dealt with is what primarily motivates the figurative use in each case and provides the source lexis for the linguistic metaphors to be used. This idea of motivation strongly underlies the terminology used by other researchers to refer to what we term interface, namely ‘double grounding’ which they see as involving ‘the simultaneous activation of structure from both the source and the target of metaphor in combination with an implied metaphoric projection’ (Brône and Feyaerts, 2005: 75). We will now deal with this pattern in detail from different angles, particularly examining its structural make-up, its quantitative incidence and its communicative impact.
2 Structural analysis From the structural point of view, the classical source target domain of the conceptual theory of metaphor is still in place in the interface pattern but, as we have pointed out above, there is an added prerequisite, namely, the necessary condition that lexis which is literally mainstream lexis from the subject matter under discussion is the lexis that is deployed figuratively. This factor also triggers further structural consequences according to our analysis. Thus, the core lexical item, no longer employed in its core sense but in a figurative one, first operates as a metonymic trigger in the source domain, acting in that domain as a salient feature that activates the whole domain. Subsequently, it then operates metaphorically in the target domain. Thus, the interface headline constitutes a discourse type in which metonymy and metaphor
138
Michael White and Honesto Herrera SOURCE DOMAIN
TARGET DOMAIN
X DOMAIN …..… .......
..........................
…….. …… ………………..
Figure 9.1 Empty schematic framework structure applicable to all interface examples.
characteristically operate inseparably. A skeletal outline of this process showing a formalization of the schema involved is presented in Figure 9.1 and will be fleshed out below in Sections 4 and 5.
3 Quantitative incidence of figurative language and of the interface pattern in business press headlines The empirical basis for the present study was elaborated by hand searching for metaphor use one issue per week of The Financial Times, El País and El Mundo during the first six months of 2003, restricting selection to headlines on business issues. Yet, the task is not as straightforward as it might seem since, among other things ‘[t]he identification of metaphor is notoriously subjective’ (Skorczynska and Deignan 2006: 91). At the very outset, choices have to be made. In our case, as a guiding principle for discerning figurative versus literal use, we adopted the strategy of categorizing as figurative both metonymy and metaphor and in general all uses that could be very broadly considered as such. Therefore, possible metaphoric instances where a high degree of conventionalization might indicate lexicalization within particular fields are still considered figurative. Categorization of what counts as metaphor is therefore rather inclusive and we adopted this strategy on the grounds that ‘however stone dead such metaphors seem, we can easily wake them up’ (Richards, 1936: 101). This analysis showed up a marked bias in favour of figurative language with an average of 76%
Business Headlines & Metaphor
Table 9.1
139
Headlines and interface patterns Headlines
Interface
Financial Times El País El Mundo
362 348 164
96 55 21
TOTAL
874
176
of headlines being figurative and only 24% literal – data which of themselves supply convincing reasons to study this type of discourse manifestation. A qualitative analysis of such data will be undertaken in Section 6. Having first set out to discover the most productive metaphoric source domains in business press headlines, serendipitously, the incidence of the type of structure we are calling interface powerfully suggested itself as a productive pattern in its own right, warranting quantitative and qualitative analysis. The quantitative incidence of the literal–figurative interface as a media strategy can be seen in Table 9.1. The resulting figures are then subjected to a chi-square test obtaining χ2 = 10.337, p < 0.01. This value allows us to claim a relationship between the interface pattern and the different newspapers. Once we have established that there is a significant association between newspapers and the pattern studied, we need to know which cell or cells contribute to this relation and in what way they do so. This is achieved by applying the Standardized Pearson Residuals test. This test will enable us to know whether this pattern is independent from some of the newspapers where it appears, or, on the contrary, whether it is associated with some or with all of them.3 Values above 1.96, regardless of the 1/2 sign, will prove significant association and that is precisely what happens with the interface pattern: Financial Times (3.17); El País (22.12); El Mundo (22.30). On the basis of these data we are entitled to claim that there is a strong relationship in all cases. In the case of the Financial Times we find a very significant association between the use of the interface pattern and this newspaper, i.e. there are more interface patterns than expected by chance; whereas the negative values in the Spanish press also show a significant association, but the other way round, as the interface incidence is lower than what would be expected by chance.
140
Michael White and Honesto Herrera
4. Cases of literal–figurative interface in business headlines in English Consider the following: (1) Manchester United boasts three-pronged attack (Financial Times, 26 April 2003) Here the term ‘attack’, while originating in the domain of war, has become so domiciled in football discourse as to be considered core lexis in that field. This lexical item now operates as a salient feature, metonymically activating the whole source domain of football. At the same time, the expression itself is metaphorically deployed, crossing from the domain of football, in its core sense of sports, to the domain of football club in the sense of a business entity. Hence the ‘three pronged attack’, which in its core sense would actually be referring to a strategic alignment of a soccer team’s forwards on the playing pitch,4 actually refers to a combination of three different income sources for the club (namely, match day sales, commercials and media income streams). Thus core football lexis for a successful outcome on the playing field is deployed figuratively to capture the idea of a business strategy for the success of a football club as a business entity. The cognitive processes involved in such a strategy are outlined graphically in Figure 9.2. The following case is yet another realization of identical structure while at the same time showing that different lexical items of the same
SOURCE DOMAIN
TARGET DOMAIN
FOOTBALL DOMAIN Players …..… attack / defence defence supporters competition rules win / lose penalty etc.
Figure 9.2.
Interface in football domain
Football club as business entity
Business Headlines & Metaphor
141
semantic field of the subject matter under discussion (football) are available for figurative use: (2) Manchester United play Cazanove at the heart of defence (Financial Times, 26 April 2003) Again, another salient feature of football, this time, ‘defence’, metonymically activates the whole source domain and, in turn, carries over metaphorically to the target domain of the club as a business entity, establishing, in this case, the correspondence of the key role of defence on the playing field with that of a renowned broker – Cazanove – in the domain of the business performance of the club. Moreover, the additional presence of lexical items which would naturally collocate with the literal use of defence on the playing field, such as ‘play’ and ‘heart of’ bolster the sensation of the literal and denotative use of defence while nevertheless being deployed figuratively. The end result is still greater cohesion and coherence in the headline, orchestrated by the literal–figurative interface technique. (See the Appendix to this chapter for a sample of key exemplifications of this same pattern.) These two examples are quite straightforward but, in fact, the underlying semantic connections can be extremely complex. Baicchi (2003: 320, 329–31) distinguishes complexity, as proper to the input of the producer, from difficulty, as proper to the receptor as processor, and she goes on to point out that one of the factors giving rise to complexity is the quantitative dimension deriving from the number of different input items involved. A close analysis of some headlines yields surprising results in this respect. For example, we consider the following interface instance to provide evidence of a complex multi-layered series of connections where the sheer number of input items is, without doubt, staggering: (3) US grapes of wrath pale beside Bordeaux’s real ills (Financial Times, 26 April 2003) Brône and Feyaerts (2005) claim that for the type of headline we are analysing, a blending analysis is preferable to the two-domain framework of Conceptual Metaphor Theory. This we feel is certainly the case in the more complex instances, such as the one at hand. The context here is the boycott in the US of French products, particularly wine (reinforced pictorially with American citizens emptying bottles of French wine into a river), in the wake of French policy opposition to the invasion of Iraq.
142
Michael White and Honesto Herrera
A blending model analysis could start with the input space of ‘grapes’ as metonymic trigger for wine, followed by ‘Bordeaux’ as twin metonymic trigger, activating Bordeaux wine in the first instance and French wine in general in the second. Then we have the composite expression ‘grapes of wrath’ activating the rich cultural scenario associated with Steinbeck’s novel and the classic movie it inspired. Furthermore, ‘wrath’, taken alone, has its own proper scenario, activating a religious domain with the Bible and hymn singing as components.5 In this scenario ‘wrath’ is not just any kind of anger, but divine anger. The latter is a justified or righteous anger: presumably God does not manifest anger arbitrarily and, hence, ‘wrath’ is the type of anger proceeding from the response of God to the crass failings of mankind (on religious metaphors for economy, see Koller, this volume). Finally, a further connection is sensed to obtain from the co-occurrence of ‘pale’ and ‘ills’, salient features in a health domain and therefore also calling up the conceptual metaphor BUSINESS IS HEALTH. The headline thus orchestrates a network of overlapping, interlocking connections and these connections cannot but reinforce its communicative impact. All these factors and how they can be seen to combine in a blending analysis are set out graphically in Figure 9.3. Because of their position, initial deployment headlines typically trigger cataphoric relationships. That means that there is a lack of prior co-textual referents and this obviously entails added difficulties. Nevertheless, it is also important to note that news reporting is an ongoing process: frequently, it takes place in a context that is shared to a high degree by the readers. In this respect, it is also interesting to consider that there are exophoric and indeed anaphoric connections available to the readers, not in the co-text, given the position ‘initial’ nature of headlines, but in the context generally. In the case of example (3), at the time the headline appeared, we can assume that Financial Times readers were highly aware of the US–French tension due to their differences on Iraq and that this tension is being overtly manifested in the US by an attempted boycott of French wine. The numerous items which the headline writer has managed to squeeze into this headline are evidence of Baicchi’s (2003) notion of complexity, mentioned above. The question is then: does this complexity also imply difficulty for the reader? In our analysis, not necessarily, the reason being that despite the complexity involved, the headline creates a frame that facilitates understanding. As Dor (2003) points out regarding tabloid headlines, these, frequently, do not provide a summary of ensuing information and, at the same time, they use puzzling and deformed syntax and lexis. Nevertheless,
US grapes of wrath pale beside Bordeaux’s real ills Generic space
Entities Relations
WRATH Input 2
Input 1 Wrath of God
Grapes of wrath
Bible Input 4
Input 3 Social sins
Human sins
Lack of American wine competitiveness French anti-Iraq invasion policy
EFFECTS
Input 6 Anti-French sentiment
Input 5 Indignation
Extermination
American patriotism
Swinging his sickle Trampling Treading on
Grapes wine Bordeaux Bordeaux wine French wine French products
Input 7 Hymn singing: blending Grapes and Wrath Input 8 American depression: Novel and film = hardship BLENDING
Figure 9.3
Grapes of wrath American indignation Boycott Difficult times for wine industry
Complex multi-layered connections in grapes of wrath metaphor.
boycott
Health pale ill
144
Michael White and Honesto Herrera
even as stand-alone texts, such headlines turn out to be easy for the reader to understand and this, according to him, is precisely because they provide and exploit frames which are culturally very familiar. The frame in which example (3) is couched, including the cultural references (i.e., the allusion to the whole scenario activated by ‘grapes of wrath’), has important ideological consequences. In fact, because of the connections made, this ideological component probably filters in with the reader unaware – who, more than likely, is favourably predisposed towards acceptance as a result of the effects of shared culture (HerreraSoler 2006a, b). Thus, for instance, the ideological frame that comes with the biblical scenario activating the idea of righteous anger is easily carried over to the US anger towards France which, because of the connection, takes on a similar righteous halo. Something akin happens in the case of the football examples. Football fans will have interiorized the costly consequences of a weak or defective defence or attack in their team’s performance and probably harbour in their minds dramatic real instances of such events. Therefore, the need for a strong defence and attack are accepted as unquestionable dogma or as a sine qua non for their football schema. Applying these concepts to the performance of football clubs as business entities will endow these (metaphoric) strategies with equal legitimacy in the business arena as those applying in the literal strategies on the playing field. Thus, because of shared schemas, the ideology involved will tend to be accepted unquestionably and practically automatically (White and Herrera, 2003; Musolff, 2004; Koller, 2004, Velasco-Sacristán and Fuertes Olivera, 2006; CharterisBlack, this volume).
5
Literal–figurative interface in Spanish evidence
The Spanish interface evidence shows the same structural patterns seen above for English. We will focus our analysis on a few examples involving rich cultural components. (4) ZARA SE ARRUGA EN LA BOLSA (Zara wrinkles up on the stock market) Wrinkling being a typical property of clothing, we again have a situation where the subject matter concerned is being tapped for figurative use. Furthermore, many concurrent semantic and pragmatic factors are simultaneously operating here, so much so that again we need to adopt the multiple-space approach of Blending Theory to handle them. As
Business Headlines & Metaphor
145
well as the literal meaning of ‘wrinkle’ in the clothing domain, arrugarse has a very expressive entrenched idiomatic meaning, deriving from a variety of polysemous senses:
arrugarse
再
literal
⇒
wrinkle
figurative ⇒ - to be intimidated, to chicken out, to be wanting in courage, not be up to a demanding situation, fail under pressure
As can easily be deduced from the nature of the figurative senses, arrugarse is a likely candidate for use in communicative circumstances of significant experiential import and this fact adds to its communicative potential. Thus, while in line with our earlier analysis, it is clear that the choice of term – arrugarse – has been motivated by the connection with the clothing domain, yet the sheer force of the idiomatic senses means both spaces are immediately activated. Rich as this is, it still does not account for all the frame components for the Spanish reader who would be highly aware of a further specifically cultural scenario. Historically, another Spanish textile company’s publicity campaign in which its use of natural fabrics was metonymically highlighted by the slogan ‘La arruga es bella’ (‘wrinkles are beautiful’)6 was so successful as to become emblematic and hence enter general encyclopaedic knowledge in Spain, a fact providing yet another element of shared knowledge amongst readers. All these factors contribute to ease of understanding and extra communicative impact and additionally show that while the interface pattern originates in a literal aspect of the subject matter, the pattern is far from merely being a mechanical application of that structural pattern but may take on board very diverse semantic and pragmatic components. The following interface example likewise belongs to the culturally filtered category, underlining the necessity for specific cultural knowledge to fully grasp the meaning: (5) ARGAL DA LA VUELTA AL CURADO (Argal turns around the cured (ham)) This headline selects a step in the process of artisan cold-curing of ham to talk about the business performance of the company Argal. The structural schema is identical to that of the former examples. In this case, a phraseological element from the subject matter – turning around the
146
Michael White and Honesto Herrera
ham – metonymically activates the whole company activity and then operates metaphorically with the source concept of the turning around of the leg of ham (in the literal sphere) corresponding to the target concept of ‘turning around’ the company’s business performance from a period of losses to a return to profit. Additionally, the ingenuity of the shift from literal ‘turn around’ to its multi-word idiomatic use is remarkable. The high profile of the culturally specific elements deriving from traditional ham-curing lore, or from the publicity world, in the foregoing examples does not mean that all the Spanish evidence falls into this category (see the Appendix to this chapter for somewhat more prosaic examples). What the foregoing kind of example does show, however, is how headlines which spring from subject-matter motivation may attain considerable complexity by the integration of additional culturally filtered elements and in that way produce headlines which the reader finds particularly satisfying, a point which will be dealt with in the following section.
6
Communicative impact
One recent insightful study of headlines exposes the paradoxical situation that while the ‘importance of the role of headlines in the communicative act performed by newspapers can hardly be exaggerated, yet the nature of this role has virtually never been explicated in the literature’ (Dor, 2003: 695). Dor’s approach to the question is driven by the principle of what he calls ‘relevance optimization’: attempting to obtain ‘the greatest possible cognitive effect for the smallest processing effort’ (Dor, 2003: 699). Unfortunately, he does not consider the type of headline we are analysing here, where processing effort may be considerable but where the cognitive effect would be greatly enhanced. The crucial issue for our chapter is if, and if so how, the interface pattern as outlined above bears on communication. An affirmative answer to the first ‘if’ could be supported simply by incidence as proof. That is, independently of whether we consider that the functions of news headlines are to summarize, attract, encourage to read on and persuade (Bell, 1991 and Van Dijk, 1985) or whether we consider them as an end in themselves for most readers (Dor 2003: 717–19), the fact that the interface pattern does appear in actual headline use in such a systematic and persistent way must imply that it is a net contributor to communicative effect. The second aspect – how metaphor in general and the interface pattern in particular bears on communication – is much more problematic and multi-layered. Let us approach this crucial issue by first looking
Business Headlines & Metaphor
147
at elements of headlines that are problematic for communication or may apparently run counter to it because of the high processing effort they demand. Generally recognized headline requirements, for example, Dor’s (2003: 700–1) first two, namely, their being ‘as short as possible’ and their being ‘clear, easy to understand and unambiguous’, often clash with the complexity, the puzzling nature, or the unorthodox syntax of many headlines. In the face of these latter aspects, metaphor in general and particularly the interface pattern can play important roles as facilitators. Conceptual metaphors, as has been demonstrated within the Cognitive Linguistics paradigm, are limited in number though their linguistic realizations may be legion. Furthermore, the facts that they are so pervasive and that they are grounded on conventionalization result in the establishment of over-arching frames which facilitate communication. Additionally, their contribution to ‘vividness’ and ‘compactness’ or solving the ‘otherwise inexpressible’ (Gibbs, 1994: 124–6) means they are particularly appropriate for a text type such as a headline which is singled out for focus, may stand alone or point forward to cataphoric resolution. The interface headline benefits from all these advantages while providing the added extras proper to its own make up. For instance, one of the tolls to be paid by titles in general and headlines in particular is the fact that they lack the great textual facilitator, anaphoric reference, and, as Baicchi (2003: 319–21) points out, they typically operate by ‘point[ing] forward to their text base for solution and direct[ing] the text receiver toward the reading process with a puzzling task: that of seeking the co-referent items and of matching them with the cataphoric co-referents contained in the title’. Our claim is that the interface pattern can mitigate this toll, providing, as it were, ‘two for the price of one’ in the sense that the intra-headline connections which they automatically establish are already a form of co-referent connection prior to any cataphoric search in the ensuing text. This is a major contribution of the interface pattern. Kövecses (2005: 96, 237) identifies the role of the ‘pressure of coherence’ in generating metaphorical source domain in the type of headline we are calling interface. We must also add the benefits of that coherence. For instance, our foregoing exemplification and analyses show how the coherence established by the rich culturally filtered frames and the interface interconnectivity trigger textual synergies (Baicchi, 2003; White, 2006) which greatly facilitate comprehension of the headline text, even as a stand-alone entity. Finally, in the case of the business press, habitual readers probably count on
148
Michael White and Honesto Herrera
substantial background knowledge of events shared with the journalist. This can mean exophoric and indeed anaphoric access to the milieu that has engendered the headline, thus tapping into default knowledge. In this way, readers may look backwards for the information they have on the issue to uncover intended meaning. At the same time journalists, by resorting to shared knowledge, are in fact operating with prominent, pervasive and persuasive ideas. The interconnected networks in shared ideas, in turn, facilitate the access and retrieval of the concepts expressed in the headline and in this way attract and influence readers, getting the headline message across and creating or reinforcing opinion. In short, our claim is that the interface pattern cashes in on a very effective interdependence of semantic, pragmatic and cognitive dimensions of figurative language and that this greatly enhances communicability: if processing effort is demanding, it is amply compensated for by cognitive effect. This gives a new vista on cognitive cost-effectiveness since added effect neutralizes greater difficulty.
7
Conclusion
This chapter has explored the use of a very productive figurative pattern, that may be labelled the literal–figurative interface, in British and Spanish business press headlines from the point of view of metaphor and metonymy use. It claims this pattern is structurally grounded on a metaphor–metonymy interconnection. It then goes on to tackle how the pattern contributes to communicative felicity. Communication is always a difficult issue to discuss in a rigorous systematic way, given the strong subjective elements inherent to its very nature. However, the significant incidence of the pattern and the strands of structural interconnection it is shown to wield provide reliable and quantifiable evidence. The combination of the literal and the figurative and the recycling of the literal as figurative trigger synergies (Baicchi, 2003; White, 2006), and such synergies are not merely the sum of each of the constituent components taken separately but result in a qualitative leap in communicative effectiveness when fused together as a combined unit. Giora et al. (2005) have shown experimentally that it is not the dramatically innovative metaphor which is judged most communicative; rather, the optimum effect seems to be produced where some conventional or easily accessible component is present but modified in some respect or other. This is just what happens in the literal–figurative interface type of headlines. Lexis, so conventional as to be saliently literal within the
Business Headlines & Metaphor
149
very field of the subject matter being discussed (football, cars, ham, wine, clothing, etc) is foregrounded by undergoing a figurative shift and this combination of the utterly conventional with its new figurative turn has an optimizing effect on communicative impact. Moreover, the wit (Brône and Feyaerts, 2005: 85) and ingenuity deployed in bringing about the interface connection raises reader awareness of the artifice involved. This, we feel, is a net contributor to optimization since not only is the underlying presence of the literal strongly sensed, but awareness is raised as to how it has been deftly mobilized for rhetorical effect. This results in a writer/reader collusion, in the sense that the writer’s consummate linguistic and rhetorical skill is overtly manifest while, in recognizing this fact, the reader can likewise reap personal satisfaction in that s/he has equally been up to the challenge in deciphering the journalist’s artifice. In so far as contrastive issues are concerned, while the widespread or near universal aspects in metaphor use are evident, the aspects specific to one or other culture are likewise shown to be highly sought by headline writers as net contributors to ease of understanding and communicative effect. Thus, while interface headlines may at times be rather complex and may be calling for considerable processing effort, all the factors analysed above offset these difficulties and favour communication. The end result is that the interface pattern is providing ‘the best (informational) value for cognitive money possible’ (Dor, 2003: 718).
Notes 1. The research for this chapter has in part been financed by the Spanish Ministry of Education, Ref:HUM2005–05957 and is developed from Herrera et al. (2006). 2. In earlier work, we had labelled it ‘interconnection’: see White et al. (2004). 3. It should be borne in mind that the Standardized Pearson Residuals test can be compared to the standard normal distributions, i.e. values above 1.96 indicate divergence from independence. 4. When Barcelona Football Club adopted this strategy persistently some years ago, this alignment was labelled ‘trident’. 5. The enigmatic meaning of ‘grapes of wrath’ when taken as a stand-alone combination, seems to have its origin in the Bible where we read ‘And the angel thrust in his sickle into the earth and gathered the vine of the earth, and cast it into the great winepress of the wrath of God’ (Revelation 14:18–20). This in turn was co-opted, with significant transformation, by Julia Ward Howe in her Battle Hymn of the Republic: Mine eyes have seen the glory of the coming of the Lord, He has trampled out the vintage where the Grapes of Wrath are stored.
150 Michael White and Honesto Herrera Ward Howe has brought about a significant conceptual integration in this case. In the Bible, the intensity of the ‘Wrath of God’ derives from the implacable manner in which he harvests and crushes the grapes. In the hymn, Ward Howe metonymically compresses the vintage process into ‘grapes’ and the ‘Wrath of God’ into ‘wrath’ per se. She thereby bypasses the explicit element of comparison, integrating it in the single concept ‘grapes of wrath’, resulting in a very powerful blend. 6. The slogan engineered a counter-expectation shift whereby the negatively profiled wrinkle took on a positive halo, emphasizing the naturalness of the fabrics.
Appendix
Subject matter
Instantiations
FOOD & DRINK
●
● ● ● ●
●
●
●
CARS
●
● ● ●
●
●
Traditional Easter chocolate binge threatens to turn sour. 19 April 2003. Heineken shows thirst for central Europe. 3 May 2003. Coffee brews on frost talk. 24 May 2003 Milk dispute sours EU savings tax deal. 22 March 2003. Gallo espera un nuevo amanecer. 5 Jan 2003 New dawn for Gallo cock Tomatazo a los precios. 23 Jan 2003 Tomato red prices. (Re. increase in price index due to tomato price hike) Una empresa fría con un mercado caliente. 2 Feb 2003. A cold company with a hot market. (Re. ice cream company performance. La Guerra del yogur se agría. 16 March 2003. Yogurt war turns sour. Behind the wheel of an American icon. (Ford company) 31 May 2003. Bridgestone bounces back in US market. 8 Feb 2003. Cadillac renews drive into Europe. 28 June 2003. El automóvil perdió su velocidad. 5 Jan 2003 Cars losing speed. (Re. fall in car sales). La industria de componentes, sin freno. No break in spare parts industry. 13 April 2003. Albertis modera su velocidad. 13 April 2003. Albertis slows down. (Re. slow growth in motorway company) Continued
Business Headlines & Metaphor
151
Appendix – Continued Subject matter
Instantiations
AERONAUTICS
● ●
●
●
TOBACCO
● ●
EADS reaches a new space frontier. 1 Feb 2003. Ryanair swoops to land Buzz in Euro 24m deal. 1 Feb 2003. EADS mantiene el cinturón abrochado. 16 March 2003. EADS keeps belt fastened. (Re. aerospace company losses). Iberia sobrevuela mejor la crisis por su reducida exposición a Asia y a EE UU. 13 April 2003. Low Asia and US risks help Iberia cruise through crisis. Last gasp for New York’s smokers. 30 March 2003. El tabaco extremeño echa humo. 15 June 2003. Red hot Extremenian tobacco. (Re. tobacco growers in protest).
References Baicchi, A. (2003). ‘Relational complexity of titles and texts: a semiotic taxonomy’. In L. Merlini Barbaresi (ed.), Complexity in Language and Text. (pp. 319–41) (Pisa: Edizione Plus-Universidad de Pisa). Bell, A. (1991). The Language of News Media (Oxford: Blackwell). Brône, G. and Feyaerts, F. (2005). ‘Headlines and cartoons in the economic press: double grounding as a discourse supportive strategy’. In G. Erreygers and G. Jacobs (eds.) Language, Communication and the Economy. (pp. 73–99) (Amsterdam: John Benjamins). Charteris-Black, J. (2004). Corpus Approaches to Critical Metaphor Analysis (Basingstoke: Palgrave Macmillan). Charteris-Black, J. (2007). The Communication of Leadership: The Design of Leadership Style (London : Routledge). Charteris-Black, J. (this volume). ‘Metaphor and Political Communication’. Deignan, A. (2005). Metaphor and Corpus Linguistics (Amsterdam: John Benjamins). Dor, D. (2003). ‘On newspaper headlines as relevance optimisers’, Journal of Pragmatics 35, 695–721. Gibbs, R. W. (1994). The Poetics of Mind (Cambridge: Cambridge University Press). Giora, R., Fein, O., Kronrod, A., Elnatan, I., Shuval, N. and Zur, A. (2004). ‘Weapons of mass distraction: optimal innovation and pleasure ratings’, Metaphor and Symbol, 19 (2), 115–41. Herrera Soler, H. (2006a). ‘Globalisers and anti-globalisers’ forums’. In C.A.Hornero, M.M. Luzón, and M.M. Ruiz (eds.), Metaphor, Blending and their Application to Semantic Analysis. (pp.61–81) (Zaragoza: Anubar) Herrera Soler, H. (2006b). ‘Conceptual metaphors in press headlines on globalisation’, Annual Review of Cognitive Linguistics, 4, pp. 1–20. Herrera Soler, H., White, M., Villacañas, B. and Amengual, M. (2006). ‘Figurative language in English and Spanish business press headlines and its effects on inter- and intra-cultural communication’. Universidad de Essen. LAUD Paper no. 661, pp. 1–27.
152 Michael White and Honesto Herrera Koller, V. (2004). Metaphor and Gender in Business Media: A Critical Cognitive Study. (Basingstoke: Palgrave Macmillan). Koller, V. (this volume). ‘Missions and empires: religious and political metaphors in corporate discourse’. Kövecses, Z. (2005). Metaphor in Culture. Universality and Variation (Cambridge: Cambridge University Press). Musolff, A. (2004). Metaphor and Political Discourse: Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Richards, I.A. (1936). The Philosophy of Rhetoric (Oxford: Oxford University Press). Skorczynska, H. and Deignan, A. (2006). ‘Readership and purpose in the choice of economics metaphors’. Metaphor and Symbol, 21(2), 87–104. van Dijk, T. A. (1985). ‘Structure of news in the press’. In T.A. van Dijk (ed.), Discourse and Communication (pp. 69–93) (Berlin : Walter de Gruyter). Velasco-Sacristán, M. and Fuertes Olivera, P.A. (2006). ‘Towards a critical cognitive-pragmatic approach to gender metaphors in advertising English’. Journal of Pragmatics. 38 (11): 1982–2002. White, M. (2006). ‘Language synergies and communication felicity’. In M. Carretero, L.Hidalgo, J. Lavid, E. Martínez, J. Neff, S. Pérez de Ayala and E. Sánchez-Pardo (eds), A Pleasure of Life in Words. A Festschrift for Angela Downing. (vol. I: 561–88) (Madrid: Universidad Complutense). White, M. and Herrera , H. (2003). ‘Metaphor and ideology in the press coverage of telecom corporate consolidations’. In R. Dirven, R. Frank and M. Pütz (eds.), Cognitive Models in Language and Thought. Ideology, Metaphors and Meaning (pp. 277–323) (Berlin : Mouton de Gruyter). White, M., Herrera, H. and Amengual, M. (2004). ‘Interrelation between content and expression in news headlines’. In S. Sainz, I. y Ángel and F. Lago (eds.), Current Trends of Languages for Specific Purposes in an International and Multicultural Context (pp. 239–49) (Granada: Universidad de Granada).
10 MRSA – Portrait of a Superbug: A Media Drama in Three Acts Brigitte Nerlich and Nelya Koteyko
Introduction This chapter focuses on MRSA, one of the superbugs that seem to ‘outwit’ modern medicine. It contributes to the emergent study of ‘infectious diseases and society’ which complements the more established study of ‘science and society’. We use approaches derived from discourse analysis, corpus linguistics and metaphor analysis (especially the study of discourse metaphors and metaphor scenarios) to investigate how an emergent risk from a bacterium was portrayed by the UK national press at three points in time: 1995 when the threat was as yet remote, 2000 when cases began to rise dramatically, and 2005 when the threat from MRSA became a hot political issue. We describe the changing rhetorical strategies used and the influences from science and policy on the use of preferred strategies, which roughly speaking changed from giving the bacteria a voice, to giving patients a voice, to giving politicians and policies a voice. In studying MRSA from a ‘discursive’ perspective we hope to contribute to a new understanding of the epidemiology of infections.1 Because we humans are big and clever enough to produce and use antibiotics and disinfectants, it is easy to convince ourselves that we have banished bacteria to the fringes of existence. Don’t you believe it. Bacteria may not build cities or have interesting social lives, but they will be here when the Sun explodes. This is their planet, and we are on it only because they allow us to be. (Bryson, 2003: 369) As this quotation from Bill Bryson’s popular book A Short History of Nearly Everything shows, human beings are not the only ‘agents’ on this 153
154
Brigitte Nerlich and Nelya Koteyko
planet. It seems that bacteria are some of its ‘prime movers’. This is a view of bacteria that has only recently started to emerge. In the 1960s chemotherapy, immunization, sanitation and the ready availability of antibiotics had reduced infectious diseases caused by bacteria and viruses to such an extent that the Surgeon General of the United States predicted that the book of infectious diseases could be closed. Bacteria and viruses were thought to have been defeated once and for all through the use of modern medicine. In hospitals, concerns such as isolating the infected, or charging staff with responsibilities for cleaning wards, were no longer given priority in an age that believed in the seemingly magical powers of antibiotics. The rise of antibiotic resistance and the increase in hospital-acquired infections, such as MRSA (Methicillin-resistant Staphylococcus Aureus), VRE (Vancomycin-resistant Enterococci) and Clostridium difficile, alongside increased recognition of the threat of emergent infectious diseases such as Severe Acute Respiratory Syndrome (SARS) and avian influenza have removed any such complacency. This chapter will focus on the issue of MRSA in particular, a bacterium that has acquired the label ‘superbug’. MRSA is a specific strain of the Staphylococcus aureus bacterium that has developed resistance to almost all antibiotics and has become widespread, particularly in hospitals. A new strain has also begun to infect people outside hospitals, so-called community-acquired MRSA. We shall focus on MRSA as a health-care-associated infection, or HCAI. As Washer and Joffe (2006) have shown, MRSA only came to the attention of the media and then the public at large at the beginning of the 1990s, although it had been infecting people since the 1960s. One UK Member of Parliament commented in 1995: ‘The first few cases of the bacteria were reported in 1991. Two years later there were 300 cases in more than 40 hospitals. Now 129 hospitals in the UK are reporting it. That is an amazing figure which shows the rapid spread of MRSA.’ (Cohen, 1995) This increase in cases is reflected in the rise in media reporting on the subject (see Table 10.1). MRSA was not yet a topic for the national press in the UK in the early 1990s. This changed in 1995 after the government had issued a first set of guidelines and after two popular science books had been published (Garrett, 1994; Cannon, 1995) that painted a picture of MRSA against a rather apocalyptic background of plagues and ‘superbugs’, a term first used in the mid-1980s, ‘usually in the context of stories about pesticides and the agricultural use of antibiotics’ (Washer and Joffe, 2006: 2145), until, in about 1997, superbug became a quasi synonym for MRSA
MRSA – Portrait of a Superbug 155
Table 10.1 Number of reports on MRSA in UK national press between 1995 and 2007 Year
Number of articles
Year
Number of articles
1990 1991 1992 1993 1994 1995 1996 1997 1998
1 2 2 1 7 21 23 13 39
1999 2000 2001 2002 2003 2004 2005 2006 2007 (Jan–Oct)
64 82 97 216 326 1221 2116 944 1304
in the media.2 From then on media coverage steadily increased until it exploded around 2004, when the Labour government announced various schemes to tackle the rise in MRSA, and 2005 when MRSA and cleanliness became party political issues in the election campaign. It has become clear in recent years that infectious diseases are not only a matter of epidemiology, virology and microbiology, but that the politics and biology of infectious diseases, in our case HCAIs such as MRSA, interact. As Brian Duerden has pointed out, microbial populations interact with human populations and human behaviour (Duerden, 2006). Human behaviour in turn is, in part, structured by human discourse. In studying discourses of MRSA we want to contribute to advances in two fields: Critical Metaphor Analysis and science and technology studies, especially the public understanding of disease which complements the more established study of the public understanding of science.
1 Methods and conceptual framework This chapter exploits, develops and empirically assesses an emerging methodological trend in discourse analysis, namely the convergence of Critical Discourse Analysis (CDA) (Fairclough, 1995; Billig, 1996; Weiss and Wodak, 2003) with Critical Metaphor Analysis (see Charteris-Black, 2004; Larson et al., 2005; Wallis and Nerlich, 2005). Using LexisNexis Professional, three data sets were collected which provided us with three synchronic ‘slices’ probing into the diachronic development of the media portrayal of MRSA.
156
Brigitte Nerlich and Nelya Koteyko
The type of metaphor analysis employed here has emerged from a critique of traditional metaphor analysis in Cognitive Linguistics. It is linked to work on metaphor in social studies of science and the sociology of expectations, especially relating to infectious diseases in humans and animals (Nerlich, 2004; Wallis and Nerlich, 2005; Larson et al., 2005; Nerlich and Halliday, 2007; Washer and Joffe, 2006). The concept of ‘metaphor scenario’ in particular (Musolff, 2006) will be used to analyse media discourses about MRSA in the UK at three distinctive points in time – a time when, as we discovered, MRSA was not yet regarded as a health risk, a time when it became increasingly viewed as a risk to patients in hospitals and a time when the fear of MRSA came to outweigh fears about hospital treatment itself (BMJ, 2004). This chapter investigates how the risks associated with the spread of MRSA were framed by the media and what measures were imagined and implemented to deal with these risks. In 1995 an unknown new risk was framed metaphorically through the personification of various good and evil forces; in 2000 this type of story telling was replaced by real people telling their own stories about fighting the infection; and in 2005 the battle entered a much more political stage. Over time, the issue of cleanliness became more prominent as a ‘weapon’ in the fight against bacterial infection, until it came to be inextricably linked to MRSA in politics as well as in the popular imagination. Why this focus on cleanliness emerged and what it means for the management of HCAI are some of the questions we try to answer in this chapter.
2
1995: personification
Washer and Joffe note in their article dealing with the social representations of MRSA in the UK media between 1995 and 2005 that: ‘In the early articles MRSA is described as an antibiotic or drug resistant condition, thereby providing a close copy of the medical scientific description, with little embellishment.’ (Washer and Joffe, 2006: 2145). Our analysis of the 1995 coverage of the issue in the UK national press seems to contradict this assessment on various levels. As we will show, neither the medical/scientific nor the media description of MRSA was devoid of ‘embellishment’. It is commonly assumed that scientific and medical discourses constitute a sober, unembellished, representation of fact, whereas media discourses are a rather more florid representation of facts as something else, either fiction or anecdote (see Rheinberger, 1997: 103).3 The following analysis demonstrates that this seemingly commonsensical
MRSA – Portrait of a Superbug 157
distinction does not hold up (see also Bucchi, 1998; Nerlich and Halliday, 2007) and, in particular, that it does not allow us to distinguish between scientific and media discourse. As Rheinberger (1997) and Bono (2005) show, both ‘representation of’ and ‘representation as’ are fundamental to scientific thinking and, we would add, both are fundamental to media communication. As Aldrige and Dingwall have pointed out (2003:438): ‘There is [ ... ] a complex symbiosis between science and spectacle’. Science and medicine commonly explore disease as an invasion, the immune system as a defence system against foreign invaders, bacteria as good or bad, friendly or unfriendly and so on. Rheinberger (1997) therefore argues that scientific activity ‘consists in producing, in a space of representation, material metaphors and metonymies’. In this respect it overlaps with the media, especially when reporting on science and disease (Hilgartner, 1990). Let us now turn to the ‘representation’ (with or without ‘embellishment’) of MRSA in the 1995 media coverage. Out of a total corpus of 21 articles, only one article mentions the issue of cleanliness . At that time cleanliness was not yet a major topic for debate. The rest of the articles (apart from a subset dealing with hospital league-tables, which will be ignored here) tell remarkably vivid stories of heroes, villains and battle. These narratives about a new ‘superbug’, provoked for the most part by books written by popular science writers, can be read as a mythical story of good and evil framed by a metaphorical war scenario through which popular science tries to explain how bacteria became powerful ‘agents’ that challenge human and technological supremacy. There are three main protagonists in this mythical story of battle and struggle: doctors who are portrayed either as heroes or villains, but who mostly come across as relatively naïve and easily duped compared to the clever bacteria and not so clever antibiotics, the two other protagonists in this myth or metaphor scenario. Bacteria can be good or bad, but they are mostly portrayed as very clever creatures that outwit doctors and display what one can only call (ironically) ‘intelligent design’ based on exploiting evolution in cunning ways. Antibiotics can be a cure or a curse, but they are mostly portrayed as a bit dim, just like the doctors who prescribe them; they are not quite as clever as the bacteria. This narrative strongly contradicts the normal narrative of human victory over nature. ‘Mindless’ replication and evolution are personified and represented as intelligent action on the part of bacteria which triumph over the human and medical mind. The primitive and primordial outwit the modern mind and modern medicine. This reminds one of ‘Orgel’s Second Rule’, namely that ‘evolution is cleverer than you are’.4
158 Brigitte Nerlich and Nelya Koteyko
On the one side of the representational (battle) stage on which the war against MRSA is being fought, are doctors engaging in overkill (overprescribing antibiotics) but losing the war against bacterial disease. On the opposite side are killer bugs fighting for survival and winning the war, using evolution as a clever tactic to outwit the doctors; they invade, colonize and spread. Caught in the middle are antibiotics which are, as one article stressed, undergoing something of a ‘crisis of confidence’, but also, less metaphorically, patients who are infected with those clever bugs that can outwit antibiotics. Table 10.2 provides an overview of the various ‘actors’, their ‘actions’ and the scenario that is exploited in metaphorically framing those actions. As one can see in Table 10.2, there were four sides in this ‘battle against MRSA/bacteria/germs’: the doctors (real agents), the hospitals (metonymical agents), the personified bacteria (who attracted by far the most metaphors as ‘intelligent agents’), and the (stupid) antibiotics. Patients as victims in this war did not feature much yet. It was not really surprising to find the use of the conceptual metaphor MANAGING DISEASE IS WAR which has become a common framing device when talking about health and illness (Chiang, 2007; see also Board of Global Health , 2006). More surprising was the use of the conceptual metaphors EVOLUTION IS A TOOL and BACTERIA ARE IMPERIALIST AGENTS to describe the actions of bacteria as clever agents engaged in this war with doctors, patients and hospitals. This echoes Bill Bryson’s assessment of bacteria quoted at the beginning of this chapter that ‘[t]his is their planet, and we are on it only because they allow us to be’.
Table 10.2
The battle scenario in 1995
Agents
Types of action
Metaphor scenario
doctors
they had thought they had won the war against bacterial disease they [engage in] colossal overkill they have been fighting it for years they help it [microbe] to become dominant by eliminating the competition they have killed off the proverbial goose… they dish out antibiotics like Smarties hospitals have been under severe pressure from bacterium hospitals fight bacterium hospitals face defeat hospitals harbour zoos of drug-resistant bacteria
war
hospitals
war war carelessness war carelessness contest war war carelessness Continued
MRSA – Portrait of a Superbug 159 10.2
Continued
Agents
Types of action
bacteria, microbes, bugs, germs
killer bacteria war ‘the bugs of war’ war biological invaders war they are survivors war lethal war deadly war bacteria are fighting back war [fighting agents] they relentlessly eat, divide and war multiply, secreting defensive poisons to thwart their attackers, hiding when necessary, and, if all else fails, mutating they rapidly mutate to withstand intelligent use of evolution chemical attacks in war against humans they kill host cells they take advantage of weakness in the defence they overwhelm us by force of numbers they evolve far more rapidly ‘intelligent design’ than homo sapiens and intentionality they adapt, mutate, evolve… [evolution is framed as an they are driven by the maxim intentional, clever thing ‘survive and reproduce’ that bacteria engage in] they have acquired the ability to make an enzyme, penicillinase staggering ability of bacteria… they draw plasmids and transposons from the vast mobile lending library in their environments they share new strength with individual and social other microbes [social agent] intelligence ‘Long before mankind discovered intelligent development of the chemicals, yeasts, fungi and rival ‘tools’ and ‘technology’ bacteria had been making antibiotics and spewing the compounds around newly claimed turf to ensure that rival species couldn’t invade their niches.’ they spread, colonise imperialism they recognise no boundaries they outwit host immune system has learned to break down penicillin ‘intelligent design’ has learned to become resistant to methicillin has colonised up to 90% of hospitals imperialism throughout the world
MRSA
Metaphor scenario
Continued
160 10.2
Brigitte Nerlich and Nelya Koteyko Continued
Agents
Antibiotics
Types of action
Metaphor scenario
is fiercely resistant to hygiene or drugs overcomes drugs strikes terror in hospitals the real enemies are antibiotics they breed disease they may turn out to betray the human race they have upset the balance of nature they turned harmless bacteria into killers they don’t want to become resistant – they do so to survive in the fact of needless overprescribing they are going through something of a confidence crisis they exert selective pressure on bacteria
war, battle, tenacity
war intentionality
war intentionality
consciousness
In this war between humans and bacteria everybody seems to be a villain, again something which is not often found in tales about battles between what is usually seen as ‘good’ and ‘evil’ – here, by contrast, there are no heroes. The doctors are enemies of patients because they overuse antibiotics; antibiotics are enemies because they turn bacteria into killers; and bacteria are enemies because they kill. This is not a chain of command; it rather is a chain of disaster. In this motley crew the bacteria are by far the smartest agents of destruction using mutation, variation and evolution as a weapon in the war, and, unlike doctors, it seems, they learn from experience. Such descriptions use ‘implicit models of evolution’ to describe the action of bacteria seen as intelligent agents and designers (Aldridge and Dingwall, 2003). What can we learn from this metaphorical and narrative framing? What are the actions that doctors, patients and policy makers are invited to envisage? Although no real plan for action is discussed, readers are left with the impression that the only way to deal with MRSA is to fight back against these primordial and intelligent agents of destruction. Bacteria are clever and have to be outwitted – a common trope in scientific discourse about the rise in antimicrobial resistance. But they are described as more than just clever, more than just evolving rapidly. They are described as using evolution as a tool in intelligently designing new deadly onslaughts on humanity. So, what weapons can be used against such brave new bacteria displaying super-evolutionary intelligence? Obviously, overprescribing
MRSA – Portrait of a Superbug 161
antibiotics, those ambivalent medical agents, has to be stopped. Less obviously, we have to give them less occasion to exert their superior evolutionary power, that is, destroy their ecological niches. This means keeping things clean. This issue, only alluded to once in this corpus, becomes central in subsequent media reporting. The 1995 reporting has set the ‘scene’ and introduced the main characters: let the battle commence.
3
2000: patient voices (and the voice of numbers!)
The corpus for the year 2000 contained 82 articles in total, of which 26 articles were written in the broadsheets and 56 in the tabloids. On some days the Daily Mail and the Express in particular published whole clusters of articles on MRSA which explored the issue from various scientific and anecdotal angles. Overall, there was a marked change in the way HCAIs in general and MRSA in particular were reported. Whereas in 1995 the unknown was personified in the context of the emergence of a new and still quite mythical risk to human health, in 2000 the risk becomes an actual personal and institutional risk. The personification of battle-savvy bacteria is complemented by patients’ personal narratives of battle against infection. The scenery changes too. From a mythological evocation of bacteria on the march we now turn to real places and spaces, from hospital corridors to spaces under patients’ beds, where bacteria lurk. The tabloids in particular, as one can expect, focus on personal interest stories: 15 stories about people dealing with or dying from MRSA were found in the tabloids and only two in the broadsheets. The main actors involved in the battle against MRSA change too. Not only is there now a one-to-one battle between patients and bacteria; institutions too become major players and spawn a host of metonymically based personifications. Hospitals as institutions, rather than doctors as individuals, are now pitted against bacteria. Between 1995 and 2000 there was a marked rise in HCAIs and a political response was needed. In January 2000 the Public Health Laboratory Service Board published a report saying that one in every 10 patients in acute-care hospitals is infected after admission and that up to 30% of HCAIs could be prevented through effective infection control programmes (PHLS, 2000). In February the National Audit Office published figures that reverberated in the press coverage throughout the year 2000 (we found 40 items in our corpus) and are still quoted today. In particular it found that about 5000 people a year died of HCAIs and mentioned handwashing as one of the most important measures in the control of MRSA (NAO, 2000).
162
Brigitte Nerlich and Nelya Koteyko
There was however also continuity between 1995 and 2000. Bacteria are still discussed as agents, especially when scientists or scientifically trained people speak out or are quoted. The war and contest scenario shifts slightly towards another scenario – that of crime. Doctors in particular are metaphorically ‘criminalized’ (‘Doctors are among the biggest culprits’; ‘shamed as the worst offenders’; ‘some of the culprits for transmission of infection are doctors’), but so are bacteria (they ‘stalk’, they ‘lurk’, are ‘at large’ etc.). Another scenario closely linked to struggle and contest is also activated, namely the race scenario, a race in which the bacteria always seem to be winning over doctors and hospitals (‘bacteria are always one step ahead’; ‘bug would change form and stay one step ahead of science’; ‘medical science is struggling to catch up’). A somewhat more creative quasi-military metaphor is used by two scientists who both say that MRSA is a ‘timebomb’, thereby stressing the urgency of action needed to fight against its spread. This more abstract concept supports the mobilization of patient and victim voices by the media which also represent increasing urgency in the management of MRSA. Whereas 1995 was mainly about doctors/hospitals and bacteria engaged in a mutual but rather abstract struggle, 2000 is about concrete hospitals, patients and cleanliness. Unlike in 1995, when only specially coated ‘killer tiles’ appeared as remedies against HCAI, various ways of dealing with MRSA are discussed, from alternative medicines to testtube antibodies. But the message that cleanliness might be key to eradicating MRSA becomes central. ‘Standards’ and ‘rates’ of cleanliness are mentioned in almost every text from our corpus and form the focus of discussion in some tabloid articles. Hygiene is discussed as the central measure to beat ‘superbugs’, and failure to maintain it by some hospitals draws severe public criticism. However, the 1995 theme of bacteria as outwitting humans continues. We can read for example that ‘The bug laughs off all attempts to control it with antibiotics’ (Mirror, 17 November 2000) and that ‘Some [bacteria] developed little pumps that squirted out the antibiotics faster than they could get in. Others developed chemicals that chopped up penicillin, and grew thick cell walls that made it impossible for the antibiotics to penetrate them’ (Mirror, 22 May 2000). Bacteria are not only clever in using evolution as a tool, they also devise metaphorical machines and technologies that can compete with advances in ‘real’ medical technology. The metaphorical scenarios of race and competition come to the fore here. In 2005, the competition frame came to dominate the debate, but in the sense of political parties competing for supremacy, using MRSA as a clever tool to gain political advantage.
MRSA – Portrait of a Superbug 163
4.
2005: politics and policies
In 1995 a new threat to human health had become visible to the media and a mythological story of good and evil was told in which bacteria, antibiotics, doctors, and hospitals were the real and metaphorical agents locked in a battle of wit. In the year 2000 a new agent entered the scene: the patient. The mythical story of bacterial supremacy was supplemented by a purification myth in which the doctors, already discredited for letting themselves being outwitted by bacteria, became criminals, and more than that, became ‘unclean’. In 2005 finally two other figures enter the scene: politicians who try to ‘wipe out’ MRSA and wipe out their opponents by promising to clean up MRSA, and matrons, the new guardians of purity in the war against the bug. Two prototypes in particular are invoked: Florence Nightingale and Hattie Jacques (an actress famous for playing the role of Matron in the comic Carry on series of films). Doctors are almost absent from the debate, but nurses and patients assume more centre-stage roles. Let us now look more closely at this development. As Table 10.1 shows, the national press output for 2005 on the issue of MRSA was enormous – 2116 texts. To make the corpus manageable for a qualitative analysis,5 we focused on the highest output for one month, April (387 articles), and focused yet again on the output of one tabloid and one broadsheet that had consistently published most articles on MRSA, namely the Daily Mail (31 articles) and The Times (28 articles). In the autumn of 2004 the then Health Minister John Reid had set as a government target that blood-borne MRSA infection rates be halved by 2008. This, together with the publication (Department of Health, 2004a) of ‘A matron’s charter: an action plan for cleaner hospitals’, a ‘Clean your hands campaign’ and a policy document entitled ‘Towards cleaner hospitals’ ( Department of Health, 2004b) had put MRSA and cleanliness on the political agenda. Both issues became the focus of the 2005 General Election. May 5 2005 was the day of the election. Leading up to this date, April 2005 was the month during which political campaigning became most intense. April 12 was the day that the Conservatives published their Manifesto, which attracted a lot of press coverage, followed by Labour on April 14. One of the central images of the Conservative campaign became a poster declaring: ‘Under Blair, there have been 5,000 deaths every year from dirty hospitals. Imagine five more years of it.’ Or: ‘I mean, how hard is it to keep a hospital clean?’ Cleanliness became central to the debate surrounding the threat posed by HCAIs.
164
Brigitte Nerlich and Nelya Koteyko
The Times reported that ‘Nurses demand a clean uniform on every shift to beat superbugs’ (27 April 2005) and the Daily Mail wailed that we would need ‘20 years to wipe out the superbug’ (27 April 2005). The scenario of battle against the microbe is now linked to that of cleaning and purity in various ways, especially via the expression ‘wiping out’ which has military as well as cleaning associations (Royal College of Nurses, Wipe it Out campaign, 2005). Cleanliness becomes a possible, plausible and above all ‘commonsensical’ weapon to ‘beat’ the superbug. The phrase ‘wipe out’ has been associated both with the metaphorical framing of war, especially the ‘War on Terror’ in which the United States wishes to ‘wipe out’ terrorism (Lakoff, 2001) as well as with the framing of germ-warfare in which we all engage now in our homes where we use disinfectants and antimicrobial ‘wipes’ (it is perhaps no coincidence that George Bush is reported to use a hand-sanitizing product after he shakes hands with his guests (Leibovich, 2006)). More generally ‘cleansing’ has become a metaphor used in wars from Hitler to Bosnia and beyond. Setting out to ‘wipe out’ MRSA, in a context where such actions are not only metaphorical but real and accompanied by ubiquitous images of cleaning, hand washing and cleansing gels, therefore demonstrate political leadership and political will. War and blame too became closely linked. Promising to wipe out/ defeat/beat MRSA becomes a party political issue and one of mutual blame, with the Conservatives blaming Labour for not keeping their promises to make hospitals cleaner and MRSA free, and Labour blaming the Conservatives for not having done enough before 1995 to eradicate MRSA. What The Times called the ‘Superbug battle’ (15 April 2005) is no longer a scientific battle against clever bacteria but a battle of blame between politicians and patients where ‘Cleaning up MRSA’ (The Times headline, 16 April 2005) has become the ultimate goal. By being discussed in the election campaign, in meetings of nongovernmental organizations , on various popular television programmes from Casualty to Panorama, to under-cover documentaries on Channel 4, MRSA became the topic of the day in 2005 and an acronym, like BSE or GM or FMD, that everybody is now familiar with in the UK. One polemicist wrote an article for The Times on 22 April entitled (in its online version) ‘MRSA: metaphor for a diseased body politic’6 and asked: ‘How has such a microcosm as the MRSA bug been blown up into one of the biggest issues in the general election?’ The answer might be found in what he himself calls the ‘politics of fear’ used by the Conservatives. Another answer might be found in the politics of blame in which both parties, Labour and Conservative, engaged.
MRSA – Portrait of a Superbug 165
Both The Times (a broadsheet) and the Daily Mail (a tabloid) discussed cleaning as a weapon in the war against MRSA, but they both, to varying degrees, also highlighted other issues, such as bed occupancy and government-required performance targets – an issue explored also by the Conservatives. Both newspapers therefore stressed scientific and policy uncertainty. Despite the uncertainties surrounding the issue MRSA is represented in 2005 as something controllable. Politicians treat MRSA rates like taxes – you can at least pledge to bring them down. Overall, in 2005 MRSA became a political issue not only for political parties but also for non-governmental organizations and patients at large who became more politicized. Cleanliness became the focus of political and popular debate. It was seen as a commonsensical way of dealing with the complex issue of the rise of rates in HCAIs (Koteyko et al., in press.). The hyperbole of the war against super-clever bacterial agents, evident in 1995, is replaced by what seems a straightforward method of dealing with HCAI. As has become evident since, this was perhaps complacent and some scientists use a different type of hyperbole to awake policy makers from their policy slumber. A colleague of ours at the University of Nottingham, for example, has introduced the metaphor of the ‘postantibiotic apocalypse’ (see Nerlich, in prep.).
5. Conclusion The ‘representations’ of MRSA in the UK national press between 1995 and 2005 can be seen as a drama in three acts, drawing on various genres, stereotypical plot lines, characters and other historical or fictional narratives. In the first act the dramatis personae are personified forces of nature as well as earthly creatures fighting them, namely doctors and hospitals engaged in a battle of evil against good, where a new type of intelligence is pitted against stupidity. In the second act the victims of the personified bacterial forces are introduced and the doctors and hospitals emerge as perpetrators of crimes against them, a crime that is not a direct act of killing but rather a crime of omission: cleaning hands or wards. In the third act the malignant forces of politics try to exploit the evil forces of nature for their own ends, but the erstwhile victims, patients and nurses that were on the sidelines in the first two acts, empower themselves. A mediator between the doctors and the potential victims of MRSA emerges and is given political and symbolic power: the ‘modern matron’, framed by the factual but mythically enhanced narrative of Florence Nightingale and the fictional one
166
Brigitte Nerlich and Nelya Koteyko
of Hattie Jacques: both evoke images of empowerment, cleanliness, and strength which are needed to finally defeat the evil and unclean forces of nature. Throughout the three acts, the various dramatis personae engage in war, battle and struggle, but also in racing and crime, the overall metaphorical scenario being one of contest. The narratives that this three-part drama draws on are the struggle between heroes and villains, between contamination and purity and between intelligence and stupidity. Metaphors, especially when linked to metaphor scenarios, allow us to understand complex phenomena in terms of everyday experiences. They and the narratives they tie together also have a performative force. As Bono has pointed out, while ‘metaphors [ ... ] serve as ‘invitations to action’, [n]arratives are users’ manuals for putting metaphors into action, for learning to work with and through metaphors.’ (Bono, 2005: 137) Representing bacteria as intelligent agents; representing the management of HCAIs as a war; representing cleaning as the ultimate solution; representing matrons as guardians of purity – all this has implications for behaviour and action, funding and policy making. In this fight against MRSA cleanliness has become a central feature, despite the lack of scientific evidence and the general scientific uncertainty surrounding its effectiveness.7 The issue of infection control, and containment of MRSA in particular, relies on the multiplicity of factors such as isolation measures, hand-washing by hospital staff, screening of patients upon arrival, screening and decolonization of healthcare workers, of which cleanliness is only one, although an important component (Loveday et al., 2006). As Voss states (2004: 521): ‘these measures work only when used in a concerted manner. Thus to single out one or a few measures, especially on the basis of limited evidence, risks potentially effective measures being neglected in future guidelines.’ Focusing on cleanliness alone may allow bacteria to reassert their cleverness, an issue which was the focus of media debate a decade ago, and should perhaps be stressed again today.
Notes 1. The chapter was written as part of the ESRC project ‘Talking cleanliness in health and agriculture’. Grant number: RES000231306. 2. The use of the term is currently changing, as it is applied more generally to other HCAIs, especially after recent rises in Clostridium difficile. 3. Hilgartner (1990, 2000), for example, studies scientific advice as a type of performance, a topic we shall come back to below. 4. See http://en.wikipedia.org/wiki/Orgel’s_rule.
MRSA – Portrait of a Superbug 167 5. The overall coverage for 2005 was analysed separately with quantitative methods of corpus linguistics (see Koteyko et al, in press). 6. On the metaphor of the ‘body politic’, see Musolff (2004). 7. In scientific literature, cleaning is also introduced as the most ‘commonsensical’ infection control measure (see Anonymous, 1972) to be undertaken in an era of rising antimicrobial resistance.
References Aldridge, M., and Dingwall, R. (2003). ‘Teleology on television? Implicit models of evolution in broadcast wildlife and nature programmes’, European Journal of Communication, 18, 4, 435–53. Anonymous (1972). ‘Contamination, cleaning, and common sense’, The Lancet, 1, 1108–9. Billig, M. (1996). Arguing and Thinking: A Rhetorical Approach to Social Psychology (Cambridge: Cambridge University Press). Board on Global Health/BGH (2006). Ending the War Metaphor: The Changing Agenda for Unraveling the Host-Microbe Relationship - Workshop Summary (Washington: Institute of Medicine). Bono, J. (2005). ‘Perception, living matter, cognitive systems, immune networks: a Whiteheadian future for science studies’, Configurations, 13, 1, 135–81. British Medical Journal/ BMJ (2004). UK health news, http://bmj.bmjjournals. com/uknews [Weblink defunct]. Bryson, B. (2003). A Short History of Nearly Everything (USA: Broadway Books). Bucchi, M. (1998). Science and the Media: Alternative Routes in Scientific Communication (London: Routledge). Bud, R. (2005). ‘The history of antibiotics’. The Medicine Chest, http://www. wellcome.ac.uk/doc_WTX026108.html. Cannon, G. (1995). Superbug – Nature’s Revenge: Why Antibiotics Can Breed Disease (London: Virgin Publishing Ltd). Charteris-Black, J. (2004). Corpus Approaches to Critical Metaphor Analysis (Basingstoke: Palgrave Macmillan). Chiang, W. (2007). ‘Conceptual metaphors for SARS: ‘war’ between whom?’, Discourse & Society, 18, 5, 579–602. Cohen, H. (1995). House of Commons Hansard Debates for 5 December 1995, http://www.publications.parliament.uk/pa/cm199596/cmhansrd/vo951205/ debtext/51205–41.htm. Dancer, S. J. (1999). ‘Mopping up hospital infection’, Journal of Hospital Infection, 43, 85–100. Department of Health (2004a). A Matron’s Charter: An Action Plan for Cleaner Hospitals (London: The Stationery Office). Department of Health (2004b). Towards Cleaner Hospitals and Lower Rates of Infection: A summary of action (London: Department of Health). Department of Health. (2005). Saving Lives: A Delivery Programme to Reduce Healthcare Associated Infection (HCAI) Including MRSA (London: Department of Health). Duerden, B. (2006). ‘Biology, politics and performance management. Tackling HCAI in the NHS in England’, www.dhsspsni.gov.uk/13-brian-duerden.pdf.
168
Brigitte Nerlich and Nelya Koteyko
Fairclough, N. (1995). Critical Discourse Analysis: The Critical Study of Language (London: Longman). Garrett, L. (1994). Coming Plague: Newly Emerging Diseases in a World Out of Balance (New York: Penguin). Hilgartner, S. (1990). ‘The dominant view of popularization’, Social Studies of Science, 20, 519–39. Hilgartner, S. (2000). Science on Stage (Stanford: Stanford University Press). King’s Fund. (2005). ‘MRSA’, http://www.kingsfund.org.uk/resources/briefings/mrsa.html. Koteyko, N. , Nerlich, B., Crawford, P. and Wright, N. (in press). ‘Not rocket science’ or ‘no silver bullet’? Media and government discourses about MRSA and cleanliness’, Applied Linguistics. Lakoff, G. (2001). ‘Metaphors of Terror’, http://www.press.uchicago.edu/ News/911lakoff.html. Larson, B., Nerlich, B. And Wallis, P. (2004). ‘Metaphors and biorisks: the war on infectious diseases and invasive species’, Science Communication, 26, 3, 243–68. Lawrence, J. (2005). ‘MRSA, politics and the press’, http://www.wellcome.ac.uk/ doc_WTX026126.html. Leibovich, M (2006). ‘In clean politics, flesh is pressed, then sanitized’, New York Times, 27 October . Loveday H.P., Pellowe, C.M., Jones S. and Pratt R. (2006). ‘A systematic review of the evidence for interventions for the prevention and control of meticillin-resistant Staphylococcus aureus (1996–2004)’, Journal of Hospital Infection, 63, 1, 45–70. Musolff, A. (2004). Metaphor and Political Discourse. Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Musolff, A. (2006). ‘Metaphor scenarios in public discourse’, Metaphor & Symbol, 21, 1, 23–38. National Audit Office and Comptroller and Auditor General. (2000). The Management and Control of Hospital Acquired Infection in Acute NHS Trusts in England (London: NAO). Nerlich, B. and Halliday, C. (2007). ‘Avian flu: the creation of expectations in the interplay between science and the media’, Sociology of Health and Illness, 29, 1, 46–65. Nerlich, B. (in prep.). ‘ ‘‘The post-antibiotic apocalypse” and the “war on superbugs”: catastrophe discourse in microbiology, its rhetorical form and political function’. Nightingale, F. (1859/1952). Notes on Nursing, revised edn (London: Gerald Duckworth and Co. Ltd). Public Health Laboratory Service Board/PHLS. (2000). Communicable Disease Report 10, 3, 21 January. Rheinberger, H.-J. (1997). Towards a History of Epistemic Things: Synthesizing Proteins in the Test Tube (Stanford: Stanford University Press). Royal College of Nursing/RCN. (2005). Wipe It Out: Good Practice in Infection Prevention and Control - Guidance for Nursing Staff (London: RCN). Voss, A. (2004). ‘Preventing the spread of MRSA’, British Medical Journal, 329, 521.
MRSA – Portrait of a Superbug 169 Wallis, P. and Nerlich, B. (2005). ‘Disease metaphors in new epidemics: the UK media framing of the 2003 SARS epidemic’, Social Science & Medicine, 60, 2629–39. Washer, P. and Joffe, H. (2006). ‘The hospital ‘superbug’: social representations of MRSA’, Social Science and Medicine, 63, 8, 2141–52. Weiss, G. and Wodak, R. (2003). Critical Discourse Analysis: Theory and Interdisciplinarity (Basingstoke: Palgrave Macmillan). Williams, R., Blowers, R., Garrod, L. P. and Shooter, R. A. (1966). Hospital Infection: Causes and Prevention (London: Lloyd-Luke Ltd).
This page intentionally left blank
Part III Metaphor Evolution in Discourse History
This page intentionally left blank
11 Shifting Identities: Metaphors of Discourse Evolution Roslyn M. Frank
Introduction Over the past two decades developments in the field of cognitive science have brought together pre-existing methodologies and theoretical approaches from a wide variety of disciplines and at the same time promoted cross-disciplinary dialogue relating to the development of new methodologies and theoretical frameworks (Bono, 1990; Maasen and Weingart, 1995, 2000). This cross-fertilization has been particularly rich in the case of researchers concerned with modelling ‘language’ and ‘language change’ in a number of new settings, for example those involved in working with artificial distributed agents associated with research projects in Artificial Intelligence (AI) and Artificial Life (A-Life), as well as in the areas of ecolinguistics, biosemiotics and theoretical biology. Whereas a great deal of attention and effort has been focused on developing these models in various subfields of cognitive science, to date less work has been carried out by Cognitive Linguists in terms of attempting to model the entity comprised by ‘language’ through crossfertilization with the evolving methodological and theoretical models found in the ‘hard’ sciences, for example, Complex Adaptive Systems theory (CAS). Nonetheless, in recent years a number of important steps have been taken in this direction, for example Croft (2000), Steels (1999) and most recently Sharifian (2003, 2008) and Frank (2008a). These initiatives represent a conscious move away from the linear, Cartesian-Newtonian mode of thinking and the linear conceptualization of causality characteristic of earlier models of ‘language’ and ‘language change’ and, as such, these steps represent movement toward (re-) descriptions of the phenomenon of ‘language’ more in terms of a selforganizing, dynamic system. The notion of a self-organizing, dynamic 173
174
Roslyn M. Frank
system is central to Complex Adaptive Systems (CAS) theory, also known as Dynamical Systems theory (Clark, 1997). Recently another avenue has opened up for applications of CAS thinking, namely, the potential that this theoretical framework has for the analysis of discourse metaphors. The latter are defined as ‘relatively stable metaphorical mappings that function as a key framing device within a particular discourse over a certain period of time’ (Zinken et al., 2008). The present chapter focuses first on the applications of CAS thinking to the notion of ‘discourse metaphor networks’. Then, an exemplary analogical sequence is explored: the evolution and discourse career of a biological concept, namely, that of the ‘gene’. Although this term first appeared at the beginning of the twentieth century, only in the past decade has it started to penetrate the discourse of Cognitive Linguistics and Critical Discourse Analysis, revealing at the same time its ability to generate extended metaphor formations in the linguistic sciences. More concretely, I will discuss several analogical expansions of this base concept of a ‘gene’, showing how they currently function as a productive source for heuristic inferences in contemporary discussions of language and language change, particularly in the case of those attempting to incorporate an evolutionary or Neo-Darwinian perspective into their overall explanatory model for language evolution or language change, sometimes referred to as a ‘population approach’ where ‘language’ is treated, analogically, from the perspective of a ‘species’ or ‘population’. Discourse metaphors provide evidence for the socio-cultural situatedness of metaphorical reasoning along with the characteristic features of context-boundedness, strategic fuzziness, and polyvocality. As others have shown, discourse metaphors often demonstrate a rich social and cultural history. They can also exhibit an uncanny conceptual staying power, which reflects their status as highly entrenched, albeit constantly changing, entities, given that the socio-cultural ground under them is always shifting (Chilton, 2005; Frank, 2003, 2005, 2008a; Nerlich and Hellsten, 2004: 262; Zinken et al., 2008; and the contributions by Banks, Cowling, Musolff and Zavadil in this volume). Hence, discourse metaphor analysis brings to the fore the importance of a diachronic perspective with a cultural orientation, where continuity is apparent in the cognitive patterns manifesting themselves in such a longitudinal analysis. This set of conditions allows the discourse metaphor formation or analogical network to interact and hence co-evolve with its socio-cultural environment. On the one hand, this socio-cultural embeddedness acts to provide stability for the network in question, i.e., there are cognitive constants
Metaphors of Discourse Evolution
175
that seem to be discursively embedded in a relatively stable reservoir of cultural beliefs and social representations. On the other hand, these environmental factors can act to destabilize the dynamics of the construct, given that discourse metaphors simultaneously provide sites for conflict, resolution and cooperation on the part of the language agents. In sum, the meanings associated with a given discourse metaphor are socio-culturally situated and co-evolve in conjunction with the cultural constructs in which it is embedded (Zinken et al., 2008; Musolff, 2004; Frank et al., 2008b). In this sense, the stability of the discourse metaphor depends, at least to a substantial degree, on the nature of the coupling that takes place between the discourse metaphor network and other cognitive artefacts, for example the socio-cultural resources that are available, internally and externally, to the language agents themselves. In order to examine the dynamics inherent to this process we can begin by looking at the advantages of incorporating a Dynamical Systems theory view (Clark, 1997: 80–123) and, more narrowly, by reflecting on the benefits of adopting a CAS model as a tool for exploring the functioning of discourse metaphors and analogies.
2
Overview of Complex Adaptive Systems thinking
Complex systems are systems in process that constantly evolve and unfold over time. Change is an integral element of their functioning. In the case of complex adaptive systems, a subset of dynamic nonlinear systems, they are adaptive in that they have an innate capacity to change and learn from experience, so to speak. Thus, they are endowed with the ability to evolve and adapt to a changing environment. Examples of CASs include social insect and ant colonies, the biosphere and the ecosystem, the brain and the cell, the immune system, financial markets, social networks, the Internet and also, in general, any human social group-based endeavour forming part of a cultural and social system. Thus, since Complex Adaptive Systems arise in a wide range of contexts, this theoretical framework is rapidly gaining ground in a variety of disciplinary areas, not only in the biological and physical sciences (Lansing, 2003: 183), but also the social sciences and to some extent as a tool for the study of artificial and natural language evolution, particularly in the field of ‘evolutionary linguistics’ (Steels, 1999, 2004). Broadly defined, a CAS is one that is self-organizing in which there are multiple interactions between many different components while the components themselves can consist of networks that in turn operate as complex (sub)systems. CAS thinking is concerned with understanding
176
Roslyn M. Frank
the global behaviour arising from local interactions among a large number of agents. This global behaviour or emergent dynamics is often quite complex; it is neither specified by prior design nor subject to centralized mechanisms of control, and, consequently, it is often difficult or impossible to predict solely from knowledge of the system’s constituent parts what the emergent global-level properties of the system will be. Therefore, the basic characteristics of a CAS include the following: (1) it is constantly constructed and reconstructed by its users; (2) it is self-organizing; and (3) it is characterized by multiple mechanisms of control, that is, control is distributed throughout the system rather than residing in a single centralized command and control centre. Furthermore, in such a system ‘global order’ derives from local interactions. Hence, when we apply the CAS theoretical model to natural language and/or, less expansively, to the cognitive networks that underlie discourse metaphors, the overall system can be viewed from two perspectives at the same time: ●
●
from the local level, which allows for description and analysis of the activity of the (individual) language agent and her cognitive architecture (idiolect + the socio-cultural situatedness of the agent herself, viewed as embedded in and, hence, inseparable from the influence of an environment that itself is subject to constant alteration); from the global level, which allows for the description and analysis of the global order while the latter, in turn, is the result of the combined activities of heterogeneously distributed agents over time.
Briefly stated, discourse metaphor networks can be understood methodologically as examples of CASs, constantly in the process of being constructed, deconstructed and reconstructed by language agents, and in which there are multiple interactions between many different components. In order to understand the functioning of the system, we need to remember that the emergent phenomenon described above has a strong causal impact on the behaviour and learning of each individual language agent. Consequently, there is a kind of ‘circular causality’ operating at all times that forms an intrinsic part of the system (Steels, 1999). At the local level, the behaviour of individual language agents determines ‘language’, that is, their choices act, cumulatively, to determine (emergent) language structure understood as operating at the global level. At the same time, when viewed from the local level the resulting emergent global-level structure co-determines the range of behaviours of the agents, that is, it acts to constrain and shape
Metaphors of Discourse Evolution
177
their interactions at the local level. In summary, circular causality is a fundamental aspect in the functioning of language and the constitution of discourse metaphors and is not unusual in other types of living systems which are themselves self-organizing and complex in nature. Therefore, a fundamental goal of research models designed for the study of natural languages, evolutionary change, and metaphor formation is to gain an understanding of these bottom-up and top-down exchanges between local and global levels of a complex system, as each provokes emergences and constraints upon the other.
3
Shifting perspectives: genes, memes and linguemes
Until recently the CAS approach to language has been more widespread as a model in the field of ‘evolutionary linguistics’, most particularly in computer simulations of the evolution of language, rather than as part of the theoretical toolkit of Cognitive Linguists examining natural language(s). So, the next logical step would seem to be its design application to the problems facing ‘evolutionary cognitive linguistics’ where explorations of natural language processes would be the goal, including those involving discourse metaphors. Here the CAS model could help reduce or even alleviate certain conceptual complexities, or perhaps better stated, drawbacks, that are inherent in some of the more recent ‘gene-centric’ or ‘meme-centric’ models often found alongside ‘speciescentric’ and ‘population-centric’ models of language (Chilton, 2005; Croft, 2002, 2006; Deacon, 2004; Mufwene, 2001, 2005; Steels, 2004). These approaches, particularly those most influenced by ‘memetics’, can be viewed collectively as a kind of discourse metaphor in evolution for they tend to appropriate the notion of agency attributed to the concept of a ‘gene’. Except in the most unusual of circumstances, one would not expect the expertise of linguists to be in the field of molecular biology, nor would one expect them to have a familiarity with the evolution of the concept ‘gene’, a concept that I suggest can be viewed as an analogy that has its own long and complex history. Rather, I believe most would agree that the intersection of these disciplinary discourses results from the way that metaphors regularly act as ‘messengers’, propagating themselves across disciplines (Maasen and Weingart, 1995). At times it has been the field of linguistics that has contributed to the metaphoric repertoire of the ‘hard’ sciences, and at other times the analogies have moved in the opposite direction (Frank, 2008a). In short, adopting a CAS perspective provides us with a means of tracing the heuristically productive role of these biologically inspired metaphors, e.g. the use of
178 Roslyn M. Frank
‘species’ analogies (Mufwene, 2001, 2005) as well as Croft’s ‘Generalized Analysis of Selection’ model with its linguemes (as linguistic counterparts of ‘memes’) and ‘lineages’ (Croft, 2000, 2002). The CAS model brings new conceptual tools into play and, at the same time it also allows for continuity with another emerging avenue of research, namely, attempts at reformulation of the meaning of a ‘meme’ and more broadly of the field called ‘memetics’. Following Dawkins (1976, 1982, 1991), a ‘meme’ refers to a unit of cultural information that is transferable from one mind to another: a unit that leaps from one mind to another. For Dawkins, examples of memes are ‘tunes, catch-phrases, beliefs, clothes fashions, ways of making pots or of building arches’ and ‘[j]ust as genes propagate themselves in the gene pool by leaping from body to body via sperms or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain’ (Dawkins, 1976: 192). In its popularized version, the memetics framework is assumed to hold that a meme: propagates itself as a unit of cultural evolution and diffusion – analogous in many ways to the behavior of the gene (the unit of genetic information). Often memes propagate as more-or-less integrated cooperative sets or groups, referred to as memeplexes or memecomplexes. [ ... ] Proponents of memes suggest that memes evolve via natural selection – in a way very similar to Charles Darwin’s ideas concerning biological evolution – on the premise that variation, mutation, competition, and ‘inheritance’ influence their replicative success. ( Wikipedia, 2007)1 This ongoing process of reformulation of the basic tenets of memetics is encountered in the work of linguists such as Musolff (2004, 2008), Chilton (2005) and Croft (2002) as well as in Mufwene (2001, 2005). Researchers in adjacent disciplines of cognitive science have noted that the concept of a meme needs to be fleshed out; that as a concept it is vague, lacking in specificity and that it operates with a misplaced sense of agency (Gatherer, 1998; Wilkins, 2005). Deacon (2004) has observed: [t]he core problem of this theory [of memetics], I think, is a kind of misplaced agency, that gives the impression that both genes and memes – replicators – can be understood without considering their embeddedness in a dynamic system which imbues them with their function and informational content. This, then, is not just a problem with memes, but a problem with the replicator concept in general,
Metaphors of Discourse Evolution
179
inherited from Dawkins’ short-circuited description of information processes in biology. (Deacon, 2004: 20) Deacon is far from alone in his rejection of the ontological grounding and misplaced agency attributed to the replicator concept, genes and memes in general. For example, speaking (2004: 174) about the ‘trendy talk about the selfish gene also known as “the replicator” ’, Moss goes on to observe that for Dawkins and his epigones it is the parasite [selfish gene] that invents the host [ ... ..]. Dawkins’ selfish replicator constitutes the quintessence of conflationary confusion. His viewpoint does not build on the advancing elucidation of molecular biology but rather depends on an enforced ignorance of it. Dawkins and his followers take their conflationary replicator (the so-called selfish gene) as an ontological bedrock. (Moss, 2004: 194) The confusion surrounding the propagation of the analogical sequence that has produced this gene meme lingueme series results from the fact that in contrast to proteins, lipids and carbohydrates, the gene ‘did not come on the scene as a physical entity at all, but rather as a place-holder in a biological theory’ (Moss, 2004: 2). Indeed: there never has been a unitary, fixed ‘object’ that acted as a causal agent for replication and transmission of genetic material. Rather the term has always been a place-holder: when the term ‘gene’ was invented, it did not correspond to any real entity, rather it was a hypothesis, albeit a strongly held one, still awaiting scientific verification. As such, it should be viewed as a device that functioned more as a heuristic guiding the direction of research than as having a concrete physical counterpart. Yet for biologists this conceptual notion became so embedded in their (earlier) ways of thinking and more importantly in the popular imagination, that today it is nearly impossible to rid ourselves of these habits of thought (and language). (Moss, 2004: 2) Stated differently, throughout the history of classical and early molecular genetics, the ‘gene’ was generally assumed to be not only a fixed and unitary locus of structure and function but also a locus of causal agency. So, we might argue, ‘it was a way of talking that tacitly grants to ‘genes’ the power to act, even in the absence of any information about how they might act’ (Keller, 2000: 46, emphasis in the original). Hence,
180
Roslyn M. Frank
from its inception the term ‘gene’ acquired properties of (unitary) materiality, agency, autonomy and permanence. And over time that reified ‘way of talking’ which assumed a context-independent ‘materiality’ for the concept became commonplace. In contrast, over the past twentyfive years subsequent genomic research has revealed that none of these assumptions are true as they were stated, and that the reality of how genomic material interacts and replicates is far more complex than previously suspected. In some ways, the processes themselves are equally as puzzling to the biologists of today as they were to their counterparts in times past (see Moss, 2004). Critiques of the tenets of memetics and its misplaced agency, such as those of Deacon, Moss and Keller, should be understood in light of these observations on the non-materiality of the referent associated with the concept of a ‘gene’. Furthermore, these comments come from members of a given epistemic community and are directed to members of similarly situated interpretive communities who are familiar with the topic at hand and the controversy in question. In contrast, Croft explains his choice of the term ‘lingueme’ to his readers who are primarily members of a different disciplinary interpretive community, as follows: In order to clearly distinguish the embodied replicator from the structure that it possesses, we must give it a name. Following a suggestion by Martin Haspelmath, I propose that the paradigm linguistic replicator be called a LINGUEME, on the analogy with Dawkins’ meme. Thus, the paradigm replicator in language is the lingueme, parallel to the gene as the basic replicator in biology; an utterance is made up of linguemes, and linguemes possess linguistic structure (Croft, 2000: 28). Citing Hull and Dawkins, Croft draws out additional analogies holding between this view of a ‘gene’ and a linguistic ‘replicator’, explaining that replication is: the process by which an entity (the replicator) produces a copy that possesses a version of inherent structure of the original entity [ ... ]. Replication can be normal (identical with the structure of the parent) or altered (not completely identical with the structure of the parent). Differential replication is the replication of a replicator at an increasing (or decreasing) relative frequency compared with other replicators. (Croft, 2000: 242)
Metaphors of Discourse Evolution
181
In this respect, we need to keep in mind that the history of the concept of a meme dates back to metaphoric extensions of what today we must view as overly simplistic and outdated definitions of the concepts of ‘gene’ and ‘gene agency’, extensions which are simply no longer viable in the post-genomic era, but rather tied to earlier gene-centred discourses which in turn were based on even earlier variations of preformationist gene concepts, grouped under the rubric of ‘gene-determinism’ (Moss, 2004; Hellsten, 2005; Strohman, 1997, 2001; Hilferty and Vilarroya, 2008). In other words, the notion of agency assigned to the gene during the 1960s and later transferred to Dawkins’ (selfish gene) meme has been called into question by advances in today’s systems biology and related fields of complexity science. At the same time philosophers of science as well as researchers in the field of biology and environmental science are increasingly attuned to the important role played by extended metaphoric networks in guiding research directions and experimental practices (see Keller, 2000; Moss, 2004). When Dawkins first came up with his notion of a meme, based analogically and phonetically on gene, biologists still thought that the ‘program’ was in the genes, and then later in the proteins ‘encoded’ by genes. This type of gene-determinism is now being challenged by a broader context-bound model, the ‘new epigenetics’ (Moss, 2004: 52) and shaped by a CAS approach (Strohman, 2001), as well as by a far greater awareness of the complexity of gene– protein–environment interaction (Nerlich and Hellsten, 2004). Consequently, even though memetics was based originally on earlier and in many senses now outdated formulations of gene agency, recent attempts by (Cognitive) Linguists to reformulate memetics are both interesting and promising. Their research agendas should be viewed as a means of testing how the highly entrenched and wildly popular term ‘meme’ might be appropriated, expanded and recast, in short, how it might be redefined and appropriated as part of a terminological toolkit that could be employed, for example, when discussing the ‘discourse career of a metaphor’ (Musolff 2004: 70). Musolff proposes converting memetics into a tool for the investigation of metaphor and conceptual evolution. His proposal would bring into view the functioning of ‘conceptual clusters’, roughly equivalent to the ‘nodes’ and ‘networks’ operating in the (sub) dynamics of a CAS modelling of discourse metaphor formation. Musolff also explores the methodological advantages of employing ‘a metaphormeme’s point of view’ (Musolff, 2004: 69–71). Other components of such a terminological toolkit might be concepts such as ‘discourse metaphor networks’ (Zinken et al., 2008), the relationship between collective
182
Roslyn M. Frank
cognition and individual activity, and the distributed nature of language (Bernárdez, 2008), as well as the heterogeneously distributed nature of ‘cultural conceptualizations’ (Sharifian, 2003, 2008). Over the past decade, population approaches to language have become more common, inspired, in part, by the writings of Dawkins (1976, 1982) with his memes, replicators and vehicles and, later, by Hull (1988) with his discussion and revision of Dawkins’ theoretical framework, specifically its application to the study of the evolution of primarily scientific concepts, and more concretely, to the way in which the scientific model of a given group of scientists is elaborated and evolves over time. However, along with the diffusion of the concept of the meme has come significant criticism of the basic tenets of memetics, or lack of said tenets, as well as a lack of consensus concerning what a meme actually is and what memetics really stands for as a field of research. Also, the proliferation of studies in which ‘mind viruses’ (Dawkins, 1991) or other forms of ‘epidemiological’ transmission are present, as conceptual frames of analysis, has brought into question the effectiveness of the original gene-meme-virus analogical construct (Deacon, 2004; Gatherer, 1998; Sperber, 1990, 2000; Wilkins, 2005). For example, Chilton has explored the possible applications of the meme and virus analogies to the analysis of metaphor and in the process accepts, at least momentarily, the (misplaced) agency attributed to the gene-meme-virus concept: But here I would hypothesize that conceptual constructs become meme-like and ‘infect’ the mind (under the right social conditions) when they have complex blending potential that recruits fundamental knowledge domains along with the core mechanisms of the metaphor. There is a further ingredient that seems to go along with textualized memes of this kind – the delivery of some kind of credibility assurance and epistemic warrant. (Chilton, 2005: 40) Picking up on the implications of the virus analogy for metaphor studies Chilton proposes a new name for it, ‘ideational epidemiology’, and then returns to the key question, that of agency: [I]deational epidemiology will study the spread of ideas in the population. Over time the distribution may change – may shrink or spread, so ideational epidemiology will be interested in patterns of spread and retreat. [ ... ] Why do some ideas or idea-clusters propagate more than others? (Chilton, 2005: 17)
Metaphors of Discourse Evolution
183
Chilton (2005: 41) concludes his discussion on a more circumspect note, stating that: ‘if there is such a thing as meme propagation one of its main modes of operation lies in the properties of metaphorical expressions [ ... ]’. Or, to bring this statement more in line with the arguments laid out in this chapter, we might take it a step further, and assume that the recognizably vague referentiality of the term ‘meme’ is better understood as a kind of ‘metaphor’ (in its broadest sense), ‘analogy’ (Musolff, 2004: 56), or a situated ‘cultural conceptualization’ (see Sharifian, 2008). In other words, conceptual constructs, such as metaphors and analogies, are constantly shifting entities that are constructed, deconstructed and reconstructed as they expand and contract in terms of their spreading activation into new domains or as they retreat from old ones. While these old subsystems can act to build up network connectivity, emergent nodes with slightly new (expanded or contracted) meanings can attach themselves and move into place. And, as Chilton (2005) has suggested, under the right social conditions the end result can be a highly entrenched and enduring discourse metaphor formation. Thus, discourse metaphor network analysis can serve to highlight processes of change, stabilization and destabilization, as well as conflict, in which ‘alternative definitions and perspectives struggle against each other, producing brief and tenuous moments of stasis rather than monolithic, permanent formations. This view of discourse is useful because it recognizes the ongoing tensions and opposing forces, rather than the moments of apparent stability, as the most salient features’ (Henze, 2004: 315), shaping the interpretative strategies brought into play for understanding a given metaphor at a specific juncture in time as well as over longer periods of time. In this way a discourse metaphor can provide a localizable framework of interpretation which, through explicit specification, can be assigned an upper temporal boundary, a point of departure for the analysis of its evolution (Musolff, 2004, 2008). Furthermore, a closer look at the complex interactions taking place in the subdynamics of a given discourse metaphor will reveal how modifications are brought about and the way that various types of attractors can operate upon each other. In some cases, given the socio-cultural situatedness of the language agents, the resulting environmental resonances may contribute to stabilizations within networks making up the overall meaning-making system. Moreover, a confluence of opinions, beliefs and motives can allow certain nodes to be selected for globalization, that is, discursive prominence can be given to a specific node or
184
Roslyn M. Frank
cluster of meaning(s) within the formation, for example a node produced by the conflation of the notions of ‘language’ and ‘species’ or analogies between ‘memes’ and ‘discourse metaphors’. These shifts in salience are the end result of aggregate actions of individual language agents operating at the local level over time – and these can be integrated into the discourse metaphor formation at a global level. Thus, a discourse metaphor network is not a monolithic entity. Instead, it should be viewed more as a phenomenon whose dynamic structure is always (somewhat) unstable – always in flux – as a result of the continual process of re-negotiation taking place between competing voices, microstructures which in turn must operate in local environments conditioned by the top-down influence of (pre-)existing emergent globalized macrostructures. Stated differently, in the case of discourse metaphors, particularly those used in scientific fields, we need to take into consideration the concrete socio-cultural situatedness of individual language agents and the interpretive (epistemic) communities to which they belong. Because the agents’ situatedness inevitably places them within a given location and time frame, along with the heterogeneously distributed nature of cultural conceptualizations in general, any given set of language agents operating locally can appropriate and employ anachronistic interpretive frameworks and/or conceptualizations that are partial or limited in some respect, for example conceptualizations that are not necessarily shared by members of the ‘expert’ scientific community. In sum, there are different levels of awareness of the historically conditioned use of the terms in question and hence different understandings of their accepted meanings.
4
Conclusions
In summary, my general observations concerning ‘gene-centric’ and ‘meme-centric’ analogies and the ‘species’ or ‘population’ models of language should not be understood as a rejection of the views on language change put forward by researchers such as Croft (2000, 2002, 2006) and Mufwene (2001, 2005). Rather, I offer these comments in an attempt to identify ways in which such models might be modified, supplemented and their explanatory power increased by the adoption of the heuristic of complex systems thinking: language conceptualized as a CAS. Indeed, Croft has already set forth the groundwork for an evolutionary model and his own research already integrates many aspects of complex systems thinking. Exactly how this methodological and
Metaphors of Discourse Evolution
185
theoretical revision might be accomplished is beyond the scope of this chapter, although hopefully the topic will be taken up and elaborated upon in more depth in the future. In conclusion, at this stage those of us who are interested in exploring discourse metaphors find ourselves actively engaged in developing a flexible methodology and at the same time endeavouring to construct theoretical framework(s) appropriate for this new subfield of Cognitive Linguistics and Critical Discourse Analysis. This juncture presents us with a unique opportunity to reflect on meta-theoretical issues. It is an opportunity that could allow us to join with the larger community of cognitive scientists who are exploring the role of language in cognition as well as the situated and collectively distributed nature of cognition and language evolution in general, the latter notion being understood as referring both to the origins and evolution of language as well as to the cognitive and cultural processes that give rise to language change. There are significant benefits that would accrue by adapting a CAS modelling technique for conceptualizing language and more specifically for analysing discourse metaphors. One of the most significant advantages would be the fact that by using a framework whose terminology is already recognized across many disciplines of cognitive science we would obtain a kind of passport that would allow discourse metaphor research to more readily cross these disciplinary boundaries and propagate, hopefully, in a synergistic fashion. At the same time, the adoption and application of CAS terminology and associated concepts, for example ‘circular causality’, ‘cultural conceptualizations’ and the ‘extended mind’ (see Clark, 1997; Clark and Chalmers, 1998), to linguistic data would allow us to begin communicating more freely in what is already rapidly becoming the methodological lingua franca of many areas of the social and behavioural sciences. Stated differently, the adoption of a CAS framework would produce a larger, more expansive conceptual platform for research into discourse metaphor networks. For those of us who are concerned with discourse metaphor networks and modelling cultural conceptualizations – and their interactive role within language and culture – we are faced with a challenge that does not differ significantly from the difficulties that others have confronted when attempting to model ‘cultural evolution’ (see Sperber and Claidière, 2006; Sperber and Hirschfeld, 2003). Unfortunately, in some studies dealing with evolutionary models of ‘culture’, at times the ‘information’ metaphor (found in the ‘conduit’ framework) tends to weaken the logic of the arguments proffered, a process by which
186
Roslyn M. Frank
there is a reification of an entity called ‘information’ which then is ‘communicated’ or otherwise ‘reproduced’ in what is often a contextual void, a kind of argumentation also typical of ‘memetics’ which assigns ‘agency’ to the ‘meme’ in a false analogy to the agency that is assumed to be located in an entity called a ‘gene’. Moreover, the problems are not simply those related to understanding language change, for example the shifts that take place over time in discourse metaphor networks, but also and perhaps more importantly to discovering the nature of the mechanisms governing the stability of linguistic meaning structure(s); how a particular self-organizing system both adapts and yet at the same time maintains a kind of equilibrium that in turn keeps ‘vital’ structures safely in place and functioning. In sum, CAS thinking, understood as a cross-disciplinary research framework, is in circulation across many subfields within the biological and information sciences and perhaps more importantly, it is gradually gaining ground in many other fields now loosely comprised by the term ‘cognitive sciences’. As Zinken (2007: 462) has observed, at this juncture it is important for us to explore these methodological issues carefully so that ‘the cognitive linguistic study of figurative language [can] enter fully into the debates of the cognitive sciences (see also Döring and Nerlich, 2005).
Note 1. For a similar, although strictly academic discussion of ‘meme agency’, see Heylighen ( 1998).
References Banks, K. (this volume). ‘Metaphors and concepts: the evolution of the body politic and the body natural in late sixteenth-century France’. Bernárdez, E. (2008). ‘Collective cognition and individual activity: Variation, language and culture’. In R. M. Frank et al. (2008b) (pp. 137–66). Bono, J. J. (1990). ‘Science, discourse and literature: the role/rule of metaphor in science’. In S. Peterfreund (ed.), Literature and Science: Theory and Practice (pp. 59–89) (Boston: Northwestern University Press). Chilton, P. (2005). ‘Manipulations, memes and metaphors: the case of Mein Kampf’. In L. de Saussure and P. Schultz, (eds.), New Perspectives on Manipulative and Ideological Discourse in Pragmatics and Discourse Analysis (pp. 15–43) (Amsterdam: John Benjamins). Clark, A. (1997). Being There: Putting Brain, Body and World Together Again (Cambridge, MA/London: The MIT Press). Clark, A., and Chalmers, D. (1998). ‘The extended mind’, Analysis, 58, 7–19.
Metaphors of Discourse Evolution
187
Cowling, D. (this volume). ‘ “Neither a borrower nor a lender be”: linguistic mercantilism in Renaissance France’. Croft, W. (2000). Explaining Language Change: An Evolutionary Approach (Essex: Pearson Education). Croft, W. (2002). ‘The Darwinization of linguistics’, Selection: Molecules, Genes, Memes. 3, 1, 75–91. Croft, W. (2006). ‘The relevance of an evolutionary model to historical linguistics’. In O. Nedergård Thomsen (ed.), Competing Models of Linguistic Change: Evolution and Beyond (pp. 91–132) (Amsterdam/Philadelphia: John Benjamins). Dawkins, R. (1976). The Selfish Gene (Oxford: Oxford University Press). Dawkins, R. (1982). The Extended Phenotype: The Gene as a Unit of Selection (Oxford: Oxford University Press). Dawkins, R. (1991). Viruses of the mind. http://cscs.umich.edu/~crshalizi/ Dawkins/viruses-of-the-mind.html [accessed 22 January 2007] Deacon, T. W. (2004). ‘Memes as signs in the dynamic logic of semiosis: beyond molecular science and computation theory’. In K. E. Wolff, H. D. Pfeiffer, and H. S. Delugach (eds.), Conceptual Structures at Work. 12th International Conference on Conceptual Structures (pp. 17–30) (Berlin: Springer–Verlag). Döring, M. and Nerlich B. (2005). ‘Assessing the topology of semantic change: from linguistic fields to ecolinguistics’. Logos and Language: Journal of General Linguistics and Language Theory, 6, 1, 55–68. Frank, R. M. (2003). ‘Shifting identities: the metaphorics of nature–culture dualism in Western and Basque models of self’, metaphorik.de, 04/2003, 66–96. http://www.metaphorik.de/04/frank.pdf [accessed 22 January 2007] Frank, R. M. (2005). ‘Shifting identities: a comparative study of Basque and Western cultural conceptualizations’. Cahiers of the Association for French Language Studies, 11, 2. 1–54. http://www.afls.net/Cahiers/11.2/Frank.pdf [accessed 22 January 2007] Frank, R. M. (2008a). ‘The language–organism–species analogy: a complex adaptive systems approach to shifting perspectives on “language” ‘. In R. M. Frank et al. (2008b) (pp. 215–62). Frank, R. M., Dirven, R., Ziemke T., and Bernárdez, E. (eds.) (2008b). Body, Language and Mind. Vol. II. Sociocultural Situatedness (Berlin: Mouton de Gruyter). Gatherer, D. (1998). ‘Why the thought contagion metaphor is retarding the progress of memetics’, Journal of Memetics – Evolutionary Models of Information Transmission, 2. http://cfpm.org/jom-emit/1998/vol2/gatherer_d.html Heylighen, F. (1998). ‘What makes a meme successful?’ In Proceedings of the 16th International Congress on Cybernetics (pp. 423–18) (Namur: Association International de Cybernétique). Hellsten, I. (2005). ‘From sequencing to annotating: extending the metaphor of the book of life from genetics to genomics’, New Genetics and Society, 24, 3, 283–97. Henze, B. (2004). ‘Scientific definition in rhetorical formations: race as “permanent variety” in James Cowles Prichard’s’, Ethnology’. Rhetoric Review, 23, 4, 311–31. Hilferty, J., and Vilarroya, Ó. (2008). ‘In search of development’. In R M. Frank et al. (2008b) (pp. 197–294).
188
Roslyn M. Frank
Hull, D. L. (1988). Science as a Process: An Evolutionary Account of the Social and Conceptual Development of Science (Chicago and London: University of Chicago Press). Keller, E. F. (2000). The Century of the Gene (Cambridge, MA: Harvard University Press). Lansing, J. S. (2003). ‘Complex adaptive systems’, Annual Review of Anthropology, 32, 183–204. Maasen, S. and Weingart, P. (1995). ‘Metaphors – Messengers of meaning: a contribution to an evolutionary sociology of science’, Science Communication, 17, 9–31. Maasen, S., and Weingart, P. (2000). Metaphors and the Dynamics of Knowledge (London/New York: Routledge). Moss, L. (2004). What Genes Can’t Do (Cambridge, MA/London: The MIT Press). Mufwene, S. S. (2001). The Ecology of Language Evolution (Cambridge: Cambridge University Press). Mufwene, S. S. (2005). ‘Language evolution: the population genetics way’. In G. Hauska (ed.), Gene, Sprachen und ihre Evolution (pp. 30–52) ( Regensburg: Universitätsverlag). http://humanities.uchicago.edu/faculty/mufwene/ publications/languageEvolution-populationGeneticsWay.pdf [accessed 92 January 2007] Musolff, A. (2004). ‘Metaphor and conceptual evolution’. metaphorik.de, 07/2004, 55–75. http://www.metaphorik.de/ Musolff, A. (2008). ‘The embodiment of Europe: how do metaphors evolve?’ In R. M. Frank et al. (2008b) (pp. 301–85). Musolff, A. (this volume). ‘Metaphor in the history of ideas and discourses: how can we interpret a medieval version of the body-state analogy?’ Nerlich, B. and Clarke, D. D. (1988). ‘A dynamic model of semantic change’. Journal of Literary Semantics, 17, 2, 73–90. Nerlich, B. and I. Hellsten (2004). ‘Genomics: shifts in metaphorical landscape between 2000 and 2003’. New Genetics and Society, 23, 3, 255–68. Sharifian, F. (2003). ‘On cultural conceptualizations’. Journal of Cognition and Culture, 3, 187–207. Sharifian, F. (2008). ‘Distributed, emergent cultural cognition, conceptualisation and language’. In R. M. Frank et al. (2008b) (pp. 109–36). Sperber, D. (1990). ‘The epidemiology of beliefs’. In C. Fraser and G. Gaskell (eds.), The Social Psychological Study of Widespread Beliefs (pp. 25–44) (Oxford: Clarendon Press). Sperber, D. ( 2000). ‘An objection to the memetic approach to culture’. In R. Aunger (ed.), Darwinizing Culture: The Status of Memetics as a Science (pp. 163–73) (Oxford: Oxford University Press). Sperber, D., and Claidière, N. (2006). ‘Defining and explaining culture (comments on Richerson and Boyd, Not by Genes Alone)’, Biology and Philosophy (published online) (25 May 2006). Sperber, D. and Hirschfeld, L. A. (2003). ‘The cognitive foundations of cultural stability and diversity’. Trends in Cognitive Sciences, 8, 1, 40–6. Steels, L. (1999). ‘The puzzle of language evolution’. Kognitionswissenschaft, 8, 4, 143–50. http://www.csl.sony.fr/downloads/papers/1999/steels-kogwis1999. pdf [accessed 22 January 2007]
Metaphors of Discourse Evolution
189
Steels, L. (2004). ‘Analogies between genome and language evolution’. In J. Pollack, M. Bedau, P. Husbands, T. Ikegami, and R. A. Watson (eds.), Proceedings of Artificial Life IX: Proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems (pp. 200–06) (Cambridge, MA: MIT Press). Strohman, R. (1997). ‘Epigenesis and complexity: the coming Kuhnian revolution in biology’. Nature Biotechnology. 15, 194–200. http://bialystocker.net/files/ kuhn.pdf [accessed 22 January 2007] Strohman, R. (2001). ‘Human genome project in crisis: where is the program of life?’ http://www.biotech-info.net/StrohmanMarch09.pdf [accessed 22 January 2007] Wikipedia (2007). ‘Meme’. http://en.wikipedia.org/wiki/Meme [accessed 14 February 2007] Wilkins, J. S. (2005). ‘ “Meme” a new “idea”? Reflections on Aunger’. Biology and Philosophy, 20, 585–98. Zavadil, J. (this volume). ‘Bodies politic and bodies cosmic: the Roman Stoic theory of the ‘two cities’. Zinken, J. (2007). ‘Discourse metaphors: The link between figurative language and habitual analogies’. Cognitive Linguistics, 18, 3, 445–66. Zinken, J., Hellsten, I. and Nerlich, B. (2008). ‘Discourse metaphors’. In R. M. Frank et al. (2008b) (pp. 363–85).
12 ‘Neither a borrower nor a lender be’: Linguistic Mercantilism in Renaissance France David Cowling
Introduction This chapter explores the extent to which the distinctive figurative language used by linguistic purists in sixteenth-century France is underpinned by what might be termed a mercantilist conception of linguistic exchange during the period of early capitalism. We explore the everyday experience of the French readers of Henri Estienne’s denunciations of what he considered to be the corrupting influence of Italian on the French language, with specific reference to their often hostile attitude towards expatriate Italian bankers and financiers, and examine how Estienne seeks to appeal, through a series of carefully chosen metaphors of economic exchange, to such ‘common ground’ knowledge. While Polonius’ words to Laertes ‘neither a borrower nor a lender be’ can hardly be stretched to constitute a comment on the contemporary state of the English language, the advice that they contain (‘For loan oft loses both itself and friend, And borrowing dulls the edge of husbandry’; Shakespeare [1998] Hamlet Act I, Scene 3) seems apposite as a description of the particular concerns of those who sought, over the course of the sixteenth and seventeenth centuries, to limit or, indeed, halt and turn back the flow of foreign loanwords into their national language. Such early linguistic purism was by no means restricted to France, and the distinctive forms of polemical language that characterize it crossed cultural boundaries with disarming ease, especially given the fact that they were frequently used to castigate
190
‘Neither a borrower nor a lender be’ 191
borrowing from the very language from which they had themselves been borrowed. Two more quotations from contemporary English writers will set the tone for my discussion: I am of this opinion that our own tung shold be written cleane and pure, unmixt and unmangeled with borowing of other tunges, wherin if we take not heed by tijm, ever borowing and never payeng, she shall be fain to keep her house as bankrupt. For then doth our tung naturallie and praisablie utter her meaning, when she bouroweth no counterfeitness of other tunges to attire her self withall, but useth plainlie her own [...]. (Sir John Cheke, letter to Sir Thomas Hoby prefaced to Hoby’s translation of The Courtier (1561); cited in Baugh and Cable, 2002: 217) Among all other lessons this should first be learned, that wee never affect any straunge ynkehorne termes, but to speake as is commonly received: neither seeking to be over fine, nor yet living over-carelesse, using our speeche as most men doe, and ordering our wittes as the fewest have done. Some seeke so far for outlandish English, that they forget altogether their mothers language. And I dare sweare this, if some of their mothers were alive, thei were not able to tell what they say: and yet these fine English clerkes will say, they speake in their mother tongue, if a man should charge them for counterfeiting the Kings English. (Thomas Wilson, Arte of Rhetorique (1553), ‘Plainnesse, what it is’; cited in Baugh and Cable, 2002: 218) We meet here (and not for the last time) the distinctive language of the early purist and, in particular, some of the metaphors that become commonplace in discussions of language contact and influence: cleanliness, purity, integrity and wholeness, along with a characteristic emphasis on metaphors of – dysfunctional – economic exchange. For John Cheke, continued borrowing of foreign terms without repayment will result in linguistic bankruptcy, and the act of borrowing itself entails a disfigurement of the language that is akin to counterfeiting. Wilson, in his own contribution to the notorious ‘inkhorn’ controversy (primarily, but not exclusively, concerned with borrowing from Latin), denounces those that seek out ‘outlandish’ (i.e. foreign) words as counterfeiters of the King’s English. While these metaphors may raise a smile today, their ‘entailments’ (or logical consequences; see Lakoff and Johnson, 1980: 157–8; Kövecses, 2002: 93–105) are more serious, and betray the
192
David Cowling
undercurrent of xenophobic hostility that characterizes much early purism. Early modern monarchs reserved cruel and unusual punishment for counterfeiters, whose activities were perceived to undermine the integrity of the country’s currency (whose devaluation was, of course, the frequently exercised privilege of the monarch); the actions of linguistic counterfeiters, Wilson suggests, represent an analogous act of lèse majesté. My intention in this chapter is to explore this link between linguistic borrowing, economic exchange and political threat in the context of French linguistic purism of the later sixteenth century, and, in particular, in the work of the celebrated humanist and hellenist, compiler of the monumental Thesaurus linguae graecae of 1572, Henri Estienne (1531–1598), who devoted a series of works in the vernacular between 1565 and 1579 to a virulent denunciation of what he saw as the corrupting influence of the Italian language on French (Estienne 1853, 1896, 1972; Clément, 1898). What I wish to suggest is that, for Estienne, the free exchange of linguistic material in the form of borrowed words is just as pernicious as unregulated economic exchange between France and Italy, which French public opinion perceived as highly damaging to French interests and, ultimately, the wealth and well-being of the French people. It has been said of Estienne that he was an ‘unrequited political theorist’ (Hope, 1971: 231): it is clear that his status as a Huguenot coloured his view of the influence of Italian Catholics, chief among them the Queen Mother Catherine de Medici, at the French royal court; what I want to explore here is the extent to which he might also be claimed to be a mercantilist avant la lettre.
1 Linguistic purism in the early modern period Before going any further, however, I need to give a clearer account of early modern linguistic purism. For George Thomas, writing in 1991, linguistic purism is: ...the manifestation of a desire on the part of a speech community (or some section of it) to preserve a language from, or rid it of, putative foreign elements or other elements held to be undesirable [...]. It may be directed at all linguistic levels but primarily the lexicon. Above all, purism is an aspect of the codification, cultivation and planning of standard languages. (Thomas, 1991: 12) A striking feature of the history of purism is the recurrence of a limited stock of metaphorical models to characterize the process of borrowing
‘Neither a borrower nor a lender be’ 193
(itself a metaphorical model, of course) and those that practise and oppose it, both across cultures and over time. Purists in the early modern period style themselves as ‘millers’ (separating the husk from the wheat), ‘gardeners’ (cultivating fruit-bearing trees and pulling up weeds), ‘metallurgists’ (recovering metals from their ore and removing impurities), ‘doctors’ (removing diseased portions of the body or purifying the blood) or even ‘genealogists’ (establishing the pedigree and legitimacy of elements of a language, and ridding it of ‘bastard’ words). All such images share the notion that specialist knowledge is required to make linguistic judgements, and all may be seen as relying, at a deeper level, on a more basic metaphor of, precisely, purity and integrity, the loss of which results in decay and degeneracy (see Thomas, 1991: 31). Thomas recognizes that purism tends to occur during periods of ‘strong national sentiment’ (1991: 43), and, when such sentiment is associated with xenophobia (as it so frequently is), it almost invariably shares the same targets. The targets tend to belong to the culture that is deemed to pose a threat to the ‘home’ culture. What is most striking about early modern purism, as I have just suggested, is that the same arguments, underpinned by the same metaphorical models, are used indiscriminately against a variety of targets from culture to culture. One model that Thomas does not, however, discuss is that of linguistic borrowing as a form of economic exchange, and the role of the purist as an inhibitor of that process. While we need to wait until the end of the sixteenth century and the reign of Henry IV to see mercantilism emerge as a clearly articulated national policy in France in the work of Antoine de Montchrestien, Barthélemy Laffemas and Turquet de Mayerne (see Desan, 1992: 11; Heller, 2003: 212–18), its basic principles, namely that the wealth of the nation depended upon its reserves of capital in the form of gold or silver bullion, and that the government should adopt a protectionist attitude towards overseas trade, promoting local manufacture and exports (except for the export of bullion) and inhibiting imports by creating import tariffs and other trade barriers, were already evident earlier in the century. If words are seen as a form of linguistic ‘capital’, then it follows that the role of the purist is to protect them against replacement by imported foreign terms, which are almost universally viewed, in this period, as inherently unnecessary, unless referring to some undesirable characteristic seen as peculiar to the culture from which the word is being borrowed. Favourite examples of this category of pejorative terms derived from Italian, which Estienne and other purists are happy to admit, are ‘assassin’, ‘charlatan’ and ‘bouffon’ (see Clément, 1967: 137, 344). The logical entailment of this metaphor,
194
David Cowling
as John Cheke was of course aware, is that a nation that fails to keep a positive balance of linguistic payments risks bankruptcy.
2
Franco–Italian relations in the sixteenth century
In the context of sixteenth-century France, the perceived cultural (and hence linguistic) threat was that posed by Italy. When assessing the reasons for Italian linguistic and cultural influence in France in the sixteenth century, it is customary to focus attention upon the contacts between the two nations that occurred as a consequence of the Italian military adventures of successive French monarchs, starting with Charles VIII in 1494, and upon the presence of Italian émigrés at the royal court of France following the marriage of Catherine de Medici to the dauphin Henry in 1533. High-level contact between the intellectual elites of the fifteenth and sixteenth centuries is also privileged by those historians who foreground scholarly debates between French and Italian humanists, and the personal and professional competition for royal favour and financial patronage that fuelled them (see Simone, 1968: 86; Mann, 1971: 60; Sozzi, 1988: 100; Balsamo, 1992: 32–3, 37). What emerges less clearly in such accounts is the everyday animus that the presence of relatively well-heeled and relatively unassimilated Italian expatriate merchants and financiers in French urban communities appears to have generated. Such low-level contact, despite the often anecdotal nature of its historiographical record, is significant because of the investment made by Henri Estienne in its exploitation in the service of his polemical aims; he repeatedly attempts to tap into petty resentment and animosity of precisely this kind through his choice of metaphors drawn from his readers’ everyday experience of Italian migrants and settlers. Rather than restricting themselves to an audience of scholars and nobles, his vernacular works seem designed to strike a chord with the middle classes, whose resentment of the perceived excesses – linguistic and moral – of court life he seeks to mobilize as part of a more general attack on Italian cultural influence. It is therefore worth analysing the types of low-level contact that would have defined and reinforced, for many of Estienne’s readers, an overwhelmingly negative stereotype of the Italian ‘invader’. In his monograph Anti-Italianism in Sixteenth-Century France (2003), Henry Heller gives a new account of anti-Italian feeling in sixteenthcentury France and of the reasons for it. While high-level contact is initially privileged (Heller, 2003: 3), Heller goes on to trace the spread of anti-Italian sentiments from the educated elite to the merchant class and
‘Neither a borrower nor a lender be’ 195
thence to the Huguenots and the nobility and, finally, to the Catholic population of major urban centres. He sees anti-Italianism not as a phenomenon restricted to intellectual and cultural debate, but, more fundamentally, as ‘a conflict over money, markets, and political power’ (Heller, 2003: 5). Significantly, too, allegedly ‘rootless’ Italians could find themselves placed in the same xenophobic boat as Jews, whose financial dealings, like those of the Italian bankers and financiers, aroused suspicion and hostility in equal measure (2003: 8–9). Indeed, during the Estates-General of Blois in 1576, the Third Estate aggressively condemned both Italian influence at court and the perceived stranglehold of prominent Italian courtiers and officials over the fiscal system, thus attempting to make the Italians responsible for the financial misery of the people (2003: 10). Huguenot publicists ascribed to the Italians responsibility for religious conflicts and economic crises. It is therefore little wonder that there was a significant market for anti-Italianism of the kind published by Henri Estienne at his press in Geneva, to which he had fled in 1551 to escape the persecution of the Catholic authorities in Paris. More detailed analysis of the forms taken by contemporary popular xenophobia will be instructive in helping to establish a picture of the preconceived attitudes and prejudices that Estienne’s careful choice of metaphors was designed to reinforce and radicalize.
3 The experiential basis for Estienne’s metaphors of economic exchange The growth of Italian influence in France in the latter part of the fifteenth and early sixteenth centuries may be ascribed both to the development of the financial markets and of trade more generally, and to the dynastic and cultural interests of successive French kings. Whereas Louis XI (1461–1483) granted fiscal and commercial privileges to Italian merchants and financiers in order to cement the position of Lyon as the leading financial and commercial centre of western Europe, Louis XII (1498–1515) and Francis I (1515–1547) invited Italian artists and intellectuals to their courts as a means both of asserting their awareness of the newest cultural and aesthetic trends, and of enhancing the desired links between the kingdom of France and its new (albeit temporary) Italian territories, whose conquest had been justified by dynastic connections and hereditary (or acquired) claims. During the early decades of the sixteenth century, the main concentration of expatriate Italian bankers and merchants, most of whom originated from Florence, Lucca, Milan and Genoa, was situated in Lyon. This milieu, as Heller reminds
196
David Cowling
us (Heller, 2003: 34), was essentially diasporic in nature, with immigrants retaining a close connection to their homeland, and naturalization relatively rare. This relative lack of integration of the immigrant population, despite their ostentatious and extravagant involvement in civic festivities, helps to account for a Calvinist backlash directed towards the Italians by such writers as Antoine Du Pinet. As Estienne was later to do, Du Pinet condemned in 1564 the extravagant sartorial fashions that he saw as imports from Italy; in addition to being morally harmful, the Italian trade in silks and luxury goods was presented as drawing gold out of the kingdom as the French paid Italian merchants – who, of course, retained close family and financial connections with their home city – inflated prices for imported Italian goods (see Heller, 2003: 30). In addition, Italian bankers, who offered temptingly high rates of interest to investors, were tainted by their association with the Grand Parti, a syndicate of mainly Lyonnese investors lending to the French crown with royal tax revenues as collateral, which notoriously collapsed in 1557, causing the ruin of numerous private individuals, including many widows and orphans (see Hauser, 1936: 312–13; 316–17; Desan, 1992: 24). Indeed, Italian bankers are frequently presented as inherently unreliable, and prone to bankruptcy. The seriousness of the problem of bankruptcy in the period and its unpopularity with the Third Estate are indicated by the demands made at successive Estates-General in 1560 and 1576 that those found guilty of fraudulent bankruptcy should be ‘punis extraordinairement et capitalement’ (Hauser, 1936: 318–19). Such accusations of peddling undesirable and unnecessary goods that served only to impoverish the French nation while playing fast and loose with French investors’ money, which came to dominate anti-Italian discourse in the latter part of the sixteenth century, and which have at their origin an essentially economic rivalry between French merchants and craftsmen and the expatriate Italian urban elite in cities such as Lyon, clearly drew upon the pervasive anti-Italianism to be found among France’s intellectual elite (see Heller, 2003: 35). Indeed, the commercial and cultural domains are frequently brought into contact in metaphorical explorations of analogies between them (see Desan, 1993: 15–16): the French poet Joachim Du Bellay, for instance, asserts that the importation of Italian cultural ‘goods’ parallels that of substandard Italian products, with the importers seeking to enrich themselves through the sale of inferior merchandise (see Du Bellay, 1908–85, vol. 6: 115–18). The French humanist Etienne Dolet, complaining at Francis I’s patronage of the Italian Giulio Camillo, remarks on the alleged French predilection for seeking out that which is foreign, novel and, typically, very expensive
‘Neither a borrower nor a lender be’ 197
to the neglect of the things that they already have at home (Dolet, 1982: 100). Interestingly, Dolet’s interest in the use of the vernacular, as attested by his treatises on translation and orthography, may be seen as part of an attempt to supplant the cultural hegemony of the Italian humanists through the development of a thriving market for French vernacular printed books, a market into which Estienne sought to break with his vernacular attacks on linguistic borrowing from Italian (see also Worth, 1988: 12–13). Dolet and Estienne thus address themselves to the same readership: the educated (but not necessarily classically educated) middle classes, whose professional activities, be they mercantile, commercial, financial or administrative, brought them into contact – and, frequently, competition – with Italian immigrants. Indeed, from the 1540s and into the 1550s, the number of complaints about unfair competition by Italian merchants, who were granted monopolies in Lyon and then Paris, and who were perceived to be in unfair competition with French merchants, increased significantly. The frustrations of this class, experienced in commercial dealings and as a result of the fiscal exactions of the French crown (in which, as we will see, Italian financiers were frequently implicated), are addressed and amplified by Estienne in his choice of metaphors used to describe the Italian cultural influence, and, in particular, Italian influence – real and perceived – on the French language. Henri Estienne’s campaign, conducted in the French vernacular, against the perceived Italian influence on French society, manners and language began in earnest in the early 1560s, and culminated in the aftermath of the Saint Bartholomew’s Day massacre of 1572. It is instructive to attempt to gauge public feeling towards the Italians in this period by means of a brief review of significant events immediately preceding and following the publication of Estienne’s works. At the outbreak of the first War of Religion in 1562, a minority of Huguenots seized control of the city of Lyon, established a Calvinist town council and consistory, and held it against royal forces until the early summer of 1563, only being expelled four years later (Heller, 2003: 29). This rebellion against royal (and Catholic) authority was presented as a service to the true interests of the king and an attempt to free him of the corrupting foreign influence of the Guise family, compromised for many by their close relationship with the Spanish crown, and the papacy. Huguenot political nationalism at this stage thus defined itself, to a large extent, as a response to foreign forces that were themselves intent on disrupting the peace of the kingdom; within the city of Lyon and elsewhere, such sentiments found expression in hostility against
198
David Cowling
the Italian population. At the same time, the perception that the French church was dominated by Italian prelates offended both Gallicans and Huguenots alike. The Estates-General of Orléans in 1560 had called for an investigation into the profligate dealings of Henry II’s Italian financiers, whose brief, ironically, had been to raise, through borrowing (from Italians) and taxation (of the French people), the enormous sums needed to finance the king’s military campaigns in Italy (Heller, 2003: 50). Indeed, the Huguenot Condé, in his Protestation de Monseigneur le Prince de Condé of 1568, attempted to justify the Huguenot rebellion of the 1560s by asserting that its chief aim had been to persuade the king of France to call another Estates-General in order to relieve the people of the fiscal oppression of the Italians (Heller, 2003: 50). This opposition to the heavy burden of taxation required to fund both the increasing extravagance of the royal court, with its insatiable appetite for credit, and foreign military adventures manifested itself also in Paris in the years preceding the Saint Bartholomew massacre. The Venetian ambassador at the court of France reported in 1569 that Italians had not been safe on the streets of Paris for two years (Heller, 2003: 80). A riot in Paris in June 1572 (two months before the massacre of Huguenots) had been directed against highly placed Italian courtiers and the Queen Mother herself, all of whom were accused of the blood libel traditionally directed against the Jewish population; this analogy between the Italians and the Jews in the popular consciousness, although, perhaps, initially surprising, was clearly rooted in the perception that both groups were engaged in usury and the exploitation of the common people. Further riots and attempted massacres of prominent Italians resident in Paris took place in 1575 and 1578, with students combining forces with commoners in the former (Heller, 2003: 80–4). While the Parisian student population had a longstanding reputation for agitation and rowdiness, its involvement in such anti-Italian activities is significant, indicating that the educated classes were not exempt from the sort of knee-jerk xenophobia that could easily spill into violence against foreigners. Estienne’s anti-Italian writings were in all likelihood pitched at a readership that was, in part at least, already radicalized and happy to have its existing prejudices reinforced and harnessed to a programme of more ambitious cultural contestation.
4
Estienne’s metaphors for linguistic exchange
In making the case for the ‘defence’ of the French language against a perceived influx of words and other linguistic material borrowed
‘Neither a borrower nor a lender be’ 199
from Italy, Henri Estienne consistently privileged, in his persuasive and satirical work, a range of commonplace metaphors that were evidently designed to appeal to such ostensibly self-evident concepts as the value of good health, ‘purity’ (as opposed to harmful adulteration), sound personal and domestic finances, and plain cooking. The choice of such metaphors was motivated, I would suggest, by two factors: first, congruence with the most salient aspects of perceived Italian dominance in financial and fiscal affairs, and influence on courtly fashion and manners, as set out above; and, secondly, a consideration of ‘audience design’, in other words, the attempt to tailor the metaphors used to the lived experience of the targeted readership (see Bell, 1984). The effectiveness of such commonplace metaphors may, I think, be explained most readily by reference to the notion of ‘common ground’, the contextual information shared by speaker/writer and listener/reader that enables the latter to decode metaphorical utterances accurately, even when the linguistic formulation of the metaphor does not appear to provide all the information needed for successful interpretation. Raymond Gibbs gives the example of a conversation about politics between two speakers of American English (Gibbs, 1994: 113–14): if one refers to a mutual female acquaintance as an ‘elephant’, this serves to indicate that the acquaintance is a member of the Republican party (whose symbol is, of course, an elephant). I would suggest that, in later sixteenth-century France, use of metaphors playing on financial impropriety, extravagance of dress or disguise, immorality of manners and indigestibility of food all made an analogous appeal to the ‘common ground’ of a French readership – not restricted to the persecuted Huguenot population – dissatisfied with the perceived political, financial and cultural influence of an Italian émigré group and ready to credit them with such vices. I will restrict myself in what follows to a consideration of one such set of metaphors, those dealing with financial practices and economic exchange. In the preface to his Traicté de la conformité du langage françois avec le grec of 1565, Estienne sets out his ideological position in the linguistic debate and, through his choice of metaphors, situates that position in the lived experience of the readership (Estienne, 1853: 17–46). Modern theoreticians of metaphor have noted the effectiveness of metaphor as a persuasive tool precisely because it is able to tap into and reinforce conventional modes of thought, to appeal to the already-known as a means of making sense of complex reality (see, for example, Lakoff and Johnson, 1980: 157). Estienne, in a lengthy section of his preface devoted to a description of the kind of language that he will not be discussing in his treatise, sets out to appeal to the existing prejudices
200 David Cowling
of his readers in respect of Italian influence on the sumptuary habits of French courtiers: But, before I begin, I wish to warn my readers that I do not intend to refer to that gaudy form of the French language that changes its livery every day depending on how our friend the courtier or our friend the lawyer choose to dress it up. Nor will I be referring to the kind of French that is disguised, masked, affected, made up and confected according to the wishes of all those other people who are as obsessed with novelty in their speech as they are with novelty in their clothes. I shall leave to one side the French that is italianised and hispanised. The reason for this is that this disguised French, by changing its dress, has at the same time lost (at least in part) the similarity that it used to display with the rich and beautiful Greek language. (My translation; original in Estienne, 1853: 20) By deploying the metaphorical models of extravagant dress and disguise, Estienne makes an appeal both to a well established tradition of French anti-aulic satire (see Smith, 1966: 51, 61, 84–5, 157) and, in particular, to Huguenot disapprobation of the perceived excesses of the French royal court, dominated (as popular opinion held) by the archCatholic faction around the Queen Mother, Catherine de Medici. In the same way that frequent changes of ‘livery’ were seen as a symptom of the supposed moral degeneracy of the French court and of the financial profligacy of its courtiers (fuelled by loans from Italian financiers), supposedly unnecessary changes to the language, specifically the borrowing of words from the modern vernaculars, as opposed to Greek, Latin, and earlier states of the French language itself, are presented as inherently undesirable and, indeed, harmful to it, since they distort its ‘true’ nature. This notion of distortion, which relies for its effectiveness on metaphors of masking and their overwhelmingly negative connotations for a readership suspicious of lavish court entertainment and the figure of the Italian charlatan, has an additional ideological burden, as is made clear in Estienne’s reference to Greek: by masking the close linguistic relationship that Estienne (whose knowledge of Greek was unparalleled in the period) claimed to perceive between that language and French, Italian influence prevents French from assuming its rightful position in the hierarchy of languages as a close relative of Greek. No-one could argue, of course, that Italian was not closer to Latin than French was; Estienne’s aim is to circumvent this difficulty by asserting,
‘Neither a borrower nor a lender be’ 201
with copious if rather (to a modern eye) unconvincing examples, that French was far closer to Greek than Italian ever could be. Aside from the obvious prestige that could be acquired for French through such an argument, there is a deeper ideological basis for the comparison between French and Greek: the latter is characterized as lending words to all other languages and borrowing words from none (Estienne, 1853: 19), thus fulfilling admirably the mercantilist conception of a perfect language. Ignoring the clear historical and cultural reasons for this state of affairs (he prefers to concentrate on Greek’s facility in the creation of neologisms), Estienne chooses to associate linguistic prestige – or what he calls ‘preeminence’ – with what we might call a positive balance of payments in the international market. This conception is underpinned by a further important model, that of borrowing. In the light of the events I have just summarized, most notably the collapse of the Grand Parti of Lyon, and the well documented anti-Italian sentiment that they had engendered, Estienne’s deployment of the metaphor of borrowing has an obvious political, as opposed to merely linguistic, resonance. The form in which he chooses to make the point, however, is clearly designed to appeal to his readers’ everyday experience: ... if we do have to borrow, why do we not reserve that honour for the two ancient languages, Greek and Latin, from which we already derive the greater part of our speech, rather than extend it to the modern vernaculars, which are, dare I say it, inferior to our own? [...] But we are like the sort of poor householder who, in order to save time, borrows things from his neighbours that he would have found at home if only he had bothered to look for them. (My translation; original in Estienne, 1853: 21–2) Borrowing implies an unequal relationship between debtor and creditor, with the latter in a position of strength, and, Estienne implies, all reasonable steps should be taken to avoid it; the French should seek to apply the same standards of thrift and prudent domestic management to their language as they do to their own households. The homely and apparently self-evident nature of these formulations belies the skill with which such commonplaces of everyday life (perceived, perhaps, in a more acute form in the context of the nascent mercantilism of later sixteenth-century France) are applied to an ideologically and confessionally loaded ‘defence’ of the French language against foreign influence. If Estienne’s warnings are not heeded, he asserts, then he fears that the French language, which has previously enjoyed such ‘vogue’
202 David Cowling
and ‘credit’, will be unable to pay back its creditors and will be obliged to ‘faire un tour de banqueroutier’ (Estienne, 1853: 32), or declare itself bankrupt: the solvency of the language depends on its ability to repay more loan words to the Italians than it has taken from them. (There is an unconscious – or, perhaps, deliberate? – irony in these lines: the words vogue, credit and banqueroute had all been borrowed from Italian.) This argument finds its logical culmination in Estienne’s Précellence du langage françois of 1579, in which he claims that, for each word that French has borrowed from Italian in the politically sensitive domain of military terms, it can adduce three or four that it has, at an earlier stage of its development, lent to that language (Estienne, 1896: 354).
5
Conclusion
Is it fair, then, to conclude that Henri Estienne is a linguistic mercantilist? He is certainly acutely conscious of the linguistic balance of payments, and uses it as a criterion for his judgement as to the relative merits of languages. He supports the enrichment of the vocabulary of French through the local manufacture of words, or at least the rediscovery of archaic terms that may usefully be reintroduced into general usage to plug a gap in the lexicon and thus remove the need for a foreign borrowing (see Desan, 1993: 110). But can we claim that he is successful in the creation of barriers to prevent borrowing in the first place? His explicitly stated wish that all foreign words be ‘banished’ from the language if they can find no-one to defend them had, as we might expect, no immediate impact. Indeed, as George Thomas reminds us, during the phase of ‘pre-standardisation’, linguistic purists ‘tend to act individually or in small, loosely organised groups. As a result, the puristic activity displays a high degree of idiosyncrasy which [militates] against its long-term impact’ (Thomas, 1991: 117). Four hundred years after his campaign, however, Estienne was hailed as a precursor by a new generation of French purists, chief among them René Etiemble, whose campaign against le franglais culminated in government legislation to inhibit the uptake of anglicisms (see Hornsby, 1998). While linguistic protectionism may be viewed as an enduring feature of the attitude of the French people towards its language, the attempt made in this chapter to reconstruct, however partially, the common ground knowledge of Estienne’s sixteenth-century readers suggests that such historically determined contextual information is essential for a proper understanding of the effectiveness of metaphor use in specific instances of discourse.
‘Neither a borrower nor a lender be’ 203
References Balsamo, J. (1992). Les Rencontres des Muses: italianisme et anti-italianisme dans les Lettres françaises de la fin du XVIe siècle (Geneva: Slatkine). Baugh, A. C. and Cable T. [1951] (2002). A History of the English Language, 5th edn (London: Routledge) Bell, A. (1984). ‘Language style as audience design’, Language in Society, XIII, 145–204. Clément, L. (1898). Henri Estienne et son œuvre française: étude d’histoire littéraire et de philologie (Paris: A. Picard; repr. Geneva: Slatkine, 1967). Desan, P. (1992). Les Commerces de Montaigne: le discours économique des ‘Essais’ (Paris: Nizet). Desan, P. (1993). L’Imaginaire économique de la Renaissance (Mont-de-Marsan: Editions InterUniversitaires). Dolet, E. (1982). Correspondance: répertoire analytique et chronologique, suivi du texte de ses lettres latines, ed. by C. Longeon (Geneva: Droz). Du Bellay, J. (1908–85). Œuvres poétiques. 8 vols, ed. by H. Chamard (Paris: Cornély). Estienne, H. (1853). Conformité du langage françois avec le grec, ed. by L. Feugère (Paris: J. Delalain; repr. Geneva: Slatkine, 1970) Estienne, H. (1896). La Précellence du langage françois, ed. by E. Huguet (Paris: Armand Colin; repr. Paris: France-Expansion, 1973) Estienne, H. (1972). Traicté de la Conformité du language françois avec le grec (1565) suivi de De Latinitate falso suspecta (1576) et de Project du livre intitulé: De la precellence du langage françois (1579) (Geneva: Slatkine Reprints). Gibbs, R. W. Jr. (1994). The Poetics of Mind: Figurative Thought, Language, and Understanding (Cambridge: Cambridge University Press). Hauser, H. (1936). ‘La Crise de 1557–1559 et le bouleversement des fortunes’. In Mélanges offerts à M. Abel Lefranc par ses élèves et ses amis (Paris: Librairie E. Droz) Heller, H. (2003). Anti-Italianism in Sixteenth-Century France (Toronto: University of Toronto Press). Hope, T. E. (1971). Lexical Borrowing in the Romance Languages: A Critical Study of Italianisms in French and Gallicisms in Italian from 1100 to 1900 (Oxford: Basil Blackwell). Hornsby, D. (1998). ‘Patriotism and linguistic purism in France: Deux Dialogues dans le nouveau langage françois [sic] and Parlez-vous Franglais?’, Journal of European Studies, XXVIII, 331–54. Kövecses, Z. (2002). Metaphor: A Practical Introduction (Oxford: Oxford University Press). Lakoff, G. and Johnson, M. (1980). Metaphors We Live By (Chicago: University of Chicago Press). Mann, N. (1971). ‘Humanisme et patriotisme en France au quinzième siècle’, Cahiers de l’Association Internationale des Etudes Françaises, XXIII, 51–66. Shakespeare, W. [1988] (1998). Hamlet. In S. Wells and G. Taylor ( eds.) William Shakespeare: The Complete Works. (Oxford: Clarendon Press). Simone, F. (1968). Umanesimo, Rinascimento, Barocco in Francia (Milan: Mursia). Smith, P. M. (1966). The Anti-Courtier Trend in Sixteenth Century French Literature, (Geneva: Droz).
204
David Cowling
Sozzi, L. (1988). ‘La Polémique anti-italienne dans l’œuvre narrative d’Henri Estienne’. In Henri Estienne: actes du colloque organisé à l’Université de ParisSorbonne le 12 mars 1987 par le Centre V. L. Saulnier (pp. 97–111) (Paris: Ecole normale supérieure). Thomas, G. (1991). Linguistic Purism (London: Longman). Worth, V. (1988). Practising Translation in Renaissance France: The Example of Etienne Dolet (Oxford: Clarendon Press).
13 Interpretations of the Body Politic and of Natural Bodies in Late Sixteenth-Century France Kathryn Banks
Introduction This chapter demonstrates that in sixteenth-century France the body politic metaphor was extremely flexible, providing answers – including contradictory ones – to a range of heated contemporary debates. As absolutist and constitutionalist conceptions of monarchy jostled for precedence, the metaphor could either locate sovereignty uniquely in the ‘head’ or disperse it through the ‘body’. At a time of civil war, it could be harnessed to support peace or, conversely, the resumption of hostilities. The healthy body could represent the unified French body politic but, as France became increasingly divided in the wars, a resemblance between disease and disharmony was highlighted as much as that between health and harmony. In addition, nature could constitute the target domain rather than the source, and interpretations of the metaphor in one domain influenced those in the other. This chapter highlights historically specific factors behind the use of the metaphor. I will show how the seventeenth century witnessed a more radical adaptation of the metaphor, as a mechanical and artificial ‘person’ provided a solution to problematic interpretations of the natural body. The human body as political metaphor can be traced back to Aristotle, Plato and beyond (Hale, 1971: 18–47 and Zavadil, this volume), and continues to be used in contemporary political discourse (Musolff, 2004: 58–71). Thinking about both nature and politics has changed enormously but, to use a metaphor from evolutionary biology, the body metaphor has evolved to adapt to new semantic or conceptual environments. This essay will examine reasons for the success of the 205
206 Kathryn Banks
body metaphor in late sixteenth-century France. These reasons are varied: for example the metaphor expresses diverse conceptions of the French constitution and diverse understandings of the French Wars of Religion. On the other hand, the factors behind the metaphor’s use may be historicized to some extent: different ones are dominant in the late sixteenth century from those in the seventeenth century or in later periods. These factors comprise not only the sorts of arguments in which the metaphor appears – for example, concerning the benefits or drawbacks of monarchy – but also the way in which a political entity is understood to be ‘like’ a natural body, in other words, the perception of the metaphor’s ‘propriety’. The case of the body metaphor suggests that while metaphors may ‘adapt themselves’ better to expressing some meanings than others, their meaning is not predetermined. It is for this reason that notions from evolutionary biology have at least a heuristic value: they emphasize the variability and adaptability of metaphors rather than their supposedly inherent suitability to express one particular meaning; they invite us to explain the reasons for healthy ‘populations’ of a metaphor while expecting that these reasons – like the metaphor itself – will vary and evolve in accordance with a wider conceptual environment. In this limited sense I respond to Susan Blackmore’s invitation to ‘take a meme’s eye view’, that is, a ‘view that looks at the world in terms of opportunities for replication’ of memes ( Blackmore, 1999: 37).1
1 Conceptions of the body politic Ernst H. Kantorowicz (1957: 207–32) argued that, from the thirteenth century, political bodies were conceived – on the model of the Church – as ‘mystical’ bodies, so that bodies politic and mystical bodies became ‘almost interchangeable notions’. The metaphor of the body thus immortalized the new, quasi-national states, and was particularly important in France where it fitted well with the traditional mysticism of French kingship. The notion of the corpus mysticum reipublicae (mystical body of the commonwealth) could exalt the king or, conversely, the constitutional forces which limited royal power. By contrast, according to Arlette Jouanna (1989: 281–312), the body metaphor inherently supported the political agents who counterbalanced royal power. Jouanna suggests that the metaphor was intrinsically tied up with support for mixed monarchy, that is, monarchy which contained some elements of aristocracy and democracy through the power of the nobles and the Estates respectively. Since the ‘limbs’ of
Bodies Politic and Natural in 16th Century France
207
the political ‘body’ played important roles, the metaphor suggested that sovereignty belonged to the whole body, and was therefore a useful tool for those who opposed nascent absolutism in late sixteenth-century France. Both of these accounts demonstrate important applications of the body metaphor. However, it should be emphasized that the metaphor served a wide variety of semantic needs in sixteenth-century France. Kantorowicz discussed bodies conceived in terms of organs such as the head and limbs, but the body natural, and hence the body politic, could also be thought about in relation to the humours of Galenic medicine. Whereas the organological body, used also as a metaphor for the eternal Church, could bestow dignity and durability upon bodies politic, a central implication of humoral theory was that humoral conflict caused illness and, ultimately, death. For early Greek thinkers, and for Machiavelli, the bodily humours represented the desires of different members of the body politic which, like humours, might conflict with one another and thereby threaten the ‘body’ as a whole (Parel, 1992: 102–54). Similarly, in his Monarchie de France, published in 1519, Claude de Seyssel reminded the young Francis I of the problem of conflicting political ‘humours’; the king should maintain the humours in harmony by satisfying all three Estates and ensuring that no one Estate exerted excessive power over the others. Seyssel did refer to bodies politic as ‘mystical bodies’ but this did not immortalize them. Rather Seyssel explicitly stated that ‘mystical bodies’ resembled ‘material human bodies’ which could die because of humoral conflict. Since bodies were ultimately subject to death, the king needed to protect France as much as possible by preventing any of its humours from transcending the others (Seyssel, 1960: 108, 154–5). As the sixteenth century progressed and France was ravaged by the Wars of Religion, depictions of the French body politic more frequently referred to disease and death, tending to present the former as a reality and the latter as an imminent threat. This is a far cry from the immortalization implied by the body metaphor in the form discussed by Kantorowicz. Similarly, the link between the body politic and mixed monarchy is not a necessary one. For Jouanna, the clearest expression of ‘the theory’ of the body politic is a 1575 justification of resistance which supported mixed monarchy and which depicted France as an imperfect body lacking some of its ‘limbs’ because princes and royal advisers were estranged from the king and thus prevented from playing their proper roles ( Anon. 1575, especially p. 98; compare Jouanna, 1989: 285–6, 291–2, 293–4). However, the metaphor was also used with contrasting
208 Kathryn Banks
implications. For example, in the following year, 1576, Arnault Sorbin responded to those engaged in resistance by asserting that the monarchy was a body of which only the monarch could be the head (Le Vray Reveille-Matin des calvinistes, cited by Church, 1969: 125–6, n. 17). Here the body metaphor reaffirms the supreme authority of the king rather than the share in sovereignty of the ‘limbs’. In fact, the body metaphor was highly adaptable and appealed widely to proponents of a variety of political views. Penny Roberts (2007: 152–64) has shown that it was used to support arguments both for war against the Protestants and also, conversely, for peace and the toleration of the Protestants: appeals were made to the king as the body’s ‘head’ or as its ‘doctor’ to ‘cure’ the body either by ridding it of its Protestant ‘poison’, ‘disease’, ‘plague’ or ‘cancer’, or, conversely, by protecting all of its constituent parts. The metaphor was also used, for example by Montaigne in his Essays, to depict the suffering of bodies politic without endorsing a particular solution or appealing to the figure of a doctor (O’Brien, forthcoming; Clark, 1970: 353). Or, the metaphor could be employed to discuss taxation and the allocation of resources: for the good of the body, the body parts needed to share nourishment, and thus the body politic should also share resources among its constituent parts.2 In addition, the clergy and jurists made competing claims to be the ‘eyes’ of the kingdom (Roberts, 2007: 149–50). Furthermore, in disputes about war and peace, the body metaphor was even more adaptable than Roberts suggests,3 since even the king’s status as ‘head’ or ‘doctor’ could be changed. The Huguenot Guillaume de Saluste Du Bartas used an innovative formulation of the body metaphor to express some of the ideas of the Protestant monarchomachs, who argued that royal power should be constitutionally limited and that kings who abused their power could legitimately be resisted. In the 1570s, following the St. Bartholomew’s massacres of 1572 in which the royal family were implicated, Protestant monarchomach theories were influential amongst both Protestants and moderate Catholics. In Du Bartas’s bestselling poem of 1578, La Sepmaine, the poet highlighted the danger of unbridled royal power by depicting the king not as head or doctor of the body politic but rather as one of the four humours. Like Seyssel, Du Bartas employed the medical notion that bodies die when one humour becomes excessively dominant; however, for Du Bartas, this forceful political humour was the king. According to Du Bartas, kings sometimes massacre the citizens of their kingdom, the ‘vassalic humours’, thus ultimately bringing about the ‘death’ of the entire body politic (Du Bartas, 1935–40, vol. II: 193–440, Day II, ll. 75–112). In Du
Bodies Politic and Natural in 16th Century France
209
Bartas’s formulation, the body metaphor undermines the notion that royal power is beneficial and that the king may ‘cure’ France’s ills, suggesting instead that he is their source.4 Thus the body metaphor was modified to serve new conceptual needs created by (a particular perception of) civil conflict and especially the St. Bartholomew’s massacre. While the body metaphor could be used to glorify the king, or to vilify the Protestants as poisoned humours, it also worked well to support criticism of the king: humoral conflict could be harnessed to imply that a ‘healthy constitution’, and indeed ‘life’ itself, were dependent upon the restraint of the dominant political ‘humour’, that is, the king. Another writer, probably Etienne Pasquier, closely imitated the entire passage (22 lines of poetry) in a 1585 Apologie pour la paix (‘Apology for Peace’), a pamphlet drawing a number of arguments for peace from analogies with the natural world.5 At around the same time, another writer with a very different – and much more aggressive – political position also found it useful to depict the king as a harmful bodily fluid. The hard-line Catholic League blamed the king for not doing enough to eradicate Protestant ‘heresy’, and became particularly powerful after the Protestant Henri de Navarre became heir to the throne in 1584. Adapting some of the arguments of the Protestant monarchomachs, League writers justified resistance to the monarchy even to the point of regicide, an act which would occur in 1589. An anonymous pamphlet in this vein in its title called for the continuation of war, ‘Origin of the Illness of France with the remedies proper to cure it, with an exhortation to the continuation of war’ (Origine de la maladie de la France, avec les remedes propres à la guarison d’icelle, avec une exhortation a l’entretenement de la guerre). The pamphlet argued that killing the king could ‘cure’ France. In order to justify this conclusion, it repeatedly depicted France as a body which needed to be ‘bled’ by the murder of the king, who was said to be allied with the body’s ‘bad humours’, namely the Protestants; readers were invited to play the role of France’s ‘doctor’, a role which had previously been that of the king himself (Anon. [1589]). The frequent use of the body metaphor in sixteenth-century France was not simply due to its ability to represent effectively one or the other rigidly defined and monolithic interpretations of, for example, kingship or France. Employing the metaphor did not inherently or consistently suggest a perception of the king or of France as glorious, or the notion that the king should, or should not, have absolute power. Instead, the recurrent use of the metaphor reflects its versatility as a tool for voicing a wide variety of answers to contemporary questions, often concerning
210
Kathryn Banks
the nature of France and its constitution. The metaphor could celebrate the body politic’s glory or, on the other hand, lament its problems; it could be used to support absolutist conceptions of monarchy as effectively as constitutionalist ones. Furthermore, while the metaphor undoubtedly shaped conceptions of the body politic and of kingship, it was also itself shaped by them: for example, contemporary apprehension of the Wars of Religion led to a new emphasis being placed upon the body politic’s subjection to disease and death, and upon a variety of ‘diagnoses’ of these ‘illnesses’ and suggestions for ‘cures’. The metaphor’s widespread use was indebted to its ability to be thus moulded in a variety of ways.
2
Political conceptions of natural bodies
The metaphor could also be used in discussions of nature rather than of politics: the body politic could be invoked as a way of describing the human body, or even natural bodies more generally. Du Bartas’s poem is entitled La Sepmaine, ou la creation du monde (‘The Week, or the Creation of the World’) and constitutes a description of the natural world structured by the seven-day Creation narrative recounted in Genesis. The body politic appears in the text when the poet explains that God separated the four elements (fire, air, water, and earth) into their respective cosmic realms because when the elements combine in cosmic bodies those bodies inevitably die. The poet employs as an example of this the human body, in which the excessive dominance of one element or humour (terms used interchangeably by Du Bartas) brings about its death, just as, according to the poet, a body politic can be destroyed by an excessive use of royal power. Du Bartas no doubt used the metaphor of the body politic because it enabled him to slip a polemical political argument into the midst of a very long poem about nature and God.6 However, the human body itself was also of great interest to Du Bartas. The creation of man is central to the Genesis narrative, and later in his poem, as he recounts that creation, Du Bartas discusses at length the powers of the human body (Du Bartas, 1935–40, vol. II: 394–402, Day VI, ll. 483–708). In the section referring to the body politic, Du Bartas is concerned with the issue of human death, which he depicts in detail. Four possible deaths are described, one for each humour which might become too dominant. In each case the poet emphasizes the suffering caused by that particular mode of death, evoking, for example, difficulty in breathing, fever, and unquenchable thirst. Each of the four passages concludes with an
Bodies Politic and Natural in 16th Century France
211
almost identical rhyming couplet stating that the suffering does not cease until man’s bones are enclosed in a ‘frozen’ tomb (Du Bartas, 1935–40, vol. II: 226–7, Day II, ll. 119–20; 127–8; 135–6; 143–4). Such a refrain is unique in the whole of Du Bartas’ work (and his two longest poems alone constitute approximately 20,000 lines).7 The body metaphor as formulated by Du Bartas thus functions as part of a baroque emphasis upon human death, a common theme in late sixteenth-century literature (Rousset (1954: 81–117), Braunrot (1973: 115–23)). Furthermore, nature more generally is a primary concern in the Sepmaine, and humoral conflict in the human body features as an example of elemental interaction, a subject which is discussed at great length. The emphasis placed upon elemental (and thus humoral) discord – and, more importantly, upon elemental discord imagined metaphorically as violent conflict – may be inspired by the Roman poet Lucretius’ depiction of ‘conflict’ between atoms: Lucretius’ De rerum natura, like the Sepmaine, is a long epic poem about the cosmos, and Du Bartas imitated passages from it (Fraisse, 1962; Kany-Turpin, 1991; Lamacz, 2002). The body metaphor was used in a similar way in a discussion of medicine also published in the 1570s: Alcmaeon holds that the equality of the faculties of the human body, like wetness, heat, dryness, cold, bitterness, sweetness, and some others, preserve and maintain health: and that, by contrast, monarchy, that is to say, predomination of one of these, creates illness: for this domination and principality brings about the corruption of the others, and is cause of illnesses (Plutarch/Amyot, 1572, f. 460v; translation and italics: KEB). This passage is from the De placitis philosophorum which (although now generally believed to be the work of Aëtius) in the sixteenth century was ascribed to Plutarch and appeared as part of his Moralia, a work very popular in the translation by Jacques Amyot (Aulotte, 1965: 256). Amyot chose to retain the metaphor of ‘monarchy’ to denote humoral dominance, and so, as in Du Bartas’ poem, the metaphorical equivalence between natural bodies and political ones appeared in a discussion of the former. The semantic exchange between the ‘natural’ and ‘political’ domains of the body metaphor appears to have been particularly strong in the sixteenth century. Either could constitute the source domain or target and, furthermore, interpretations of the metaphor in one domain could
212
Kathryn Banks
influence those in another: for example, Du Bartas’ application of it in his discussion of nature was ‘recycled’ by the author of the aforementioned Apologie pour la paix in his discussion of politics (Anon., 1585: 25). Adaptations of the metaphor could move very swiftly between the discussion of nature and that of politics, thus further increasing the possibilities for the metaphor’s ‘replication’, as well as the strength of semantic influence of both domains upon each other.
3 Other motivations of the body metaphor In his discussion of ‘Tacitism’ in the history of political thought, Peter Burke suggests that the body metaphor reflected a desire to establish ‘scientific’ laws for politics like those which existed in medicine (Burke, 1969: 167–8, and 1991: 482, 486). In other words, in the texts Burke studied, the metaphor’s success was indebted to a particular understanding of politics as a ‘science’. However, the metaphor was also used in discourses where theorizing was a less important impetus than polemic. Such discourses included ones intended to persuade through rhetorical strategies employed orally rather than through texts which could be re-read and subjected to careful scrutiny: Simon Vigor, perhaps the most famous preacher in Paris during the 1560s, used corporeal images to argue that the French king should not tolerate the ‘putrid infection of heresy’, which threatened the entire social order (Diefendorf, 1991: 153–8). While many of the thinkers whom Burke discusses were strongly inspired by the contemporary applicability of the ‘laws’ they were establishing, and while polemical arguments did gain strength from the ‘laws’ of medicine, current relevance was arguably a much stronger impetus for Vigor than any conception of politics as a ‘science’: his statements about the ‘body politic’ were designed to support arguments about what should be done in France. Thus, for some, including Vigor, any ‘laws’ concerning the body were more than usually motivated by the immediate concerns of the ‘scientist’ in question, whereas, for others, abstract knowledge was a more important incentive. Moreover, the term ‘body’ was an obvious one to use, whether or not one ‘dreamt of a science of politics’. In my discussion I have referred to ‘bodies politic’ and ‘natural bodies’ rather than to ‘bodies’ (understood to be natural ones) and ‘states’. The term état (state) was usually used to describe the condition of a body politic; according to Howell A. Lloyd, the idea of the state as a distinct and substantive entity was formulated in the 1570s by Jean Bodin – but Bodin himself used the term inconsistently, and subsequent sixteenth-century thinkers continued
Bodies Politic and Natural in 16th Century France
213
to refer to the corpus mysticum reipublicae rather than the état (Lloyd, 1983: 146–71). Furthermore, the similarity between the body politic and the body natural seems to have been considered an ontological one, that is, one regarded as real rather than heuristic or figurative.8 It is part of a wider equivalence perceived between the natural world and the human one, between nature and society; another example of this is the similarity between the king and the sun.9 The epistemological importance attributed to similarity in the sixteenth century varied from discourse to discourse (Maclean, 1998); nonetheless the similarity between the human body and the political one was convincing enough that royal doctors gave advice on how to ‘cure’ France, and other doctors composed treatises on this subject, as if their knowledge of the human body equipped them to diagnose the political body (Soll, 2002; Burke, 1991: 482, 486). By contrast, in more recent years, while the body metaphor appears for example in the form of heart-based metaphors in debates about the European Union (Musolff 2004a: 83–114), I know of no heart surgeon who believes that their profession qualifies them to comment upon the way in which the European Union should function. This ontological status of the body metaphor was, I would argue, an important force behind its massive replication in the sixteenth century. Any metaphor shapes perceptions of its target domain, whether or not the similarity implied is grounded in the way I have described. However, in the contemporary period one can at least attempt to resist this process by highlighting the metaphorical status of particular utterances. This is precisely what Susan Sontag does powerfully with reference to illness: ‘only in the most limited sense is any historical event or problem like an illness’; ‘Of course, one cannot think without metaphors. But that does not mean there aren’t some metaphors we might well abstain from or try to retire’ (Sontag, 1991: 85, 91). More generally, we tend to consider that ‘reality [ ... ] can be best described in simple, non-metaphorical terms’ and that ‘figurative or poetic assertions are distinct from true knowledge’ (Gibbs, 1994: 1). It is for this reason that highlighting the cognitive importance of metaphor can be considered to undermine an entire edifice of Western thought based on ‘reason’ (Lakoff and Johnson, 1980: 195–222). It is more difficult to locate an opposition between metaphorical language and logic in sixteenth-century thinking. The sixteenth century saw the humanist introduction of figures based on analogy into the discipline of logic (although this was resisted by more traditional scholastic thinkers). Furthermore, metaphors were logically grounded in similarities perceived to exist between, for example, the
214
Kathryn Banks
natural world and the societal one. As J. Rousset stated, ‘the old analogical cosmos [ ... ] logically grounded the validity of the metaphorical spirit depending upon the similitudes and correspondences between all orders of reality, from stones to man and from man to the stars’ (Rousset, 1976: 67; my translation). Thus, use of the body metaphor was supported not simply by a perceived equivalence between two domains but also by the notion that such equivalences were ‘real’. The strong perception of a similarity between natural bodies and political ones seems to have been so ‘self-evident’ that it did not need to be stated: discussions grounded in this underlying similarity did not always explicitly refer to the body politic as a ‘body’. References to a body per se signify much (although ‘body’ acquired more specific overtones in certain contexts, for example those of mortality in apocalyptic discussions of the general decline of the world). On the other hand, extending the metaphor, or using related metaphors of ‘limbs’ or ‘illness’, for example, could be used to make more specific – and polemical – arguments, such as locating the source of France’s problems in the ‘poison’ of ‘heresy’. Thus, while the strong apprehension of similarity between body natural and body politic was arguably important in encouraging the use of the body metaphor, the development of the metaphor was often motivated by the desire to depict France in more specific – perhaps controversial – ways.
4
Evolution of the body metaphor
While the late sixteenth century saw a wide variety of interpretations of the body metaphor, the depiction of a natural body was ubiquitous. By contrast, in seventeenth-century European thought, the state tended to be considered as a person as much as a body (Lloyd, 1983: 22), and natural metaphors were combined with mechanistic ones. In the genre of political thought at least, the political body in the seventeenth century was less exclusively a natural one than in the sixteenth. The new metaphor of a person expressed a new concept of the state; however, it would be misleading to assume that the changes in metaphorical thinking were secondary to changes in political thought. Indeed, the new version of the body metaphor can be seen to respond to interpretations (and interpretative problems) of the old one, firstly, the conflict between absolutist and constitutionalist analyses and, secondly, the emphasis upon disease and death. Lloyd (1983: 146–68) suggests that the new idea of the state was needed because the concept of a body composed of members of the political community could give rise to conflicting interpretations of
Bodies Politic and Natural in 16th Century France
215
the location of authority, a problem which became very acute in late sixteenth-century France. Indeed, actual uses of the body metaphor show that it served to locate sovereignty either solely in the ‘head’ or also throughout the ‘body’, and that these contrasting interpretations had become increasingly controversial and polarized rather than leading the discourse community towards any solution which might be shared. By contrast, in Hobbes’ seminal Leviathan, the concept of the state as artificial person locates sovereignty firmly in the sovereign chosen to represent that person: in order to avoid the warfare inherent in living in the condition of nature, the multitude instituted the artificial person of the state, and authorized the sovereign to represent that person (Hobbes, 1996: 111–15; Skinner, 2002: 177–208). Late sixteenth-century body metaphors suggest that the image of the natural body was also problematic insofar as it was increasingly interpreted as diseased and threatened by death. Especially in texts which did not appeal to a ‘doctor’ to employ a particular ‘medicine’, the body metaphor served to analyse the political situation but not to point to a solution. Where writers depicted bodies politic in general (rather than the French body in particular), the French wars did not appear as an exception to any general rule of harmony in bodies politic: instead, in Du Bartas’ case for example, it was a general rule that royal choices (in favour of the king’s own desires rather than the law) might destroy bodies politic.10 Furthermore, a pessimistic ‘baroque’ conception of the body natural – that is, one which emphasizes the omnipresence of death – could reinforce a conception of the body politic as ‘living’ a precarious and threatened existence. In short, the body metaphor suggested that political problems, such as royal violence, were ‘natural’. The anonymous author of the 1585 Apology is clearly uneasy about this. Having argued that nature teaches us to live in peace (pp. 23–35), he acknowledges that nature also contains violence, attempting to circumvent the apparent contradiction by observing that violence is an imperfection caused by the Fall of Man (pp. 75–83); arguably, though, the problem nonetheless weakens the argument. Thus the body metaphor creates a bleak image of bodies politic; the fatalistic and aggressive illness metaphors which Sontag ascribes uniquely to the modern period (Sontag, 1991: 77–87) are used, but in a context in which their power is less likely to be diminished by arguments against metaphor. The problem of disease can be solved by the concept of the state as an artificial person for whom mechanistic metaphors as well as organic ones are appropriate. When, in the opening lines of his Leviathan (1996: 9), Hobbes explains that the commonwealth is ‘an Artificiall Man’, he is
216
Kathryn Banks
quick to add that this artificial man is ‘of greater stature and strength than the Naturall’. Later he explains that ‘diseases’ can be avoided if this artificial man is properly created (1996: 221). The natural body is less central to the body metaphor, so that the difference between the (artificial) state and the (natural) body can be highlighted. In the face of civil warfare depicted as a natural disease of bodies politic, Hobbes’s artificial person is crafted so as to enjoy a greater longevity than natural bodies. Arguably, writers about politics were henceforth less likely to tar nature and politics with the same brush. By contrast, in late sixteenth-century France, natural bodies were central to the replication of the body metaphor; as we have seen, the metaphor also owed its success to religious discord, to conflict between constitutionalist and absolutist conceptions of monarchy, to the desire to establish ‘scientific’ laws for politics, and to a perceived ontological similarity between political bodies and natural ones.
Notes 1. Memes were posited by Richard Dawkins in his Selfish Gene (1976) and denote a second ‘replicator’ in addition to the gene; memes are elements of culture that compete with other memes to replicate themselves through imitation in human hosts. For critical discussion of the meme concept in conceptual history and discourse history, see Roslyn Frank in this volume. 2. In 1560, this argument was used by Jean Lange, the speaker for the Third Estate at the Estates of Orleans (Church, 1969: 98, n. 52), and also by Etienne Pasquier in his Pourparler du prince (1560: ff. 95r–96v). 3. Roberts describes the king as the ‘undisputed head’ of the body politic (2007: 149, 164). 4. See Banks (2008). 5. Anon. (1585: 25). For the attribution to Pasquier, see Thickett (1956: 13, 75 and 1979: 105–6). 6. It appears after almost 1000 lines of a poem approximately 6,500 words long. 7. Du Bartas (1935–40, vol. II: 226, note to ll. 119–20). 8. My use of ‘ontological’ in this context is not to be confused with that which refers to the metaphorical consideration of events, emotions or ideas as entities and substances (for example Lakoff and Johnson, 1980: 25–32): ‘ontological’ here refers to the status of the similarity rather than that of a source domain, which is more concrete than the target one. See Banks (2008). 9. For example, one commonplace collected by Erasmus reads as follows: ‘just as God in the heavens created the sun as the most beautiful and most delightful image of himself, so in the republic he created the king to represent him through his wisdom, his justice and his liberality’ (Lycosthenes, 1614: 97, my translation). 10. Sontag is thus mistaken to suggest that in the early modern period metaphors of illness always gave optimistic ‘prognoses’ and presented ‘disease’ as
Bodies Politic and Natural in 16th Century France
217
manageable, thus reinforcing a general rule of harmony from which illness was a deviation (Sontag, 1991: 77–80).
References Anon. (1575). La Resolution claire et facile sur la question tant de fois faite de la prise des armes par les inferieurs (Basle: les héritiers de Jehan Oporin). Anon. (1585). Apologie de la Paix. Representant tant les profficts et commodités que la Paix nous produict, que les malheurs, confusions, et desordres qui naissent durant la Guerre (Paris: Jean Richer). Anon. ([1589]). Origine de la maladie de la France, avec les remedes propres à la guarison d’icelle, avec une exhortation a l’entretenement de la guerre (Paris: Jacques Varangues). Aulotte, R. (1965). Amyot et Plutarque: La Tradition des Moralia au XVIe siècle (Geneva: Droz). Banks, K. (2008). Cosmos and Image in the Renaissance: French Love Lyric and Natural-Philosophical Poetry (Oxford: Legenda). Blackmore, S. (1999). The Meme Machine (Oxford: Oxford University Press). Braunrot, B. (1973). L’Imagination poétique chez Du Bartas: éléments de sensibilité baroque dans la ‘Création du monde’ (Chapel Hill: University of North Carolina Press). Burke, P. (1969). ‘Tacitism’. In T. A. Dorey (ed.), Tacitus (pp. 149–71) (London: Routledge & Kegan Paul). Burke,P. (1991). ‘Tacitism, scepticism, and reason of state’. In J. H. Burns (ed.), The Cambridge History of Political Thought 1450–1700. (pp. 479–98) (Cambridge: Cambridge University Press). Church, W. F. (1969). Constitutional Thought in Sixteenth-Century France (New York: Octagon Books). Clark, C. E. (1970). ‘Montaigne and the imagery of political discourse in sixteenthcentury France’, French Studies XXIV, 337–55. Diefendorf, B. B. (1991) Beneath the Cross: Catholics and Huguenots in SixteenthCentury Paris (New York and Oxford: Oxford University Press). Du Bartas, Guillaume de Saluste (1935–40). La Sepmaine. In The Works of Guillaume De Salluste Sieur Du Bartas, ed. by U. Tigner Holmes, J. Coriden Lyons, R. White Linker, et al. (vol. II, pp. 193–440). (Chapel Hill: University of North Carolina Press). Fraisse, S. (1962). Une Conquête du rationalisme: l’influence de Lucrèce en France au seizième siècle (Paris: A. G. Nizet) Frank, R. M. (in this volume). ‘Shifting identities: metaphors of discourse evolution’. Gibbs, R. W., Jr. (1994). The Poetics of Mind: Figurative Thought, Language, and Understanding (Cambridge: Cambridge University Press). Hale, D. G. (1971). The Body Politic: A Political Metaphor in Renaissance English Literature (Paris: Mouton). Hobbes, T. (1996), Leviathan, ed. R. Tuck (Cambridge: Cambridge University Press). Jouanna, A. (1989). Le Devoir de révolte: La Noblesse française et la gestation de l’État moderne, 1559–1661 (Paris: Fayard).
218 Kathryn Banks Kantorowicz, E. H. (1957). The King’s Two Bodies: A Study in Mediaeval Political Theology (Princeton, NJ: Princeton University Press). Kany-Turpin, J. (1991). ‘Une Réinvention de Lucrèce par Guillaume du Bartas’. In Y. Bellenger (ed.), La Littérature et ses avatars: discrédits, déformations et réhabilitations dans l’histoire de la literature. (pp. 31–52) (Paris: Klincksieck). Lakoff, G., and Johnson, M. (1980). Metaphors We Live By (Chicago and London: University of Chicago Press). Lamacz, S. (2002). ‘La Construction du savoir et la réécriture du De rerum natura dans la Sepmaine de Du Bartas’. Bibliothèque d’humanisme et renaissance LXIV, 617–38 Lloyd, H. A. (1983). The State, France, and the Sixteenth Century (London: George Allen & Unwin). Lycosthenes, C. (1614). Parabolarum sive similitudinum, quae ex Aristotele, Plutarcho, Plinio ac Seneca, gravissimis authoribus, olim ab Erasmo Roterodamo collectae, ... loci communes (Lyon: Joannem Gazeau). Maclean, I. (1998). ‘Foucault’s Renaissance episteme reassessed: an Aristotelian counterblast’, Journal of the History of Ideas, 59, 149–66. Musolff, A. (2004a). Metaphor and Political Discourse. Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Musolff, A. (2004b). ‘Metaphor and conceptual evolution’. metaphorik.de 7, 55–75. http://www.metaphorik.de/07/ (accessed: 2 August 2007). O’Brien, J. (forthcoming). ‘Intestinal disorders’. In E. Vinestock and D. Foster (eds.) Writers in Conflict in Sixteenth-Century France: Essays in Honour of Malcolm Quainton (Durham: University of Durham). Parel, A. J. (1992). The Machiavellian Cosmos (New Haven and London: Yale University Press) . Pasquier, E. (1560). ‘Pourparler du prince’, in E. Pasquier, Les Recherches de la France, Livre Premier, plus, un Pourparler du prince (Paris: Vincent Sertenas). Plutarch/Amyot (1572). Les Œuvres morales & meslees, trans. by Jacques Amyot (Paris: Michel de Vascosan). Roberts, P. (2007). ‘The kingdom’s two bodies? Corporeal rhetoric and royal authority during the Religious Wars’, French History, 21, 147–64 Rousset, J. (1954). La Littérature de l’âge baroque en France: Circé et le paon (Paris: J. Corti). Rousset, J. (1976). L’Intérieur et l’extérieur: essai sur la poésie et sur le théâtre au XVIe siècle (Paris: Librairie José Corti). Seyssel, Claude de (1960). La Monarchie de France, ed. by J. Poujol (Paris: Librairie d’Argences). Skinner, Q. (2002).Visions of Politics, vol. 3: Hobbes and Civil Science (Cambridge: Cambridge University Press). Soll, J. (2002). ‘Healing the body politic: French royal doctors, history, and the birth of a nation 1560–1634’, Renaissance Quarterly, 55, 1259–86. Sontag, S. (1991). Illness as Metaphor and Aids and its Metaphors (London: Penguin. Thickett, D. (1956). Bibliographie des Œuvres d’Estienne Pasquier (Geneva: Droz). Thickett, D. (1979). Estienne Pasquier (1529–1615): The Versatile Barrister of 16thCentury France (London: Regency Press). Zavadil, J. (in this volume). ‘Bodies politic and bodies cosmic: the Roman Stoic theory of the “two cities’’ ’.
14 Bodies Politic and Bodies Cosmic: The Roman Stoic Theory of the ‘Two Cities’ Jeffery Zavadil
Introduction In this chapter I examine a development in Roman political thought, often called the theory of the ‘two cities’, that involved a change in how ‘bodies politic’ were conceived and that has had extensive effects on Western political thought. For most of classical antiquity, the locus of political loyalty was the parochial city-state, but during the Roman imperial period Stoic and other philosophers more strongly emphasized the cosmos as the proper location of ethical identity: the universe was a great heavenly city, governed by a rational natural law (just as an earthly city is governed by a man-made law), whose population encompassed all humanity. Although this conceptual shift also appeared in other schools of thought (including Cynic and Christian), Stoic philosophers most clearly expressed it. I will focus on the writings of Cicero, Seneca, Epictetus, and Marcus Aurelius.1 In these Stoic texts the concept of the ‘two cities’ was definitively articulated with organic metaphors: by taking advantage of the common term ‘body’ of the metaphors THE CITY IS A BODY and THE COSMOS IS A BODY, the metaphor THE COSMOS IS A CITY was created. The ‘body politic’ was thereby turned into the ‘body cosmic’, and the target domain of the body politic and other organic political metaphors shifted from the single city to the world as a whole. Although body–cosmos metaphors had long been an element of Platonic and Stoic thought, the existence and permanence of the Roman Empire called for an expanded sense of political identity, which this organic ‘two cities’ metaphor made possible. The ‘two cities’ has had significant political consequences historically, and even today influences debates over cosmopolitanism. 219
220
1
Jeffery Zavadil
Background ancient political concepts
Today political identities largely remain tied to the nation-state, but the focus of political loyalty for most of antiquity was the city-state or polis (in Greek). Each polis was a self-governing body that made its own laws and independently conducted diplomacy, trade, and warfare (Thucydides, 1982). In the smaller, more communal polis, public and private were not as separate as today, and in philosophy there was little distinction between political and ethical theory. Democratic Athens and other poleis limited citizenship to free males but also demanded more from citizens, expecting them to participate in the assembly, serve on juries, follow public religious rites, and take up arms in defence of the city. The age of the polis came to an end in the late 300s BC with Phillip of Macedon’s unification of Greece and the establishment of Hellenic empires in the wake of Alexander’s conquests. Formerly independent city-states were then incorporated into entities that were cosmopolitan (encompassing a cosmos of poleis). But the independent, self-governing polis lived on as an ideal for a long time. In Italy, the Roman republic was still an independent city-state by the time it gained dominance over the Italian peninsula in 286 BC and began its Mediterranean expansion. As Rome grew, citizenship was gradually extended to conquered peoples, either as a reward for military support or for political reasons, as when in 212 AD Caracalla gave imperial citizenship to all free males so they could be subject to the 5% death tax. Although much Roman culture was Hellenic, Rome grew too large to maintain the classic Greek view of citizenship. Citizenship could no longer mean active participation in governing, as most ‘citizens’ lived too far outside of Rome to attend its assemblies, which were eventually eliminated under the emperors anyway (Dahl, 1998: 13–14). Rather, Roman citizenship became a legal status that conveyed certain privileges, such as the right to make contracts or sue, and immunity from torture and crucifixion. Nor did expansion mean that the Roman imperium became a nationstate in the modern sense, for there was always a distinction between Rome, the city and its environs, and the territories that it governed: Rome proper was a political community that held imperium or dominion over its provinces and territories. Furthermore, there was a metonymy at work: ‘Rome’ could refer to the city, to the empire that the city controlled, or to both, as the city could be thought of as standing in for the whole of the territory that it ruled. Thus ancient Romans, to a degree, always saw Rome as a city-state. However, a recognition of the
Roman Stoic Theory of the ‘Two Cities’ 221
Roman Empire as a cosmopolitan entity also gradually emerged – the establishment of Diocletian’s tetrarchy in 284 AD and Constantine’s removal of the capital to Constantinople in 324 AD both suggest a weakening of the identification of Rome the city with Rome the empire. The point is this: for much of classical antiquity, what ancient thinkers had in mind as their target domain when they used organic political metaphors was a city. A permanent empire that so clearly exceeded the scope of the classical city-state therefore called for a new target domain for organic political metaphors.
2 Greek bodies, politic and cosmic The metaphor of the body politic is a very ancient one and was common in both classical Greece and Rome. In Greece, fragmentary and undeveloped examples of bodies politic survive from Pre-Socratic philosophers and poets.2 Early on, the body was simply used as a conceptual pattern for understanding highly complicated and abstract political relationships. Just as a body consists of complex relationships between interdependent limbs and organs, so with a city: the different classes and/or professions in a city’s division of labour worked together to give the body politic life. This was illustrated by Aesop’s sixth century BC ‘Fable of the Belly and the Members’, perhaps the earliest reference to the body politic in Western thought: Back when all the parts of the human body did not function in unison as is the case today, each member of the body had its own opinion and was able to speak. The various members were offended that everything won by their hard work and diligent efforts was delivered to the stomach while he simply sat there in their midst, fully at ease and just enjoying the delights that were brought to him. Finally, the members of the body revolted: the hands refused to bring food to the mouth, the mouth refused to take in any food, and the teeth refused to chew anything. In their angry effort to subdue the stomach with hunger, the various parts of the body and the whole body itself completely wasted away. As a result, they realized that the work done by the stomach was no small matter, and that the food he consumed was no more than what he gave back to all the parts of the body in the form of blood which allows us to flourish and thrive, since the stomach enriches the blood with digested food and then distributes it equally throughout the veins (Aesop, 2002: fable 66).
222
Jeffery Zavadil
Note how the different social classes are conceived of as different parts of the body, the working class as the members and the leisure class as the stomach, and how the story advises against revolution. The metaphor here is straightforwardly anatomical, using human anatomy to depict political relationships, in this case to provide an ideological justification for class privilege. Both Plato and Aristotle, ancient Greece’s greatest influences on political thought, frequently used a large variety of body-related metaphors in their political philosophies, drawing on the source domains of anatomy, medicine, physicians, diet, exercise, and biology. Both philosophers thought that the role of an individual was defined in relation to the whole city, just as the functions of a limb or organ was defined in relation to the whole body. For Aristotle, wholes of any kind were ontologically superior to individual parts: ‘[T]he state is by nature clearly prior to the family and to the individual, since the whole is of necessity prior to the part; for example, if the whole body be destroyed, there will be no foot or hand, except homonymously, as we might speak of a stone hand; for when destroyed the hand will be no better than that.’ (Aristotle, 1984: 1253a 19–27).3 For Aristotle, parts were defined by their function, while a whole was a self-sufficient entity. A city was like a living body in that it achieved self-sufficiency by being composed of differentiated parts, citizens, that performed different functions and that were not self-sufficing outside it: someone subsisting outside a polis would have to be either a beast or god, and was human only homonymously (Aristotle, 1984: 1253a 29–30). This functionalism would persist into Roman times. While these early examples of organic metaphors were political, a cosmological body metaphor also appears in Plato’s Timaeus, where body, soul, city, and universe all exhibit a similar hierarchic pattern (Plato, 1953: 30c–34b, 44d–45b).4 The human body, Plato thought, looks as it does only because it is a copy of the cosmos, which is envisaged as a perfect sphere governed by a rational intelligence. The human torso, also roughly spherical, takes its basic shape from the universe, except that the rational head has been pinched off at the neck to form another sphere attached to and governing it, and limbs have been added for locomotion and the manipulation of objects (Plato, 1953, 44d–45b). The intermediate level of the city, too, was organically arranged and should be ruled by its most rational class, Plato’s famous philosopher-kings. Plato understood himself as describing the small parts of the world as copies of the larger cosmic pattern, but modern science suggests that what he was doing was anthropomorphizing the cosmos by projecting
Roman Stoic Theory of the ‘Two Cities’ 223
the human body onto it. From Plato’s perspective, however, organic political metaphors were a subset of a larger array of cosmologicalorganic metaphors ranging up and down the entire ladder of existence. Hence the metaphor THE BODY IS A COSMOS was asserted early on, and also the metaphor THE CITY IS A COSMOS, for both body and city were patterned after the universe and so were similar to it. However, Plato did not think the reverse, that THE COSMOS IS A CITY – it was not that the universe took the form of a city, but that the city was a copy of the universe.5 Greek and Roman Stoic philosophers also commonly thought of the universe as an organized body ruled by a rational mind, an organic understanding that emerged from Stoic physics: the Stoics believed the rational principle of the universe to be a physical substance, either the element fire, or alternatively pneuma, ‘breath’, which combined fire and air. Thus the divine breath of the living universe was its rational vital essence. From what sources we have, early Stoic philosophers did assert the metaphor THE COSMOS IS A CITY, but their influence did not become widespread until the Roman imperial period (Pangle, 1998: 240–2).6
3
The Roman body politic
The Romans, following the Greeks, used the metaphor of the body politic to describe their city: according to Seneca, Caligula once said that he wished the Roman people had a single neck so he could strangle them all at once (Seneca, 1995: III.19.2). This synecdoche, in which a single body represented the many bodies that made up society, is just one example of the many bodily metaphors that appear in Roman political texts, most of which treated Rome, the city itself, as the relevant target domain of the metaphor. One example from the early Roman republican period involved Aesop’s ‘Fable of the Belly’ (Livy, 1967: II.xxxii).7 In 494 BC, a Roman army, consisting mainly of plebeian enlistees, successfully defended Rome from multiple military threats. The patrician Senate thoughtlessly refused to consider debt relief in return for their service, so in response the army encamped itself on the Sacred Mount three miles away and crouched ominously over Rome. This threw the populace into a panic, so the Senate sent an envoy, Menenius Agrippa, to negotiate. He told the plebeians a near-verbatim version of Aesop’s ‘Fable of the Belly’. Menenius likened the different classes of Rome to different parts of the body, saying that the plebs were the hands and arms and the patrician Senate was the stomach; if the arms did not feed the belly then Rome as a whole would perish. This story was so persuasive, Livy recounts, that the plebs ended their revolt and rejoined
224 Jeffery Zavadil
the city. The civic body of interest here, of course, was the city of Rome proper. The organic political metaphors of the late republican orator and statesman Marcus Tullius Cicero (106–43 BC) also mainly had the city-state as their target. While Cicero employed a variety of political metaphors, organic metaphors helped to constitute important parts of his political outlook.8 Cicero was active during the period of civil strife that led to the downfall of the republic. Conflict was between two class-based political factions, optimates with a political base in the upper classes and populares, elites that included politicians such as Julius Caesar, who found support among the plebeians through land distribution, debt relief, and so on. In the following example, Cicero criticizes the political tactics of his time, which included property confiscation and proscription (Cicero was himself killed a few months after writing this): Suppose that each limb were disposed to think that it would be able to grow strong by taking over to itself its neighbour’s strength; necessarily the whole body would weaken and die. In the same way, if each one of us were to snatch for himself the advantages that other men have and take what he could for his own profit, then necessarily fellowship and community among men would be overthrown (Cicero, 1991: III. 21–2). And so no one should violate the purpose for which the republic existed, which was to protect property, both private and public (Cicero, 1991: I.20–2, I.31; Wood, 1983: 739–56). Cicero also criticized tyranny with body metaphors. Tyrants were wild animals, threats to the community that must be hunted down and killed: ‘Although he [the tyrant] has the outward appearance of a man, he outdoes the wildest beasts in the utter savagery of his behaviour.’ (Cicero, 1998: 2.48). He then described tyrants as festering sores on the body politic which must be cut out: ‘For just as some limbs are amputated, if they begin to lose their blood and their life, as it were, and are harming the other parts of the body, similarly if the wildness and monstrousness of a beast appears in human form, it must be removed from the common humanity, so to speak, of the body.’ (Cicero, 1991: III.32). Cicero thus thought the assassination of would-be tyrants such as Caesar was justified (Atkins, 2000: 508). Again, the city is the target domain of the body metaphor. Cicero, however, also explicitly evoked Plato’s cosmology from the Timaeus to frame his view of natural law. Cicero understood the universe
Roman Stoic Theory of the ‘Two Cities’ 225
to be ruled rationally by a single mind, just as a body is governed by a rational mind. This mind gave the universe a rational, natural law, governing it the way that a head governs a body or that man-made law governs the republic. This suggests that the ‘natural law’ tradition has a body metaphor at its root: the universe is ruled by a rational law as the body is ruled by a rational intelligence (Cicero, 1998: 1.23, 2.16).9 A more direct mixing of cosmos and city, mediated by organic metaphors, appears in the writings of the Stoic philosopher Lucius Annaeus Seneca (4BC–65 AD), who was tutor and advisor to the emperor Nero (who eventually ordered Seneca to commit suicide). Many of Seneca’s ethical metaphors were medical: throughout his treatise On Anger, for example, he discusses the emotional vice of anger as an illness of the soul to be treated with both preventative and therapeutic philosophical medicine. Seneca’s cosmology was typically Stoic, so the idea that humans inhabit a rational universe and should conform to its laws permeates his social thought (Cooper and Procopé, 1995: xviii). Whereas Cicero had argued that mixed government based on the rule of law was the form of government that best coincided with nature (Cicero, 1998: 1.56–69), Seneca, at least sometimes, favoured monarchy: Compare the way in which the body is entirely at the service of the mind. It may be ever so much larger and more impressive. The mind may remain hidden and tiny, its very location uncertain. Yet hands, feet, and eyes do its business. The skin that we see protects it. At its command, we lie still. Or else we run relentlessly about, when it has given the order ... In the same way, the vast multitude of men surrounds one man as though he were its mind, ruled by his spirit, guided by his reason; it would crush and shatter itself by its own strength, without the support of his discernment (Seneca, 1995b: I.3.5–4.1). While the emperor, a single individual, might appear tiny and insignificant in comparison to the populace of Rome, metaphorically he was the most important element in the imperium, the mind of its body. Just as the mind could overrule the impulses and pains of the body to achieve great things, so the princeps, through his reason, discernment, and spirit, could induce the people to overcome their impulsive and riotous tendencies and unify their action. We see here a justification of ‘top-down’ hierarchy by appeal to interdependence: the people need their emperor lest they lose direction and the ability to act. Thus Seneca mapped the body onto the polity in a way that provided ideological justification for the principate.
226
Jeffery Zavadil
This dependence did work both ways, however. Seneca stressed that the emperor was as dependent on Rome as it was on him, and therefore he should not harm the political body any more than he would his own person. This assertion of an identity of interests between ruler and ruled helped justify monarchy by imbuing it with an element of mercy: ‘if, as it has just been shown, your commonwealth is your body and you are its mind, you can see, I think, how necessary mercy is. You are sparing yourself, when you appear to spare another. So even culpable citizens should be spared in the same way as unsound limbs. If there is ever a need for blood to be let, you should stay the knife, lest it cut beyond what is needed.’ (Seneca, 1995b: I.5.1, I.13.4). Seneca therefore advocated clemency: emperors should not order harm to their subjects because that is like a head ordering harm to its own body. History shows how poor a student Nero was in learning this lesson.
4
A tale of two cities
During the first and second centuries AD the Stoic metaphor THE COSMOS IS A CITY was becoming more prominent; since both cosmos and city were bodies, philosophers started to more strongly assert and develop the idea that the cosmos was a city, rather than that cities were patterned after an organic cosmos. Furthermore, the increasing popularity of Stoicism made this outlook available outside the confined circles of philosophers and their students. This example from Seneca, a reminder to his readers to avoid causing harm, illustrates both a body politic and the Stoic metaphor of the ‘two cities’: It is a sacrilege to harm your country; it is therefore a sacrilege to harm a citizen, too (for he is part of your country – and its parts will be sacred, if the whole commands veneration); and therefore it will be a sacrilege to harm even a human being, since he is a citizen of that greater city of yours. What if the hands wished to harm the feet or the eyes the hands? The members of the body all accord with each other, since it is in the interest of the whole that the parts be preserved (Seneca, 1995b: II.31.7; emphasis added). Here we see the theme of complex interdependence again, but in this passage Seneca extended the body metaphor beyond the city by appealing to the ‘greater city’ of the world. He concluded that men as men ‘were born for fellowship’, not just Roman men.10 For the Stoics, the greater universe was coming to be understood not just as a body but also as a body poli-
Roman Stoic Theory of the ‘Two Cities’ 227
tic, one that encompassed all humanity, not only the citizens of a single city, and thus allowed for the emergence of a universal moral identity. Other Stoic thinkers used organic metaphors in similar ways. The slaveturned-philosopher Epictetus of Heirapolis (c.50 BC–c.30 AD), who had an immense and lasting influence on Roman ethics that crossed class boundaries, urged his students to remember that their duties might sometimes be unpleasant, but served a greater good. An individual, as part of a social body, was obligated to accept his appointed lot and to fulfil his duties without complaint: ‘Only when we are viewed as isolated individuals I will allow that it is natural for the foot (for instance) to remain clean. But if you take it as a foot, and not as a mere isolated thing, it will be fit that it should walk in the dirt, and tread upon thorns; and sometimes that it should even be cut off, for the good of the whole body; otherwise it is no longer a foot.’ (Epictetus, 1944: II.5). Likewise, ‘if you consider yourself as a man, and as a part of a whole, it will be fit, on the view of that whole, that you should at one time be sick; at another, take a voyage, and be exposed to danger; sometimes be in want; and possibly die before your time.’ (Epictetus, 1944: II.5). Thus one should accept life’s difficulties, for a member of a whole is defined by its function, whether the part be a foot or a man. Removed from the living body of the cosmos, the part ceases to be a part and becomes a dead and inanimate object; recall Aristotle’s comment that a foot or hand amputated from the body is a foot or hand only homonymously and not actually. Epictetus concludes with an explicit statement of the two cities: ‘For what is a man? A part of a commonwealth; first and chiefly that which includes gods and men; and next, of that which you immediately belong, which is a miniature of the universal city.’ (Epictetus, 1944: II.5; emphasis added). Thus moral duties based on functionalism were becoming universal as well a civic: in this passage moral identification is made ‘first and chiefly’ with the body cosmic, and only secondarily with a parochial body politic. There are many contemporaneous examples of this shift, but I will only give one other. The journal kept by the emperor and Stoic philosopher Marcus Aurelius Antoninus Augustus (121–180 AD), today called the Meditations, contains many kinds of organic political metaphor, including the two cities: ‘the world is a kind of political community’ (Aurelius, 2002: IV.3; Gill, 2003: 42–3). But for Marcus, as for earlier Stoics, this world-state also corresponded to a world-body: Constantly regard the universe as one living being, having one substance and one soul; and observe how all things have reference to
228 Jeffery Zavadil
one perception, the perception of this one living being; and how all things act with one movement; and how all things are the cooperating causes of all things which exist; observe too the continuous spinning of the thread and the contexture of the web (Aurelius, 2002: IV.40). The interdependence implied by the metaphors of living being, fabrics, and webs was reflected in the social nature of humanity, and Marcus concluded that people were intended by nature to cooperate and not abuse each other: ‘For we are made for cooperation, like feet, like hands, like eyelids, like the rows of the upper and lower teeth.’ (Aurelius, 2002: I.1, VII.13). Since the rational individual was part of a greater universal body, to be virtuous was to be a healthy, functioning organ of that body, while to be vicious was to be a sickly, dysfunctional one: ‘The human soul does violence to itself [ ... ] when it becomes an abscess and, as it were, a tumour on the universe, so far as it can.’ (Aurelius, 2002: 1.16). Through these metaphor-based arguments, Marcus reminded himself that even as emperor of Rome he was the mere toe of a universe that had its own plans for him. In addition to having responsibilities as the leader of an earthly political community, he had responsibilities as a human being: ‘But my nature is rational and social; and my city and country, so far as I am Antoninus, is Rome, but so far as I am human, it is the world. The things then which are useful to these cities are alone useful to me.’ (Aurelius, 2002: VI. 44). With this statement Marcus exhibited the cosmopolitan shift in political loyalties prompted by the metaphor of the ‘two cities’, and what an amazing statement it was: where one would expect an emperor to display unqualified loyalty to the imperial centre, here he also expressed, in the same breath, his commitment to the larger cosmos of which it was a part.
5
Conclusion
During the first two centuries of the Roman imperial period a blurring of terms was underway that extended the scope of ethical and political obligation: no longer was the polis the definitive whole of which individuals were parts, but the cosmopolis was becoming that definitive whole. The fact that this conceptual shift was achieved through organic metaphors is a very important point. The Stoic thinkers developed the metaphor THE COSMOS IS A CITY from the pre-existing conceptual metaphors THE CITY IS A BODY and THE COSMOS IS A BODY by exploiting the entailment that both were bodies to then infer that cities and the cosmos must also have a similar political structure. Whereas this similarity
Roman Stoic Theory of the ‘Two Cities’ 229
had previously implied that the constitution of the city ought to accord with the structure of nature, for the Stoics it meant that the world itself was structured as a great city. One could thus posit a universal law binding all, and one could feel a cosmopolitan ethical identity with other humans regardless of parochial political identity. Ancient cosmopolitan thought, therefore, was centred on an organic political metaphor, and may not have been possible without it. There were more lasting historical effects. The ‘two cities’ metaphor altered ancient conceptions of good citizenship, which were based on seeking a glorious remembrance from posterity for great civic deeds, in the form of ancestor worship or public memorials like columns, arches, and statues. The Stoics understood that this was ultimately meaningless: taking a long view of time, they knew that even those who memorialize you will themselves die and eventually you will be completely forgotten: ‘Near is your forgetfulness of all things; and near the forgetfulness of you by all’ (Aurelius, 2002: VII.19, 21). Civic glory was nothing to someone with such long time horizons, but without it, some alternative source of moral motivation was needed. Marcus Aurelius perhaps gave the best example of the Stoic response: indifferent to imperial fame, he was instead moved by Stoic philosophy to conform to the universal natural law – that is, to bring his mind into accord with the rational, universal mind (Aurelius, 2002: VII.19, 21; White, 2003: 124, 128, 152; Schofield, 1991: 242–6). This had the effect of valorizing philosophic virtues over political ones: rather than live an active, glory-seeking public life in the city of one’s birth, it was better to live ethically as a citizen of the world and leave the ruling to emperors and their minions. Thus there is a shift towards a political (although not ethical) passivity. Among Christian writers in the early Middle Ages such as Boethius, this became a full-blown theodicy that gave the coup de grâce to the active civic life of antiquity, elevating in its place the contemplative life of the medieval monk in his cell (Boethius, 1969: I.VI, IV.VI). The ‘two cities’ metaphor had an even longer-term influence than this. It is detectable in other patristic writers such as Augustine, who in the City of God used the source domain of the city to describe the universal church and/or God’s heavenly realm. Furthermore, in City of God we see an early statement of the Great Chain of Being, the idea that all creation, from large to small, followed an organic pattern repeated at each level – in the person, family, city, and world (Augustine, 1998: XIX.13, 938–40; XIX.25, 961). The greater world, the macrocosm or major mundus, was a model that each sub-part copied and so became a little world, the microcosm or minor mundus, an idea with clear Platonic
230 Jeffery Zavadil
roots. The Great Chain of Being became a fundamental part of the feudal worldview, and propped up its system of ranks and orders by depicting them as divinely ordained: a serf was exactly where God wanted him to be, as was a knight, baron, or king. Even today the ‘two cities’ has influenced liberal ideas of equality and human rights that depend on conceiving of all humanity as citizens of the world. This is sometimes explicitly acknowledged in contemporary debates in cosmopolitan political theory: Martha Nussbaum, for example, still evokes Stoic ideas about world citizenship in her arguments for cosmopolitan education, including the idea of a universal body politic (Nussbaum, 2002: 6–11). In sum, the history of the theory of the ‘two cities’ reveals that it was articulated in organic terms: like the ancient city, the world community is a body politic in which all the people of the earth share a common interdependence and a common nature. This specific incident of conceptual transformation illustrates at least three things about metaphors within discourses. First, it shows how changing political circumstances can prompt new combinations of metaphoric patterns, as Stoic philosophers, faced with a narrow body politic metaphor that no longer fit their reality, strove to create an ethical system for imperial Rome. Second, it shows how a change in a metaphor can have real-world implications for how people conceive of themselves in relation to their social and moral universe. Lastly, this example demonstrates how a relatively simple alteration of a metaphor can influence political and moral thought for centuries and even millennia, underscoring the importance of historically informed discussion about the metaphors within current discourse.
Notes 1. I will focus on the Roman Stoa rather than the ancient and middle Greek Stoa because: (1) early Stoic texts have been lost; (2) Stoicism became the most significant philosophical school in Rome; and (3) it was the Roman version of Stoicism that was influential in later Western political thought. See (Colish, 1985: 4, 12; Pangle, 1998: 240; Inwood, 2003: 1; Sedley, 2003: 30–1). For a summary of Stoicism, see (Colish, 1985: 9–11). 2. See the writings of Hesiod, Theognis, Alcmaeon, Heraclitus, Protagoras, Prodicus, Antiphon, and Xenophanes (Gagarain and Woodruff, 1995: 4, 38–9; Barnes, 2001: 31, 37, 62, 178, 211–20). 3. All citations of Aristotle refer to Bekker numbers rather than page numbers. 4. All citations of Plato refer to Stephanus numbers rather than page numbers. 5. In his Republic, Plato’s most influential political work, the idea that the cosmos is a body is largely absent – it is only alluded to elliptically at 592b (Plato, 2000).
Roman Stoic Theory of the ‘Two Cities’ 231 6. One of the first philosophers to think of the universe as a political community was Diogenes the Cynic, who declared himself a citizen of the world, and who is thought to have had a strong influence on Zeno of Citium, the founder of Stoicism. 7. Shakespeare repeats the tale in Coriolanus, Act 1, Scene 1; see also Act 3, Scene 1. See also Hale (1973). 8. Cicero was not a Stoic, strictly speaking. His positions were closest to the sceptical Platonism of the New Academy, but he was eclectic and incorporated much Stoic philosophy into his thought (Atkins, 2000: 484; Colish, 1985: 4, 12). 9. For Roman natural law theory, see (Colish, 1985: 31–6, 97–103). 10. The sexist element of these early body metaphors is typical of the age, and implicit sexist biases permeate much Stoic thought, of course. One the other hand, Stoics were sometimes admirably egalitarian in their theoretical views on gender. See Schofield (1991: 43–6).
References Aesop. (2002). Aesop’s Fables, trans. By L. Gibbs (Oxford: Oxford University Press). Aristotle. (1984). Politics. In J. Barnes (ed.), The Complete Works of Aristotle (pp. 1986–2129) (Princeton: Princeton University Press). Atkins, E.M. (2000). ‘Cicero’. In C. Rowe and M. Schofield (eds.), The Cambridge History of Greek and Roman Political Thought (pp. 477–516) (Cambridge: Cambridge University Press). Augustine. (1998). The City of God Against the Pagans, ed. by R.W. Dyson (Cambridge: Cambridge University Press). Aurelius, M. (2002). Meditations, trans. by G. Hays (New York: Modern Library). Barnes, J. (ed.), (2001). Early Greek Philosophy (London: Penguin). Boethius. (1969). The Consolation of Philosophy, trans. by V. E. Watts (New York: Penguin). Cicero. (1991). On Duties, ed. by M. Griffin and E. M. Atkins (Cambridge: Cambridge University Press). Cicero. (1998). ‘On the Republic’. In N. Rudd (trans.), Cicero: The Republic, The Laws (pp.1–94) (Oxford: Oxford University Press). Colish, M.L. (1985). The Stoic Tradition From Antiquity to the Early Middle Ages (Leiden: E. J. Brill). Cooper, J. M. and J.F. Procopé (eds.) (1995). Seneca: Moral and Political Essays (Cambridge: Cambridge University Press). Dahl, R. (1998). On Democracy (New Haven: Yale University Press). Epictetus (1944). ‘Discourses’. In T. W. Higginson (trans.), Discourses and Enchiridion, (pp.1–330) (Roslyn, NY: Walter J. Black). Gagarain, M. and P. Woodruff (eds.) (1995). Early Greek Political Thought from Homer to the Sophists, (Cambridge: Cambridge University Press). Gill, C. (2003). ‘The School in the Roman imperial period’. In B. Inwood (ed.), The Cambridge Companion to the Stoics (pp. 33–58) (Cambridge: Cambridge University Press).
232 Jeffery Zavadil Hale, D. G. (1973). ‘Analogy of the body politic’. In P.P. Wiener (ed.), Dictionary of the History of Ideas (pp. 67–70) (New York, Scribner). Inwood, B. (2003). ‘Introduction: Stoicism, an intellectual odyssey’. In B. Inwood (ed.), The Cambridge Companion to the Stoics (pp.1–6) (Cambridge: Cambridge University Press). Livy. (1967). Livy in Fourteen Volumes, trans. by B.O. Foster (Cambridge, MA: Harvard University Press). Nussbaum, M. (2002). ‘Patriotism and cosmopolitanism’. In M. Nussbaum (ed.), For Love of Country? (pp. 3–17) (Boston: Beacon Press). Pangle, T. (1998). ‘Socratic cosmopolitanism: Cicero’s critique and transformation of the Stoic ideal’, Canadian Journal of Political Science, 31.2, 235–62. Plato. (1953). Timaeus. In B. Jowett (trans.), The Dialogues of Plato, vols. 1–4, 4th edn (pp. 631–780) (Oxford: Clarendon Press). Plato. (2000). Republic, ed. by G.R.F. Ferrari, trans. by T. Griffith (Cambridge: Cambridge University Press). Schofield, M. (1991). The Stoic Ideal (Cambridge: Cambridge University Press). Schofield, M. (2003). ‘Stoic ethics’. In B. Inwood (ed.), The Cambridge Companion to the Stoics (pp. 233–56) (Cambridge: Cambridge University Press). Sedley, D. (2003). ‘The School, from Zeno to Arius Didymus’. In B. Inwood (ed.), The Cambridge Companion to the Stoics (pp. 7–32) (Cambridge: Cambridge University Press). Seneca. (1995a). Moral and Political Essays, ed. by J.M. Cooper and J.F. Procopé (Cambridge: Cambridge University Press). Seneca. (1995b). ‘On anger’. In J. M. Cooper and J.F. Procopé (eds.), Moral and Political Essays (pp. 1–116) (Cambridge: Cambridge University Press). Seneca. (1995c). ‘On mercy’. In J. M. Cooper and J.F. Procopé (eds.), Moral and Political Essays (pp. 117–64) (Cambridge: Cambridge University Press). Shakespeare, W. (1922) Coriolanus, ed. by W.J. Craig and R.H. Case (London: Methuen). Thucydides (1982). The Peloponnesian War, ed. by T.E. Wick (New York: The Modern Library). White, M. (2003). ‘Stoic natural philosophy’. In B. Inwood (ed.), The Cambridge Companion to the Stoics (pp. 124–52) (Cambridge: Cambridge University Press). Wood, N. (1983). ‘The economic dimension of Cicero’s political thought: property and state’, Canadian Journal of Political Science 16.4, 739–56.
15 Metaphor in the History of Ideas and Discourses: How Can We Interpret a Medieval Version of the Body−State Analogy? Andreas Musolff
Introduction The use of political body and illness metaphors in the Middle Ages and the Renaissance has been portrayed in the history of ideas as a preliminary stage to ‘modern’ uses, due to its alleged grounding in ‘humoral’ medicine and micro-/macrocosm cosmology. This chapter reviews these claims through analysing the use of the body–state metaphor in the twelfth-century cleric John of Salisbury’s treatise Policraticus. The analysis shows that John’s metaphor use is highly rhetorical and that his scenario of medical treatment for a diseased state even resembles present-day usage in some respects. In conclusion, we discuss ways in which a discourse-oriented perspective can link cognitive and historical approaches to metaphor analysis.1
1 Political body metaphors in history The mapping of body-related concepts onto political entities is one of the most famous, and notorious, conceptual metaphors in the history of ideas. Even in antiquity, the image of the state as a (human) body inspired not just a few isolated comparisons and applications but informed complete parables such as that of the ‘fable of the belly’, in which the other parts of the body rebel against the stomach, only to learn that their revolt leads to their own demise (Hale, 1971: 269; Koschorke et al., 2007: 15–20). 233
234 Andreas Musolff
In Western traditions, the body−state metaphor2 became a commonplace in poetry and philosophy throughout the Middle Ages, humanism and the Renaissance, most famously, perhaps, in Shakespeare’s dramas (Hale, 1971; Guldin, 2000; Koschorke et al., 2007; and the contributions by Banks and Zavadil in this volume). In modern times, the metaphor has been employed extensively in political propaganda, with an horrific climax in Nazi ideology (Coker, 1967; Sontag, 1991; Rash, 2006; Musolff, 2007). It has also motivated a set of lexicalized English expressions that are still used today without evoking strong ‘imagistic’ associations, such as body politic, head of state/government, organ (of a party) etc. (Deignan, 1995: 2; Brewer’s Dictionary of Phrase and Fable, 1999: 149, 713). The body−state metaphor appears to have completed its ‘evolution’ or ‘life cycle’, as described by William Croft and D. Alan Cruse (2004: 20−45): ‘Once a metaphor takes hold in a speech community and gets repeated sufficiently often [ ... ] its character changes. First, its meaning becomes circumscribed relative to the freshly coined metaphor, becoming more determinate; second, it begins to be laid down as an item in the mental lexicon; third, it begins a process of semantic drift, which can weaken or obscure its metaphorical origins.’3 However, in the case of political body imagery, there appears to be some life left in the old metaphor: not only is its source motivation still transparent to language users, but also its productiveness seems unbroken. Whether it is a British politician mocking himself as being the ‘toenail in the body politic’ (Boris Johnson, MP, quoted in The Independent on Sunday, 20 November 2005), or observers of the European Union who debate about where the heart of the Union lies and whether it is healthy or diseased (Musolff, 2004) – the metaphor’s mapping potential is routinely revived and extended time and again. Susan Sontag, in her celebrated (1991) essay, Illness as Metaphor, showed that some of its worst applications, i.e. the depiction of social problems or problem groups as deadly illnesses (e.g. cancer) of the state that must be fought and exterminated, have continued to appear not just in totalitarian propaganda but also in international political discourse, in the media jargon of the 1960s and 1970s and even in her own journalistic writings (Sontag, 1991: 82−5). This longevity gives the lie to the ‘obituaries’ that the metaphor has received occasionally, e.g. from David Hale who asserted that it ‘had lost most of its validity by the middle of the sixteenth century’, and that later applications ‘were [ ... ] void of any implications rising from the analogy’ (Hale, 1971: 131–6). Rather than ‘dying’, the body– state metaphor ‘survived’ the transition from the Renaissance to the
Interpreting the Medieval Body–State Analogy
235
Enlightenment and into modern times up to the present day. Adopting naturalistic terminology (and imagery), we could say that it adapted to the evolutionary pressures of its conceptual and socio-historical environment. Sontag already formulated a strong hypothesis about this conceptual evolution, which still needs empirical testing; she detected a qualitative difference between classical and modern uses: ‘classical formulations which analogize a political disorder to an illness – from Plato to, say, Hobbes – presuppose the classical medical (and political) idea of balance. [ ... ] Treatment is aimed at restoring the right balance [ ... ]. The prognosis is always, in principle, optimistic. Society, by definition, never catches a fatal disease’ (Sontag, 1991: 77). Since the Enlightenment, however, ‘the use of disease imagery in political rhetoric implies other, less lenient assumptions’ and especially the modern idea of revolution ‘shattered the old, optimistic use of disease metaphors’ (1991: 80). These new fatalistic diagnoses of political diseases culminated in totalitarian ideologies, which preferred the imagery of all-powerful ‘master illnesses that required matching radical cures’ (1991: 73−4). Sontag’s reading of modern, and especially twentieth-century, uses of the metaphor as a perversion of classical traditions of conceptualizing politics forms a narrative that may be compared to that of earlier historians of ideas. For example, Arthur O. Lovejoy and E.M.W. Tillyard in their studies of Neoplatonist traditions, including the body−state metaphor and the Great Chain of Being, spoke of such traditions coming ‘full circle’ in the contemporary ‘policy of great states and the enthusiasms of their populations’ (Lovejoy, 1936: 313), or resembling ‘trends of thought in central Europe’ that had brought ‘the world into its present conflicts and distresses’ (Tillyard, 1982: 117). Tillyard formulated a further historical hypothesis that complements the one put forward by Sontag. Whilst she proposed that from the Enlightenment onwards dramatic analogies of political crises with ‘master illnesses’ had increased, Tillyard maintained that during the Renaissance the traditional notions of a stable order of ‘correspondences’ between all levels of the Great Chain of Being, including those of human, social, and heavenly bodies, had changed their epistemological status. Whilst in the Middle Ages this system of correspondences was ‘accepted knowledge’ in a quasi-literal sense, humanism and the Renaissance transformed it into an inventory of rhetorical and poetic metaphors. According to Tillyard, ‘the correspondence between macrocosm, body politic, and microcosm’ acquired ‘a double function’ – it still served, as in medieval times, to express the idea of a cosmic order, but ‘[its Renaissance users] no longer allowed the details to take the
236
Andreas Musolff
form of minute mathematical equivalences: [ ... ] equivalences shaded off into resemblances’ (Tillyard, 1982: 107). In this perspective, the modern versions of body−state metaphors, including twentieth-century ‘distortions’, again appear to have a conceptually different status from those of pre-modern times. Semantically, they are more flexible and can be adapted to contrasting rhetorical registers. Sontag’s interpretation would provide a specification of this hypothesis to the effect that in political discourse since after the Renaissance watershed, body/illness metaphors have been increasingly separated from their traditional focus on problems of maintaining and restoring a healthy balance in the state and instead have been conceptually remoulded so as to express ideological evaluations of whether a given socio-political entity deserves to exist at all. In this study, we shall test the premises of Tillyard’s and Sontag’s historical hypotheses, i.e. that in pre-modern times the body–state mapping was part of a stable cosmic, eschatological knowledge system and less adaptable to rhetorical variation, by analysing one of the most famous medieval sources for the body–state analogy, i.e. the work of the cleric and philosopher John of Salisbury (c. 1115–180). Before eventually becoming Bishop of Chartres, John, as secretary to two Archbishops of Canterbury, Theobald and Thomas Becket, was involved in the conflicts between king and Church that led to the assassination of the second of these archbishops and to John’s own exile in France.4 During his exile he wrote Policraticus, a Latin treatise contrasting, as the subtitle states, the ‘frivolities of courtiers’ (to be avoided) with the ‘footsteps of philosophers’ (to be followed).5 In Books V and VI of the Policraticus, John develops the notion of the state (res publica) in its medieval form in terms of the analogy to the human body, on the supposed authority of an ‘instruction’ to the Emperor Trajan by Plutarch.6 Does this use of the analogy from around the middle of the twelfth century serve as evidence for the presumed high degree of integration of the metaphor into a quasi-literal ‘worldview’ and an ‘optimistic’ bias in favour of restoring balance in the state?
2
The body of the ‘res publica’
In the second chapter of Book V of the Policraticus, John gives a summary of the body-state analogy in the form of an extended simile, attributed to Plutarch. It deserves being quoted in detail, as it provides the reference point for John’s conclusions about desirable and undesirable states of affairs in politics and about political strategy.
Interpreting the Medieval Body–State Analogy
237
(1) For a republic is [ ... ] a sort of body which is animated by the grant of divine reward and which is driven by the command of the highest equity and ruled by a sort of rational management. [ ... ] The position of the head of the republic is occupied [ ... ] by a prince subject only to God and to those who act in His place on earth, inasmuch as in the human body the head is stimulated and ruled by the soul. The place of the heart is occupied by the senate, from which proceeds the beginning of good and bad works. The duties of the ears, eyes and mouth are claimed by the judges and governors of the provinces. The hands coincide with officials and soldiers. Those who always assist the prince are comparable to the flanks. Treasurers and record keepers [ ... ] resemble the shape of the stomach and intestines; these, if they accumulate with great avidity and tenaciously preserve their accumulation, engender innumerable and incurable diseases so that their infection threatens to ruin the whole body. Furthermore, the feet coincide with peasants perpetually bound to the soil, for whom it is all the more necessary that the head take precautions, in that they more often meet with accidents while they walk on the earth in bodily subservience; and those who erect, sustain and move forward the mass of the whole body are justly owed shelter and support. Remove from the fittest body the aid of the feet; it does not proceed under its own power, but either crawls [ ... .] or else is moved with the assistance of brute animals. (John of Salisbury, 1990, pp. 66–7) In this presentation of the body-state analogy, the image of the ‘republic’ as a human body is systematically developed, succinctly ordered and at crucial points backed up with further arguments. We find a transparent hierarchical perspective from the head ‘down’ to the feet and advice regarding important practical implications (avoidance of malfunctioning of the stomach/treasurers; necessity to look after the feet/peasants). The specifications that the head/prince is ruled by the soul/Church and is assisted by the heart/senate show that his position is not that of an absolute monarch in a modern sense but instead viewed in a religious and Church-oriented perspective. This latter point has drawn the attention of historians who are interested in the conflicts between monarchical and Church power in the High and Later Middle Ages. Kantorowicz (1997: 94–7, 194–200) and Hale (1971: 40) have linked John’s qualification of the head’s power to his condemnation of tyranny as opposed to lawful kingship and to his involvement in the conflict between the English kings Stephen and Henry II and the Roman Church as represented by the archbishops of Canterbury, which culminated in Thomas Beckett’s ‘murder in the cathedral’.
238 Andreas Musolff
A further focus of historical interest has been John’s attention to problems caused by the stomach and the concern for the feet: in these cases, the influence of John’s teachers at Paris and at the School of Chartres, notably Robertus Pullus (1080–1154) and Guillaume de Conches (c. 1080–1154), has been cited (Liebeschütz, 1950: 22; Nederman, 1990: xvi; Bass, 1997: 203–6). Some regard the Policraticus as part of a reformist movement of ‘medieval humanism’ that transcended traditional paternalistic concepts of state and society.7 The main ‘lesson’ to be learnt from the body-state analogy – which is re-stated time and again in the Policraticus and is particularly evident from John’s emphasis on the care due to the lowest members, the feet, is that notwithstanding their hierarchical relationships, all body parts depend on each other and must work together in order to enable the body to stay healthy and function properly: (2) The health of the whole republic will only be secure and splendid if the superior members devote themselves to the inferiors and the inferiors respond likewise to the legal rights of their superiors, so that each individual may be likened to a part of the others reciprocally and each believes what is to his own advantage to be determined by that which he recognises to be most useful to others. (John of Salisbury, 1990: 126) So far, the body-state metaphor in Policraticus seems to fit perfectly into the schema of a ‘medieval-as-pre-modern’ worldview with a stable perspective on social and political entities as parts of the system of correspondences between ‘microcosm’ and ‘macrocosm’, which could be believed quasi-‘literally’ as a whole and unbroken Chain of Being. Consequently, Tillyard (1982: 103) saw no problem in presenting John’s book as ‘one of the most elaborate medieval statements’ of the analogy. However, there are elements of the analogy in Policraticus that do not fit easily with its interpretation as a standard version of the pre-modern understanding of the body-state metaphor, which we shall study next.
3
Crisis and disease in the political body
One internal complication of John’s presentation of the political body metaphor in Policraticus is the occurrence of a competing version to the head-to-feet model in the form of the ‘fable of the belly’, which goes back to ancient Aesopian traditions and which was turned into a masterpiece of political rhetoric four centuries after John of Salisbury by
Interpreting the Medieval Body–State Analogy
239
Shakespeare in Menenius’ dispute with rebellious citizens in Coriolanus.8 The central, most important organ in the fable, in its ancient and also in later versions, is not the head, which is top in John’s hierarchy of body parts, as in example (1), but the belly, i.e. the stomach. The gist of the fable is that the ‘working’ members of the body (i.e. common citizens) protest against the perceived laziness of the belly (i.e. a ruling class or institution) that receives all the food. However, when the other members move from protest to rebellion, it quickly becomes clear that the belly is at least as essential as any other body part for the well-being of the whole. Once it is deprived of nourishment all the other members also fall ill, and the only chance of rescue lies in the resumption of the ‘orderly’ distribution of resources. Within Policraticus, the fable appears as a lesson taught to the author by none other than the then reigning Pope, Adrian IV, in a purported conversation between John and the pontiff. On the Pope’s own bidding, John, after having reported on complaints against the Church concerning corruption and simony, challenges the head of the Church himself: ‘If you are father, therefore, why do you accept presents and . payments from your children?’ (John of Salisbury, 1990: 135) The Pope responds by telling the fable and deducing from it the ‘obvious’ conclusion: ‘Measure neither our harshness nor that of secular princes, but attend to the utility of all’ (1990: 136). Both Hale (1971: 40) and Guldin (2000: 57–8) note an apparent discrepancy of the belly vs. the head as occupying the supreme position of power. Viewed from a text/ discourse-oriented perspective, however, there is hardly a conflict of meaning. In the first place, the purported contexts are different: one is a quotation from the Pseudo-Plutarch, the other a report of an alleged conversation with the Pope. John uses the fable as a separate ‘example’ story, which he could assume to be known to his learned readers: it does not compete with, but rather corroborates, the main analogy. Secondly, the target referents are not quite the same: one is the (secular) prince, the other is the Pope, who is a ‘prince of the Church’ but has different and, from John’s point of view, perhaps even more important, functions than the worldly prince. He is God’s representative on earth: as leader of ‘those who act in [God’s] place on earth’ he represents the soul that ‘rules’ the head/prince. Thus John is talking about two distinct, though related institutions at the target level. On the other hand, the argumentation import of the two passages is relatively similar, i.e. the emphasis on the necessity of mutual cooperation among all body members. The fable of the belly and the main head-to-feet analogy even match in one conceptual-source aspect: in both narratives,
240 Andreas Musolff
the stomach and intestines represent socio-economic management. In the secular state, the monarch’s ‘treasurers and record-keepers’ fulfil this task (see example 1), whilst in the Church it is performed by the Pope’s administration. What matters most for the conclusiveness of the analogy are the dangers for the health of the body if the stomach/belly/bowels do not function properly. Whether they are starved of food (as in the fable of the belly) or keep it for themselves (in John’s main analogy) – in both cases they ‘engender [...] diseases so that their infection threatens to ruin the whole body’ (John of Salisbury, 1990: 67). John uses the analogy not as a rigid classificatory schema but as an argumentation warrant for socio-political conclusions that he considers necessary to maintain the well-being of the commonwealth. His warning of ‘diseases’ that can ‘ruin the whole body’ leads us to a further aspect worth investigating with view to Sontag’s hypothesis quoted earlier – i.e. that in ‘classical’ formulations up to the period of Thomas Hobbes, five centuries after John of Salisbury, the ‘prognosis is always, in principle, optimistic’ and that the political body, ‘by definition, never catches a fatal disease’ (Sontag, 1991: 77). In this perspective, the purpose of the analogy as presented in the Policraticus can be interpreted as a warning addressed to the leadership of the State and Church to enable them to avoid or manage such diseases and thus ‘to encourage rulers to pursue a more rational policy’ (Sontag, 1991: 78). However, some instances of illness imagery pertaining to the bodystate analogy in Policraticus are so drastic that a benign or optimistic reading may appear too far-fetched. In Book V, Chapter 7, John quotes Plato as having warned of an oppressive magistrate being equal to a ‘swollen head’ that makes it ‘impossible for the members of the body to endure it either at all or without difficulty and even leads them to suicide, for ‘if the affliction would be incurable, it is more miserable to live than to die’ (John of Salisbury, 1990: 63). John utters further dire warnings as regards shielding the heart from injury, ensuring responsible direction of the hands and eyes and caring for the feet (1990: 84, 105, 126). After the rendition of the fable of the belly, in which the death of the whole body is only avoided narrowly (1990:135–6), John reasserts the dangers of injury to all body parts if the head is wounded: ‘a blow to the head, as we have already said, is carried back to all the members and a wound unjustly afflicted upon any member whomsoever tends to the injury of the head’ (1990: 137). If, on the other hand, the life of the body as a whole is at stake on account of a fundamental health crisis in the state, i.e. in the case of a rebellion, the ‘guilty’ members are not just dispensable, but it is the Prince’s duty to amputate and eliminate them.
Interpreting the Medieval Body–State Analogy
241
Invoking biblical authority (Matthew 18:9), John insists on the most radical form of therapy: (3) That the members are likewise to be removed is clear from that which is written: ‘If your eye or your foot offend you, root it out and cast it away from you.’ I think this is to be observed by the prince in regard to all of the members to the extent that not only are they to be rooted out, broken off and thrown far away, if they give offence to the faith or public security, but they are to be destroyed utterly so that the security of the corporate community may be procured by the extermination of the one member. Who will be spared, I say, by him who is commanded to do violence against even his own eyes? Indeed, neither the ears nor the tongue nor whatever else subsists within the body of the republic is safe if it revolts against the soul for whose sake the eyes themselves are gouged out. (John of Salisbury, 1990: 140–1) The justification John provides for his zealous exhortation in favour of amputation and elimination is the fact that the injury of rebellion concerns the soul, namely ‘when God is offended by abuses of criminals or the Church is spurned’ (1990: 141). The severity of the therapy can hardly be surpassed and its description shows little sign of optimism or concern for rational disease management. In John’s view, ‘rebellion’ was probably what Sontag called a political ‘master illness’ in the modern metaphorical sense: ‘Now, to liken a political event or situation to an illness is to impute guilt, to prescribe punishment’ (Sontag, 1991: 82). The epitome of this pessimistic vision can be found in the last book of the Policraticus, where the exact counter-image of a proper political body is drawn, namely the ‘republic of the impious’: (4) Its tyrannical head [ ... ] is the image of the devil; its soul is formed of heretical, schismatic and sacrilegious priests and, to use the words of Plutarch, prefects of religion, assailing the laws of the Lord; the heart of impious counsellors is like a senate of iniquity; its eyes, ears, tongue and unarmed hand are unjust officials, judges and laws; its armed hand is violent soldiers [ ... ]; its feet are those among the more humble occupations who oppose the precepts of the Lord and legitimate institutions. (John of Salisbury, 1990: 193–4) Here, the body-state analogy is fully present but it receives the complete opposite of any optimistic evaluation. From the head down to the feet,
242 Andreas Musolff
the devil’s anti-state forms a body mirroring that of the proper state, but now the function of every body part is destructive, oppressive and unhealthy. The same source scenario as that of the healthy ‘res publica’ thus also serves to depict the opposite target of the tyrannical state, to highlight its utter depravity.
4 Policraticus as part of a history of discourse about states as human bodies Prima facie, the investigation of the historical hypotheses formulated by Tillyard and Sontag concerning the turning points in the conceptual development of the body–state metaphor seems to have produced a negative result as regards John of Salisbury’s Policraticus.9 John’s depiction of illnesses, injuries and therapies can hold its own against any modern uses in terms of ferocity and varied application. He clearly knew every rhetorical trick to twist and turn the core concepts of the metaphor so as to fit his argumentation purposes, even by depicting the image of a tyrannical, devilish head and body to drive home his ethical message (see example 4). Nor did he shy away from using different source aspects of head and belly/stomach for similar argumentation purposes or from mixing his state–body analogy with further metaphors, such as those of rulers as fathers (1990: 81, 135), of the king as an image of equity or justice (1990: 28–38), or of the state as a beehive (1990: 127–9, 132). Moreover, he was willing to acknowledge the limits of the metaphor: he applied the concept of the feet not only to peasants but to all ‘those who exercise the humbler duties’ in society including weavers, artisans, servants, procurers of food and managers of private households (1990: 125). This extension of the target ‘scope’ (Kövecses, 2002: 108–9) of the metaphor leads John to mock his own application of it: ‘there are so many of these occupations that the number of feet in the republic surpasses not only the eight-footed crab, but even the centipede’ (1990: 126). John’s awareness of the absurdity of counting and distinguishing all the feet of the republic is evidence that for him the body–state analogy was not at all a matter of ‘equivalences’ but a rhetorical figure that served its purpose up to a certain point and then could be turned into a joke. John was clearly also aware of the conceptual and literary traditions in which he formulated his ideas, as evidenced by the sophisticated use of the fable of the belly and of a host of biblical and classical references in Policraticus; indeed, he may have invented a classical authority, the Pseudo-Plutarchian instruction, to create a tradition where there had been none. What does this tell us about the status of his text in the
Interpreting the Medieval Body–State Analogy
243
history of ideas and in the history of discourses organized by the body– state metaphor? Conceptual history has successfully established a corpus of research that allows us to locate the formulation of the body–state metaphor by John of Salisbury both within the contemporary setting of politicotheological theory, such as the School of Chartres, and within longterm traditions that connect Neoplatonism, Pauline concepts of the Church as Christ’s body and medieval political philosophy. These medieval concepts were the background for Renaissance and Enlightenment traditions, some of which have provided the platform for powerful conceptualizations of society and politics up to the present day. However, the narrative of a continuous development, which is characteristic of the ‘history of ideas’ approach rests on the assumption that the conceptual unit, i.e., here, the body–state metaphor, can be isolated, that it stays essentially the same and that its history can be described by unfolding its immanent entailments through time. This assumption is, however, at least debatable. Sure enough, it is possible to assemble a corpus of chronologically continuous texts containing body– state mappings. It may also be plausible to argue that in a broad sense, the experiential basis of the source domain input in these metaphors remains roughly similar at the level of physiological experience. However, there is ample evidence of significant changes in popular medical knowledge, and hence of the source for body-based metaphors, in Western cultural traditions over the last millennium (Geeraerts and Grondelaers, 1995; Kövecses, 1995 and 2002: 186–7). The Hippocratic and Galenic traditions that a twelfth-century cleric could take for granted have paled into obscurity and have survived only in the terminology of melancholic, choleric etc. temperaments whose etymology and place in the system of medical philosophy and cosmology are little known to most contemporary speakers. The fact that we can still recognize the salient body parts that John of Salisbury referred to is by no means sufficient to reconstruct the meaning of body/illness metaphors in his political philosophy. For that, we have to take into account his presumptive aim of explicating the complexities of state and society ‘to his fellow clerical bureaucrats about how to avoid potential misfortunes’ (Nederman, 1990: xix). On the other hand, if we want to avoid anachronism, we must refrain from projecting contexts of use that may have been of importance later (such as the alleged turning-point of its rhetorical and poetic use in the Renaissance, or the alleged distance of pre-modern medical thought to ‘modern’ concerns for ‘master illnesses’ such TB, cancer or AIDS) onto the Policraticus.
244 Andreas Musolff
This attention to the specific contemporary-discourse context of metaphor use may also help as a warning against the fallacy of reifying a conceptual development into a naturalistic model of concept development that posits a deterministic view of semantic drift, be it from more to less vivid conceptual richness or from metaphoric to lexicalized or literalized meanings (for a detailed critique of such approaches, see Frank, in this volume). As we have seen, John of Salisbury employed the analogy in a highly complex and imaginative way, reshaping it several times to fit his rhetorical and ideological purposes. The author of the Policraticus was aware of an already established tradition of viewing society and state as a body, as his references to the fable of the belly and canonical interpretations of St Paul (I Corinthians,:12, 12–7) show. After John wrote his book, a further eight hundred years of uses of the body– state metaphor have been documented. He thus stands neither at the beginning nor at the end of the metaphor’s conceptual career, and his formulations of it are neither more elaborate than that of earlier users, nor are they less complex, vivid or conclusive than those of later users. If the Policraticus is to be situated meaningfully within an historical account that hangs together by the thread of successive debates based on the body–state metaphor, we must not force it into the narrative of a linear progressive (or regressive) path of evolution. Instead, we can use it to inform ourselves about the cognitive potential of that metaphor for the interpretation of politics and thus help to ‘abolish the situation in which things worth knowing are not experienced’ (John of Salisbury, 1990, 3). To achieve this, we should combine the cognitive emphasis on the grounding of the body–state mapping in universal, experientially based conceptual structures with context-sensitive interpretations that explicate their function as discursive actions in specific historical circumstances. This eclectic approach will enable us to formulate and evaluate empirical hypotheses concerning the historical dimension of metaphor. Even if the global statements about major shifts and discontinuities in the history of political body metaphors, as formulated by historians of thought, may have to be revised in view of specific textual evidence: we need more, not fewer, such hypotheses – and their critique.
Notes 1. I am grateful to Jörg Zinken, Roslyn Frank, Kathryn Banks and David Cowling for helpful comments and criticisms of earlier versions of this paper. The research was supported by a grant for research leave from the Arts and Humanities Research Council.
Interpreting the Medieval Body–State Analogy
245
2. The term ‘metaphor’ is used here as a fundamental category covering the mapping of distinct domains of knowledge and experience in thought and discourse (Lakoff and Johnson, 1980 and 1999; Kövecses, 2002). Related terms from rhetorical traditions, e.g. ‘allegory’, ‘analogy’, ‘simile’, ‘parable’, are used as characterizations of specific manifestations of metaphor in individual texts. 3. For the general evolutionary model of language change that underlies the concept of semantic ‘life cycle’, see Croft (2000: 13–39). 4. For John of Salisbury’s life and career, see Wilks (1984) and Nederman (1990: xvi–xviii); for the textual history of Policraticus, see Nederman (1990: xviii–xix). 5. Cf. John of Salisbury (1990), the English translation, and (1909), the Latin original. John’s primary addressee was Thomas Beckett, to whom it was dedicated (John of Salisbury, 1990: 230). 6. Liebeschütz (1950: 23–4) declared the ‘instruction’ to be a literary fiction designed ‘to disguise the personal character of his political criticism’. Kantorowicz (1997: 199), Hale (1971: 39–41), Nederman (1990: xxi) and Koschorke et al. (2007: 78) accept this interpretation; for a differing view compare Bass (1997: 206). 7. For a strong emphasis on John’s ‘radically innovating and compassionate social vision’, see Bass (1997: 207–10); Liebeschütz (1950: 1, 24) also stresses John’s role in the ‘medieval renaissance’. Kantorowicz (1997: 200, 311), on the other hand, highlights John’s dependence on the concept of the Church as Christ’s body, derived from St. Paul’s Epistles (especially Romans, 12, 4–5; 1 Corinthians, 12, 12–28). Struve (1984: 309) and Hale (1971: 39–40) underline John’s loyalty to Church authority. Guldin (2000: 57–8) stresses the ‘paternalistic’ tenor , in contrast to Christine de Pizan’s fifteenth- century Le livre du corps de policie (Pizan, 1994). 8. Shakespeare (1983, Coriolanus: I,1:101–69). For the ancient tradition of the fable and Shakespeare’s adaptation of it, see Hale (1971: 27–8, 96–107), and Koschorke et al. (2007: 15–35, 52–4). 9. This relativization of Sontag’s characterization of ‘pre-modern’ uses in general – she did not mention John of Salisbury specifically – does not, of course, invalidate her analysis of modern uses nor her critique of the metaphorization of illnesses such as TB and cancer and also AIDS through war imagery (1991, passim).
References Bass, A. M. (1997). ‘The metaphor of the human body in the political theory of John of Salisbury: context and innovation’. In B. Debatin, T. R. Jackson and D. Steuer (eds.), Metaphor and Rational Discourse ( pp. 201–13) (Tübingen: Niemeyer). Brewer’s Dictionary of Phrase and Fable (1999), ed. by Adrian Room (London: Cassell). Coker, F. W. (1967). Organismic Theories of the State. Nineteenth-Century Interpretations of the State as Organism or Person (New York: AMS Press). Croft, W. and Cruse, D. A. (2004). Cognitive Linguistics (Cambridge: Cambridge University Press).
246
Andreas Musolff
Deignan, A. (1995). COBUILD English Guides. Vol. 7: Metaphor Dictionary (London: HarperCollins). Geeraerts, D. and Grondelaars, S. (1995). ‘Looking back at anger: cultural traditions and metaphorical patterns’. In J. R. Taylor and R. E. MacLaury (eds.), Language and the Cognitive Construal of the World (pp. 153–79 ) (Berlin: de Gruyter). Guldin, R. (2000). Körpermetaphern: Zum Verhältnis von Politik und Medizin (Würzburg: Königshausen & Neumann). Hale, D. (1971). The Body Politic: A Political Metaphor in Renaissance English Literature (The Hague: Mouton). John of Salisbury (1909). Policraticus sive De nugis Curialium et vestigiis philosophorum, ed. by C. C. I. Webb. 2 vols (Oxford: Oxford University Press). John of Salisbury (1990). Policraticus. Of the Frivolities of Courtiers and the Footprints of Philosophers. ed. and trans. by C. J. Nederman (Cambridge: Cambridge University Press). Kantorowicz, E. H. (1997). The King’s Two Bodies: A Study in Mediaeval Political Theology. With a new Preface by William Chester Jordan. (Princeton, NJ: Princeton University Press). Koschorke, A., Lüdemann, S., Frank, T. and Matala de Mazza, E. (2007). Der fiktive Staat. Konstruktionen des politischen Körpers in der Geschichte Europas. (Frankfurt/Main: Fischer). Kövecses, Z. (1995). ‘Anger: its language, conceptualization, and physiology in the light of cross-cultural evidence’. In J. R. Taylor and R. E. MacLaury (eds.), Language and the Cognitive Construal of the World (pp. 181–96) (Berlin: de Gruyter). Kövecses, Z. (2002). Metaphor. A Practical Introduction. (Oxford: Oxford University Press). Lakoff, G. and Johnson, M. (1980). Metaphors We Live By. (Chicago/London: University of Chicago Press). Lakoff, G. and Johnson, M. (1999). Philosophy in the Flesh. The Embodied Mind and its Challenge to Western Thought. (New York: Basic Books). Liebeschütz, H. (1950). Mediaeval Humanism in the Life and Writings of John of Salisbury (London: The Warburg Institute). Lovejoy, A. O. (1936). The Great Chain of Being. A Study of the History of an Idea (Cambridge, MA: Harvard University Press). Musolff, A. (2004). ‘The heart of the European body politic. British and German perspectives on Europe’s central organ’, Journal of Multilingual & Multicultural Development, 25, 5 & 6, 437–52. Musolff, A. (2007). ‘Which role do metaphors play in racial prejudice? The function of anti-Semitic imagery in Hitler’s “Mein Kampf” ’, Patterns of Prejudice, 41, 1, 21–44. Nederman, C. J. (1990). ‘Editor’s introduction’. In John of Salisbury, Policraticus. Of the Frivolities of Courtiers and the Footprints of Philosophers (pp. xv–xxviii), ed. and trans. by C.J. Nederman (Cambridge: Cambridge University Press). Pizan, C. de (1994). The Book of the Body Politic, trans. by K. Langdon Forhan (Cambridge: Cambridge University Press). Rash, F. (2006). The Language of Violence. Adolf Hitler’s Mein Kampf (New York: Peter Lang).
Interpreting the Medieval Body–State Analogy
247
Pope, M. (1983). ‘Shakespeare’s medical imagination’, Shakespeare Survey, 38, 175–86. Shakespeare, W. (1983). Coriolanus. In The Complete Works of William Shakespeare (pp. 758–97), ed. with a glossary by W. J. Craig (London: Pordes). Sontag, S. (1991). Illness as Metaphor. Aids and its Metaphors (London: Penguin). Spurgeon, C. F. E. (1993). Shakespeare’s Imagery and What It Tells Us. (Cambridge: Cambridge University Press). Struve, T. (1984). ‘The importance of the organism in the political theory of John of Salisbury’. In M. Wilks (ed.), The World of John of Salisbury (pp. 303–17) (Oxford: Basil Blackwell). Tillyard, E. M. W. (1982). The Elizabethan World Picture (Harmondsworth: Penguin). Wilks, M. ( ed.) (1984). The World of John of Salisbury (Oxford: Basil Blackwell). Williams, G. W. (1984). ‘Shakespeare’s metaphors of health: food, sport and lifepreserving rest’, Journal of Medieval and Renaissance Studies, 14, 187–202.
This page intentionally left blank
Commentary
This page intentionally left blank
16 Studying Metaphor in Discourse: Some Lessons, Challenges and New Data Raymond W. Gibbs, Jr. and Julia E. Lonergan
There is something surprising about the wonderful contributions in this volume on ‘metaphor and discourse’. On the one hand, the various studies of how metaphor permeates diverse types of discourse provide additional evidence for theories that emphasize the prominence of metaphor in thought, language, and culture. Yet on the other hand, it has taken metaphor scholars some time now to recognize that metaphors are not decontextualized entities, to be studied in isolation, as they attempt to see what each one may mean and how it works. Metaphors are, as the chapters in this volume clearly articulate, products of discourse, and thus are thoroughly contextualized, enough so that it is now questionable whether metaphor should ever be studied apart from the contexts in which they live. In saying this, we do not simply imply that discourse contexts give specific meanings to metaphors, as if metaphors are pre-existing entities of language or thought, with context adding seasoning to flavour each specific metaphor as it is used. Instead, and more dramatically, metaphors are inseparable from context because there is no division between metaphor and discourse, given that metaphors are both products of discourse and creators of discourse. Realization of this true state of affairs offers metaphor scholars several important lessons, and also raises some challenges as we continue to study metaphor in discourse. A first challenge concerns the implications of our specific studies for theories of metaphor. Metaphor scholars often seek to make theoretical generalizations about the fundamental nature of metaphoric thought and language from their specific discussions of particular metaphors in different discourses (e.g. German political ideas, ancient Roman political 251
252
Raymond W. Gibbs, Jr. and Julia E. Lonergan
thought, English concepts of the ‘superbug’, Shakespeare’s sonnets). This is similar to the type of inferential generalization psychologists, like us, engage in when we collect and analyse the data from a specific group of people (e.g. young college students participating in experiments). Thus, we are not usually interested in the specific performance of these students in psychological studies, but wish to draw more general conclusions about populations of people (e.g. all adults or all English-speaking adults with normal health). Drawing these inferences requires that certain cautions be taken into consideration given the possible errors that may arise from improperly assessing the risks associated with assuming that our specific subjects, and their performance, are representative of these larger populations. In a similar manner, metaphor scholars must be careful to not improperly assume that what one finds in any individual study of particular metaphors in specific discourses necessarily reflects something accurately about all metaphors in all discourses. One mistake that is often made in this regard is that metaphor scholars assume that metaphors can be taken out of discourse, and examined as individual entities, much as dead butterflies are pinned down in glass cases, to understand something about how they work. Just as butterflies are best understood out in the wild, metaphors are best studied and analysed within their natural contexts, as speakers and writers and producers of nonlinguistic forms of metaphors are engaged in real cognitive and social work. For instance, it seems reasonable to closely examine a particular discourse, or set of discourses, related to a specific topic or knowledge domain, and pick out the metaphors that appear to have a dominant role in structuring that discourse. One may go on from that point and draw tentative conclusions about the ways that particular speakers or writers, for example, might be thinking in metaphorical terms to provide some causal explanation for the metaphors that have been discovered. Furthermore, several contributors to this volume have correctly, in our view, emphasized that at least part of the motivation for many patterns of metaphor in discourse are due to social-cultural and ideological forces that speakers and writers live and function within. Linguistic metaphors, for instance, do not arise solely from individual minds based on how those people think (e.g. conceptual metaphors), because all thought and language is situated in social-cultural contexts that are inseparable from cognition in the same way that metaphor and discourse are inseparable. Once again, the study of metaphor in discourse is gaining recognition that the conceptual is not so distinct from the social-cultural, enough so that one can no longer continue
Commentary 253
to assume that individual metaphors can be pulled out of a discourse and analysed apart from that discourse to infer conceptual metaphors. Studying metaphor in discourse, therefore, opens up the social-cultural and discursive roots of metaphor for metaphor scholars with an interest in the cognitive, psychological, or embodied aspects of metaphor. Once more, saying that metaphors have part of their roots in the socialcultural does not mean that discourse context shapes the meaning of metaphor, but really functions as the very source of metaphor. A second, related implication of studying metaphor in discourse is the recognition that metaphor identification must take into account the positioning of metaphors in discourse and their relations to other metaphors in that and other related discourses. For example, the standard practice of selectively focusing on certain metaphorical expressions, or even single conceptual metaphors, without much attention to articulating the criteria by which that identification has been done, the frequency of that metaphor in discourse, or where in the discourse metaphors are situated, may no longer be acceptable. Consider the problem of metaphor identification. Judging the empirical adequacy of any individual analysis is usually done as a matter of belief in one’s own intuition, something supported by counter-examples. Yet a clearer description is needed of how metaphor analysts came up with their classifications. What are the criteria for determining what counts as a metaphorical expression in the language? Following this, what are the criteria for positing that a conceptual metaphor of some sort underlies the creation and use of a set of systematically related linguistic expressions? With a few exceptions (Cameron, 2003; Pragglezjaz Group, 2007), metaphor scholars have not provided criteria or guidelines by which they conduct their individual analyses of metaphor in language and thought. Providing such criteria will be essential toward placing interdisciplinary metaphor research on a firm empirical footing, which implies that the results of our analyses can be verified and replicated. There are, in fact, several proposals from computational linguistics on metaphor identification that may be useful for scholars studying metaphor in discourse. One program, MetaBank (Martin, 1994) contains a knowledge base of metaphor based on the Master Metaphor List compiled in the early 1990s (Lakoff et al., 1991), and can search for a particular metaphor in any larger corpus having metaphors in its knowledge base. These searches are done by choosing a set of particular words typically associated with a specific conceptual metaphor and then simply locating sentences containing those words. Once target
254 Raymond W. Gibbs, Jr. and Julia E. Lonergan
sentences are selected, MetaBank examines them to determine whether they express literal statements, instances of the target metaphor, or some other novel metaphor. Finally, MetaBank retains this data for its other search attempts to build up a more comprehensive list of probe words that are clearly associated with particular conceptual metaphors. MetaBank has been used in the analysis of metaphors in UNIX-based e-mail communications and the distribution of metaphors in the Wall Street Journal (Martin, 1994). A more recent computational program, CorMet, analyses large corpora of different domain-specific documents from the Internet to infer the selectional preferences of verbs associated with that domain (Mason, 2004). ‘A selectional preference is a verb’s predilection for a particular type of argument in a particular role’ (Mason, 2004: 231). For instance, in a data base of FINANCE documents, CorMet would encounter a statement like ‘Funds poured into his bank account’ and would analyse the verb ‘poured’ to see that it has a selectional preference, in this domain, for financial assets. CorMet then compares the selectional preference of ‘poured’ for FINANCE with that of the same verb in other domains, such as in LABORATORY WORK (e.g. ‘He poured the solution into a test tube’), and then would infer a metaphorical mapping from liquids to money. Across different explorations, such as the above, CorMet can infer different sets of co-occurring inter-concept mappings (e.g. between liquid and money, and containers and investments) to specify higher-order structures of conceptual metaphor. One analysis of CorMet’s performance against the Master Metaphor List showed that the program found similar metaphors as seen in the Master List 77% of the time, although CorMet was not able to detect some metaphors such as THEORIES ARE ARCHITECTURE OR BUILDINGS. Yet in another test, CorMet did not falsely infer metaphorical mappings between randomly selected pairs of domains. Unlike MetaBank, CorMet does not require any conceptual knowledge base, other than WordNet, such as the Master Metaphor List, to infer and identify metaphorical expressions based on enduring conceptual metaphors. Most generally, each of the above programs suggests that computational tools may be beneficial toward semi-automatically identifying certain expressions in discourse as being metaphorical, and indeed sometimes as being linked to widespread conceptual metaphors. We refer to these computational tools as just one way to better ground our individual analyses of metaphors in discourse to, again, make these studies more systematic and reliable, and to possibly avoid the present tendency to just assume that some word or phrase or piece of discourse
Commentary 255
conveys metaphorical meaning just because we think it does. Providing more explicit criteria for the metaphor analyses we conduct will surely be a major challenge in future studies of metaphor in discourse. One topic that arises in the study of metaphor in discourse is mixed metaphor. Several of the chapters in this volume either note the mixing of metaphors in different discourse analyses or explicitly aim to generally account for how and why mixed metaphors are produced and understood. Because mixed metaphor is such a prominent feature of discourse, and because it concretely illustrates the importance of not separating the cognitive from the social and pragmatic, we focus the rest of this commentary on this topic. There are many ways in which metaphorical words, phrases, and broader metaphorical themes can get jumbled together in real speech and writing, some of which can evoke humorous reactions. Consider the following example, first appearing in The Times newspaper (12 October 1998), and later reprinted in The New Yorker magazine as an item in its long-running filler entitled ‘Block that Metaphor!’: ‘As I look at it with a broad brush, there are a lot of things going south at the same time,’ said Morris Goldstein, a former International Monetary Fund official and senior fellow at the Institute for International Economics in Washington. ‘Where is the good news coming from? Is there no silver bullet out there?’ Although your typical English teacher would advise people to not speak or write in this manner, the fact is that mixed metaphors of various types litter speech and writing and people appear able to somehow understand something about what others intend to communicate in these situations. Although there is various psycholinguistic research demonstrating that switching metaphors does demand extra processing effort on the part of listeners and readers (Gentner, et al., 2002; Langston, 2002; Nayak and Gibbs, 1990), there has not been much empirical work looking at exactly what people really interpret when, for example, they read the bit of discourse reproduced above. Lonergan (in prep.) has conducted an initial experiment to explore the consistency in ordinary readers’ understanding of mixed metaphors, such as the above, with the broader aim of explaining why people do mix, and what implications this has for theories of metaphor use and understanding in real discourse. In Lonergan’s study, a group of 20 college students were presented with eight different instances of mixed metaphor taken directly from the ‘Block that Metaphor!’ column during the years 1996 to 2006. These
256 Raymond W. Gibbs, Jr. and Julia E. Lonergan
vignettes were first analysed according to the ‘metaphor identification procedure’ outlined by the Pragglejaz Group (2007) to indicate which phrases really conveyed metaphorical meaning. Each vignette was then presented to participants clause-by-clause (e.g., ‘As I look at it with a broad brush’), with each phrase shown on a separate page of a booklet. The participants’ task was to read each phrase and write down a response to ‘What is the speaker’s meaning?’ for each one. There were no restrictions on how much a participant could write in response to each phrase, but they typically wrote one or two lines. For instance, consider some of the participants’ responses to the first phrase, ‘As I look at it with a broad brush’,’ from the above vignette: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
He is saying how bad the problem is. I understand it to mean ‘generally’ The meaning has to do with the broad scope of things. Have an open mind. The speaker looks at it with an open mind and general understanding As one looks at it generally and quickly. The individual is looking very broadly. He is looking without account for small things. Looking at the big picture, in broad terms. He is looking with bushy eyebrows intensely at something. He is looking at it with an open mind. As I look at it with a broad mind that is sharp.
There is something consistent in this group of protocols, despite differences in whether participants chose to state their interpretations in metaphorical (e.g., ‘big picture’) or non-metaphorical (e.g. ‘looks at it generally and quickly’) terms (i.e. 50% metaphorical, 50% non-metaphorical). A subsequent Cognitive Linguistic analysis on the protocols for this example suggested that there were two key conceptual metaphors, SEEING IS UNDERSTANDING and IDEAS ARE OBJECTS, shaping people’s interpretations of ‘As I look at it with a broad brush’. These two conceptual metaphors have been well established as supra-individual concepts within the Cognitive Linguistic literature (Kövecses, 2002). Overall, 88% of the participants’ responses to all phrases across all the vignettes maintained the ‘scope of the source domain’ (Kövecses, 2005). At the very least, this generally suggests that people have fairly regular readings of the metaphorical phrases, despite the fact that they appear, on the surface, to be rather jumbled. But most importantly,
Commentary
257
participants’ protocols specifically cohered to the idea that the OBJECTS in IDEAS ARE OBJECTS have ‘dimensions’, or the subcategory metaphor of ‘object dimensions’. A subcategory is a restriction on the conceptual metaphor that still maintains a perceptual relation to the source domain. Thus, in the above example, the OBJECT (or theme) is a VOLUNTARY-PERCPETUAL-EVENT OF SIGHT. The submetaphor is structured according to the metonymic notion of a BRUSH AS AN INSTRUMENT OF SIGHT. Students’ protocols, again, give evidence of their recognition of this submetaphor through their lexical choices of phrases like ‘big picture’, ‘broad scope’, and ‘no small things’. A different example of participants’ conceptual metaphorical, and submetaphor, understandings of the mixed metaphor is seen in their responses to the second phrase in the above vignette, ‘There are a lot of things going South at the same time’. In response to this phrase, participants mentioned various things referring to the general idea that DOWN IS BAD, and the more specific submetaphor of GOING SOUTH IS DOWNWARD MOTION, as illustrated by their protocols of what the speaker meant such as ‘going downhill’, ‘lots of things that are hard to see’, ‘many things are in decline’, and ‘many things going bad at the same time’. Finally, consider participants’ responses to the last phrase in the above mixed metaphor vignette, ‘Is there is no silver bullet out there?’ They stated that the speaker meant, ‘no magic’, ‘no pot of gold’, ‘no hero’, ‘no high plateau’, ‘nothing genuine’, and ‘nothing sturdy’. Each of these responses reflects people’s understandings of the phrase as referring to the conceptual metaphor IDEAS ARE OBJECTS, and more specifically to the submetaphor RARE IDEAS ARE RARE OBJECTS. Although the participants’ lexical choices in offering their interpretations differed, they remained consistent with the idea that solutions to the problem referred to by the speaker are conceived of as rare objects, or objects that within Western culture are typically seen as rare. Overall, 90% of all participants’ responses to the individual phrases in the different mixed metaphors they read stabilized around the same mappings noted by the relationships between the subcategory metaphors and the conceptual metaphors. These data are especially relevant to two related claims by Kövecses (this volume) and Barnden (this volume) who attempt to address the challenging problem of metaphor variation in discourse. For instance, Kövecses (2005: 237) states: Metaphorical source domains for particular targets may arise from what I call the ‘pressure of coherence.’ This means that speakers try
258
Raymond W. Gibbs, Jr. and Julia E. Lonergan
(and tend) to be coherent with various aspects of the communicative situation in the process of creating metaphorical ideas. The communicative situation has to be understood as minimally composing the audience, the medium, the topic, and the setting. All of these may play a role in creating metaphors that may not exist in the most conventionalized, the standard, variety of a language or in the creation of differential uses of metaphors in communicative situations that are characterized by different purposes and affective relationships between author and audience. Kövecses’ ‘pressure of coherence’ theory provides a global explanation for why it is completely reasonable for speakers, such as the one in the above vignette, to mix their metaphors, and express complex ideas in non-standard ways. At the very least, our own results on mixedmetaphor understanding suggest that people are quite capable, and remarkably consistent, in coming up with reasonable interpretations of what speakers hope to communicate by their juxtaposing of ideas that on the surface seem completely bizarre in relation to the topic referred to (e.g. in this case, international economics being talked about via ‘broad brushes’, ‘things going south’, and ‘no silver bullet’). But more specifically, Kövecses (this volume) states, ‘When we actually use these metaphorically constituted concepts in real discourse, it is often the case that we choose metaphorical expressions that are not constitutive of our understanding of the target concept in question in discourse but that are based on an already existing metaphorical understanding of a model of a target domain’. This notion explains, then, why speakers refer to specific submetaphors, such as RARE IDEAS ARE RARE OBJECTS, and employ phrases like ‘There is no silver bullet out there’, to refer to the idea that there is no easily available solution to deal with the economic crisis, even though the specific phrase ‘silver bullet’ has little conventional relation to ideas or thoughtful solutions to economic dilemmas. Barnden (this volume) also offers a very relevant suggestion about the same kind of variation we saw in our mixed metaphor corpus, when he noted, ‘metaphor often involves non-parallelism’. Thus, many metaphors in discourse obtain their meanings less from systematic mappings from source to target domain (a form of parallelism), and more from compensatory within-source-domain reasoning. In previous writings, Barnden and Lee (2001) claim that many metaphor variants are ‘mapping transcendent utterances’ where various ancillary assumptions constrain the imaginary situation in the source domain.
Commentary
259
We see our mixed metaphor corpus and participants’ interpretations of these vignettes to be quite compatible with these ideas. More specifically, mixed metaphors, such as the one vignette presented above, reflect exactly the sort of within-source-domain reasoning, often shaped by many of the ancillary assumptions referring to causes, events, mental/emotional states, and value judgments, posited by Barnden and Lee (2001; and Barnden, this volume) to be ‘view-neutral mapping adjuncts’. These adjuncts are default mappings that are ‘view-neutral’ in not being tied to any specific metaphorical view (i.e. conceptual metaphor). For example, the meaning of ‘Is there no silver bullet out there?’ in the above mixed metaphor, arises from both the conceptual metaphor of IDEAS ARE OBJECTS, and the more specific submetaphor of RARE IDEAS ARE RARE OBJECTS, but also from view-neutral mapping adjuncts such as ‘value judgment’, indicating that the level of importance assigned by individuals to the source domain (e.g. the rarity and importance of silver bullets) maps onto the level of importance assigned to the target domain (e.g. the low likelihood of finding a solution to the economic crisis). Other view-neutral mapping adjuncts, such as ‘function’ suggest that the purposes or functions of entities in the source domain (e.g. function of special bullets to kill special, dangerous people) gets mapped onto the target domain of finding a special solution to deal with the special economic problem at hand. Most important, our participants’ interpretations of ‘Is there no silver bullet out there?’ tap into these view-neutral mapping adjuncts quite nicely when they referred to other instances of these mappings, such as other things of great value and functions such as ‘magic solution’, ‘pots of gold’, ‘no hero’, and ‘high plateau’, in addition to some non-metaphorical descriptions of these same ideas such as ‘there is nothing to take the pain/suffering away’ and ‘there is not one solution to the problem’. These interpretations reflect additional withinsource-domain reasoning similar to that which enabled the speaker of the original mixed metaphor to say what he did in the first place. Finally, it is critical to note that our participants’ interpretations of the mixed metaphors were not simply based on their static understanding of what each individual metaphorical statement means. Instead, people read the different, and mixed up, metaphors according to the very brief contexts in which they appeared, and clearly recognized the evaluative work that these metaphors were providing (see Musolff, 2004). When asked at the end of reading the vignettes line-by-line to write a summary description of what the speaker meant to communicate,
260 Raymond W. Gibbs, Jr. and Julia E. Lonergan
participants consistently were able to articulate the broader value judgment or evaluation of the topic suggested by the speaker. For this reason, we can reliably conclude that mixed metaphors convey coherent ideas about often abstract topics, with ordinary readers being able to both understand what individual metaphors mean in context, and gain an overall sense of coherence in conveying social evaluations. The beauty of this work, as is the case with the chapters in this volume, comes from the tight interplay between widely pervasive metaphorical schemes (i.e. conceptual metaphors) and very specific sorts of within-source-domain reasoning, with the latter being tightly constrained by various ancillary assumptions which surely reflect different socio-cultural knowledge and beliefs. In this way, the empirical study of metaphor in discourse drives home the important lesson that a great deal of our thinking is tied to acts of speaking and writing, or communicating, where it may not be possible to strip away the conceptual from the pragmatic, or thoughts away from the discourse contexts in which they occur.
References Barnden, J. and Lee, M. (2001).Understanding Open Ended Usages of Familiar Conceptual Metaphors: An Approach and Artificial Intelligence System (Technical Report CSRP-01–05), Dept of Computer Science, The University of Birmingham). Cameron, L. (2003). Metaphor in Educational Discourse (London: Continuum). Gentner, D., Imai, M. and Boroditsky, L. (2002). ‘As time goes by: understanding time as spatial metaphor’, Language and Cognitive Processes, 17, 537–5. Kövecses, Z. (2002). Metaphor: A Student’s Introduction (New York: Oxford University Press). Kövecses, Z. (2005). Metaphor and Culture: Universality and Variation (New York: Cambridge University Press). Lakoff, G., Espenson, J. and Schwartz, A. (1991). The Master Metaphor List. (Technical Report, University of California, Berkeley). Langston, W. (2002). ‘Violating orientational metaphors slows reading’, Discourse Processes, 34, 281–310. Lonergan, J. (in prep.). ‘Mixed metaphor and conceptual metaphor theory’. Doctoral thesis in preparation (Dept. of Psychology, University of California, Santa Cruz). Martin, J. (1994). ‘Metabank: A knowledge base of metaphoric language conventions’, Computational Intelligence, 10, 134–49. Mason, Z. (2004). ‘CorMet: A computational, corpus-based conventional metaphor extraction system’, Computational Linguistics, 30, 23–44.
Commentary 261 Musolff, A. (2004). Metaphor and Political Discourse: Analogical Reasoning in Debates about Europe (Basingstoke: Palgrave Macmillan). Nayak, N. and Gibbs, R. W. Jr. (1990). ‘Conceptual knowledge in the interpretation of idioms’, Journal of Experimental Psychology: General, 119, 315–30. Pragglejaz Group (2007). ‘MIP: A method for identifying metaphorically used words in discourse’, Metaphor and Symbol, 22, 1–39.
This page intentionally left blank
Index ad-hoc category, 2 advertizing/marketing, 118, 152 Aesop, 221, 223, 231, 238 Agerri, R., 93 Aitchison, J., 12, 23 Aldridge, M., 160, 167 ambiguity, 44–6, 50, 56, 68, 72 analogy, 26–8, 30, 80, 101, 108–9, 177, 180, 182–3, 186, 198, 213, 234, 236–42, 244, 245 argument/argumentation (theory), 25, 28, 32, 44, 98–9, 102–3, 105, 107–13, 117, 183, 185–6, 193, 201–2, 206, 208–10, 212, 214–16, 228, 230, 237, 239–40, 242, 254 Aristotle, 6, 7, 60, 62, 67, 75, 99, 101, 114, 205, 222, 227, 230–1 artificial intelligence/AI, 79–90, 173 Asher, N., 41, 57 Atkins, E.M., 224, 231 Augustine, 229, 231 Aulotte, R., 211, 217 Aurelius, M., 231 Baicchi, A., 142, 147, 148, 151 Balsamo, J., 194, 203 Banks, K., 174, 123–4, 244 Barcelona, A., 32, 76 Barnden, J., 80, 84, 86, 88, 92, 93, 122, 132, 257, 258–9 Barnes, J., 230 Barr, A., 93 Barthes, R., 100–2, 112 Bass, A.M., 238, 245 Baugh, A.C., 191 Becket, T., 236–7, 245 Beer, F.A., 59, 61–3, 66 behaviour-oriented perspective, 30 Bell, A., 146, 199 Bernardez, E., 182 Bible/Genesis/New Testament, 131, 142, 149–50, 210, 241 Billig, M., 61, 155
Black, M., 5, 65 Blackmore, S., 206 Blair, A., 97, 103, 108, 111, 113, 163 blending blend(ed) space, 14, 45, 50–6, 150 Blending Theory, 40–1, 50, 56, 141–2, 144 conceptual compressions, 54 conceptual integration, 48, 54, 56, 150 conceptual network, 51, 54 cross-space mapping, 44 generic space, 50–1, 143 hyper-blend space, 54 mental/conceptual spaces, 41, 43, 52–4, 86, 142, 144–5 Bluck, S., 72 body politic, 97–9, 164, 167, 205–8, 210, 212–16, 219, 221, 223–4, 226–7, 230, 234–5 Boers, F., 20, 21 Boethius, 229 Bono, J. J., 157, 166, 173 Bourdieu, P., 99 Bovair, S., 28 Bowdle, B.F., 27, 29, 80 Brooke-Rose, C., 44 Bruner, J., 5 Bryson, B., 153, 158 Bucchi, M.,157 Bud, R., 167 Burke, P., 212, 213 Business Week, 124, 132 Cable, T., 191 Cameron, L., 25, 32, 131, 253 Cannon, G., 154 capitalism, 107, 134, 139 Carston, R., 2 Cassirer, E., 100 Castro, F., 97, 103, 105, 107, 109, 110 Chain of Being, 229–30, 235, 238 263
264 Index Charteris-Black, J., 32, 97, 100, 104–5, 118, 122, 136, 144, 155 Chiang, W., 158 Chiappe, D.L., 27 Chilton, P., 15, 32, 52, 57, 61, 65, 67, 103, 174, 177–8, 182–3 Church, W.F., 208, 216 Cicero, 219, 224–5, 231 Claidière, N., 185 Clark, A., 174, 175, 185 Clark, C.E., 208, 217 Clark, H.H., 7 Clément, L., 192, 193 Cognitive Theory cognitive-critical discourse analysis, 120–1, 130 cognitive effect, 40, 45, 51, 56, 146, 148 Cognitive Linguistics, 11, 15–16, 22, 25–6, 29–32, 41, 44, 48, 56, 60, 62, 71, 72, 135, 147, 156, 173–4, 177, 185–6, 233, 256 cognitive-psychological approach, 27, 31 Cohen, H., 154 coherence conceptual coherence, 41, 43–4, 49, 56 cultural coherence, 41 pressure of coherence, 11–12, 18–19, 21–3, 147, 257, 258 Coker, F.W., 234 Colish, M.L., 230–1 communism, 60, 106, 108 Complex Adaptive Systems (CAS), 173–8, 181, 185–6 conceptual metaphor, see metaphor context context-model, 28–9, 34–7 contextual issue, 87, 89–90, 92 co-text, 55, 127, 142 socio-cultural context, see culture corpus linguistics concordancing, 122, 124–9, 131 corpus-based analysis, 121, 126, 129, 153, 167, 258–9 corpus-search, 44, 253 purpose-built corpus, 121–5, 131, 157, 161–3 text-corpus, 243
Cooper, J.M., 63, 225 Cosmos, 211, 214, 219–20, 222–3, 225–8, 230; see also macro-/ microcosmos Costall, A., 6 Coulson, S., 40 Cowling, D., 174, 244 Croft, W., 173, 177, 178, 180, 184, 234, 245 Cropper, C.M., 121 culture cultural conceptualizations, 182, 184, 185 cultural knowledge, 55, 145, 260 cultural model, 12, 14–15, 22 intercultural/cross-cultural, 12, 14, 107 socio-cultural context, 174–6, 183–4, 252, 260 Dahl, R., 220 Dancer, S.J., 167 Dancygier, B., 72 Darwin/(Neo-)Darwinism, 174, 178 Davidson, D., 92 Dawkins, R., 178–82, 216 De Beaugrande, R., 120 De Landtsheer, C., 59, 61–4, 66, 69–73 De Sola Pool, I., 61 Deacon, T.W., 177–80, 182 Deignan, A., 42, 122, 136, 138, 234 Desan, P., 193, 196, 202 Diefendorf, B.B., 212 Dingwall, R., 157, 160 Dirven, R., 31, 120 discourse corporate discourse, 116–31 Critical Discourse Analysis, 1, 3–4, 27–37, 42–3, 120, 155, 174, 185 discourse formation, 48 discourse history, 171, 212, 233, 242–4 discourse metaphor network, 174–6, 183–4 discourse practice/genre, 42, 46, 48 macrodiscourse, 42, 46, 55–7 media/newspaper discourse, 12–14, 18, 21–3, 61, 65–6, 70–4, 135, 137, 139, 146, 153–66, 234, 255
Index 265 discourse – continued microdiscourse, 41–2, 56 (socio-) political discourse, 6, 15–16, 56, 59–60, 64–5, 70, 73, 101, 136, 205, 234, 236 scientific discourse, 2, 60, 156–7, 160, 165–7, 182, 184 Dolet, E., 196–7 domain (cross-) domain mapping, 25–7, 29–31, 34–7, 50 source domain, 13–15, 17, 20–2, 29, 32, 34, 48, 79–83, 91–2, 108, 120, 124–7, 129, 137–8, 140–1, 147, 211, 229, 243, 256–60 target domain, 12–20, 22, 29, 32, 34, 50, 79–80, 83, 92, 118, 120, 123–6, 129, 132, 137, 138, 140–1, 205, 213, 219, 221, 223–4, 258, 259 Dor, D., 142, 146, 147, 149 Döring, M., 186 Du Bellay, J., 196 Du Bartas, G.D.S., 208, 210–12, 215–16 Duerden, B., 155 Dynamical Systems Theory, 174–5; see also Complex Adaptive Systems Edelman, M., 63 ellipsis, 28 embodiment, 4, 23, 41, 48, 68, 100, 120, 180, 253 emotion, 14, 45, 48–50, 52, 63–4, 65–6, 68–70, 74, 87, 97–8, 100, 102–7, 110–13, 129–30, 216, 225, 259 emotive/emotional language, 61, 63–4, 67, 105 Emrich, C.G., 63, 69 Enright, D., 106 entailments, 5, 14, 104, 108–9, 191, 243 Estienne, H., 190, 192–7, 199, 200, 201, 202 ethos, 98–9, 101, 103, 105, 110, 112–14 Eubanks, P., 121 Evans, V., 48 evolutionary model, 184 Fairclough, N., 42, 120, 121, 155 Falkenhainer, B., 87
fascism, 60, 63–4, 116–17 Fass, D., 80 Fauconnier, G., 41, 42, 52, 55 Feigenbaum, E.A., 93 Feldman, J.E., 48 Fletcher, C.A., 27 Flood, C.G., 100 Fox, J., 130 Fox, R., 130 Fraisse, S., 211 frame, 45, 47–50, 55–7, 62, 66–7, 108, 142, 144–5, 162 Frank, R.M., 173–5, 177, 216, 244 Fuertes Olivera, P.A., 144 Gay, W.C., 64 Gagarain, M., 230 Galison, P., 40 Garrett, L., 154 Gatherer, D., 178, 182 Gaus, H., 63 Geeraerts, D., 243 Gentner, D., 27, 29, 80, 255 Gibbs, R.W. Jr, 4, 7, 26–7, 29–30, 41, 62–3, 81, 147, 199, 213, 255 Gill, C., 227 Giora, R., 81, 148 Glasbey, S., 80, 84, 86, 92–3 Glucksberg, S., 27, 81 Goatly, A., 26, 87 Goldman, S., 27 Grady, J., 11, 15, 46, 120 grammatical metaphor, see metaphor Guldin, R., 234, 239, 245 Guttenplan, S., 2, 4 Hope, T.E., 192 Hale, D.G., 205, 231, 233–4, 237, 239, 245 Halliday, M.A.K., 26, 31, 156, 157 Halmari, H., 106 Hardie, A., 122, 131 Hastings, A., 63 Hauser, H., 196 Heller, H., 193–8 Hellsten, I., 5, 60–1, 63, 174, 181 Henze, B., 183 Herbeck, D.A., 65 Herrera, H., 144, 149 Heylighen, F., 186
266
Index
Hilferty, J., 181 Hilgartner, S., 157, 166 Hirschfeld, L.A., 185 history conceptual history, 181, 216, 230, 242 cultural history, 174 discourse history, see discourse history of science, 40, 177, 179 Hitler, A., 63, 68, 108, 164 Hobbes, T., 215–16, 235, 240 Hobbs, J.R., 80 Holyoak, K.J., 80 Hornby, N., 84 Hornsby, D., 202 Hovland, C., 61 Hull, D.L., 180, 182 Humboldt, W. von, 7 humanism, 192, 194, 196–7, 213, 234–5, 238 humour/humoral/humorous, 33, 106–7, 207–11, 233, 255 icon, 44, 101 ideology, 56, 63–5, 68–72, 97–100, 102–5, 107, 112–14, 117, 121, 131, 144, 199–201, 222, 225, 234–6, 244, 252 Ilyin, M.V., 61, 65, 67 image schemas, see schema Indurkhya, B., 4 Ingold, B., 64, 74 Inwood, B., 230 Isaksson, M., 122 Ivy, R.L., 65 Joffe, H., 154, 156 John of Salisbury, 233–4, 236–45 Johnson, M., 11, 14, 23, 25, 41, 46, 59, 60, 62, 191, 199, 213, 216, 245 Jouanna, A., 206, 207 Kamp, H., 41 Kantorowicz, E.H., 206, 207, 237, 245 Kany-Turpin, J., 211 Keane, M., 80 Keller, E.F., 179, 180, 181 Kennedy, J.M., 27 Keysar, B., 81
Kieras, D., 28 Kintsch, W., 27, 28 Kittay, E.F., 66 Koch, E., 73 Koch, J., 117, 118 Koeller, W., 62 Koller, V., 32–3, 35, 120–2, 131, 142, 144 Korzybsky, A., 60 Koschorke, A., 233–4, 245 Koteyko, N., 165, 167 Kövecses, Z., 11–15, 18, 23, 25, 31, 62, 147, 152, 191, 242–3, 245, 256–8 Krasnoboka, N., 73 Kristiansen, G., 120 Lakoff, G., 2, 6, 11, 14, 15, 18, 23, 25, 31, 41, 46, 56, 59, 60, 62, 81, 164, 191, 199, 213, 216, 245, 253 Lamacz, S., 211 Langacker, R.W., 41 Langlotz, A., 81 Lansing, J.S., 175 Larson, B., 155, 156 Lascarides, A., 41 Lasswell, H.D., 60, 61 Lawrence, J., 168 Lee, M., 80, 86, 88, 93, 258, 259 Leezenberg, M., 2, 81 Leibovich, M., 164 Leudar, I., 6 Levin, S., 1, 56 Liebeschütz, H., 238, 245 Livy, 223 Lloyd, H.A., 212–14 logos, 98–9, 102, 108, 112–14 Lonergan, J., 255 Loveday, H.P., 166 Lovejoy, A.O., 235 Low, G., 25, 38 Luke, A., 61 Luke, T.W., 77 King, M.L., 101 Lycosthenes, C., 216 Maasen, S., 173, 177 Macdonnell, H., 121 Maclean, I., 213 Machiavelli, N., 89, 207
Index 267 macrocosm-microcosm, 164, 229, 233, 235, 238 Manfredi, D., 3, 5 Mann, N., 194 Mann, W., 29 Martin, J.H., 80, 101, 180, 253–4 Mason, Z., 122, 254 Matthiessen, C.M.I.M., 26, 31 Mautner, G., 118 McGlone, M.S., 3, 5 media/newspaper discourse, see discourse Melz, E.R., 80 meme, 177–84, 206, 216 mental space, see Blending Theory mercantilism, 187, 190, 192, 197, 201–2 metaphor conceptual metaphor (analysis), 2, 3, 7, 13, 46, 81, 104, 120, 142, 158, 253–4, 257, 259 conceptual metaphor theory, 2, 11, 40, 62, 80, 141 extended metaphor, 174, 181 generic metaphor, 46 grammatical metaphor, 26–8, 30–1, 33, 35–7 lexical metaphor, 26–8, 31, 34–7 metaphor identification, 253, 256 metaphor power (analysis), 65–6, 68–9, 71–4 metaphor power (index), 65, 69 metaphor power (model), 59, 65, 70, 72 metaphor thought, 30 metaphor understanding, 1–6, 79–82, 86–7, 258 metaphor variation, 11–13, 17, 20, 22–3, 257 universality of metaphor, 14 metonymy, 32, 63, 137–8, 148, 220 mission statements, 119, 121–6, 130–1 Mooy, J.J., 62, 65 Moss, L., 179–81 MRSA, 154–67 Mufwene, S.S., 177–8, 184 Murphy, G., 25
Musolff, A., 3, 5, 14, 16, 61, 63, 67–8, 93, 97, 108, 120, 122, 136, 144, 156, 167, 174–5, 178, 181, 183, 205, 213, 234, 259 myth/mythical thought, 55, 63–5, 71, 97–103, 107, 111–14, 157, 163 Narayanan, S., 80 nationalism, 117, 197 Nayak, N., 255 Nederman, C.J., 238, 243 Nelson, F., 121 Nerlich, B., 155–7, 165, 174, 181, 186 non-parallelism, 79, 81–2, 84–7, 89, 91–2, 258 Norvig, P., 89 Novick, L.R., 80 Nussbaum, M., 230 Nuyts, J., 120 O’Brien, J., 208 O’Halloran, K., 120 Oakely, 40 Orwell, G., 60 Pangle, T., 223, 230 Panther, K.-U., 32 paradigmatic, 32 Parel, A.J., 207 Pasquier, E., 209, 216 pathos, 98–9, 102, 104–5, 107, 112–14 Pellowe, C.M., 168 Perfetti, C., 27, 28 personification, 104–5, 111, 156–8, 161, 165 Piao, S.L., 123 Pinker, S., 3 Pizan, C.de., 245 Plato, 205, 222–3, 230, 235, 240 Plutarch/Pseudo-Plutarch, 211, 236, 239, 241–2 political meaning/semantics, 59–60, 62, 65 political discourse, see discourse political ideology, see ideology polysemy, 44, 49 Pörings, R., 31 Pragglejaz Group, 32, 253, 256
268 Index pragmatics (communicative/discourse) situation, 17–18, 20–2, 28–9, 33–4, 37 conversational maxim, 29 pragmatic function/dimension, 29, 120, 144, 255, 260 relevance (theory), 2, 7, 41, 81 salience, 20, 184 speech act, 29, 72, 100–1 psycholinguistic, 3–5, 30, 66, 255 purism, 190, 192–3 quantitative analysis, 59–60, 64, 121–2, 125, 130–1, 139 Quinn, N., 14–15, 23 Radden, G., 32 Rayson, P., 122 reference/referential function, 28–9, 33–7, 65, 117, 142, 144, 147, 180, 183, 214, 227, 239 Renaissance, 55, 190, 233–6, 243, 245 Reyle, U., 41 Rheinberger, H.-J., 156, 157 rhetoric, 29, 32–7, 41, 60–4, 70–4, 97–101, 103, 107–13, 149, 153, 212, 233, 235–6, 238, 242–4 Richards, I.A., 62, 138 Roberts, P., 208, 216 Roberts, W., 60 Rogers, P.S., 122 rogue state, 110–13 Rousset, J., 211, 214 Russell, S., 89 Sacks, S., 41 Saddam Hussein, 108 scenario non-physical scenario, 82 source scenario, 80, 82–8, 92–3, 242 source scenario conclusion, 86 target scenario, 80, 82–3, 85, 88 Schäffner, C., 61, 65, 67 schema conceptual schemas, 50, 117, 121, 138, 144–5, 238, 240 image schemas, 48, 57
Schoen, D.A., 4 Schofield, M., 229, 231 science cognitive science, see Cognitive Theory popular science, 154, 157 medical science/medicine, 153–4, 157, 162, 207, 212, 215, 222, 225, 233 scientific discourse, see discourse Seddon, M., 119 Sedley, D., 230 Seliger, M., 99 semantic/semantics, 29, 32–3, 35, 44–5, 47, 49, 52, 54–5, 59–62, 64, 66, 69–70, 74, 81, 90, 122–6, 128–9, 163, 141, 144–5, 148, 205, 207, 211–12, 234, 236, 244–5 Semino, E. 122, 131 semiology, 102 semiotic perspective, 30–1 Seneca, 219, 223, 225, 226 Seyssel, C de., 207–8 Shakespeare, W., 43, 44, 56, 190, 231, 234, 239, 245, 252 sonnet, 26, 40, 42–53, 55–6 Sharifian, F., 173, 182, 183 Shimko, K.L., 61, 65, 69 simile, 26–8, 30, 34, 37, 236 Simone, F., 194 Simon-Vandenbergen, A.-M., 26, 31 Skinner, Q., 215 Skorczynska, H., 122, 138 Smith, P.M., 200 socialism, 117 Soll, J., 213 Sontag, S., 97, 213, 215–17, 234–5, 240–2, 245 Sozzi, L., 194 Sperber, D., 41, 81, 182, 185 Spinnen, B., 116 Stefanowitsch, A., 122 Steels, L., 173, 175–7 Steen, G.J., 25–32, 62 Stern, J., 2, 81, 87 Stoicism, 219, 223, 226–31 Strohman, R., 181 Struve, T., 245
Index 269 style, 11–12, 60–1, 63–5, 69–74, 97, 107, 193 Suedfeld, P., 72 supra-individual, 15, 256 Sutherland, C., 63 Swales, J., 122 Sweetser, E., 72 syntagmatic, 32 Systemic Functional approach, 26, 29–31 Taran, S., 64, 71, 101 Thagard, P., 80 Thatcher, M., 106, 108 Thickett, D., 216 Thomas, G., 192–3, 202 Thompson, S., 29 Thornburg, L., 32 Thucydides, 220 Tillyard, E.M.W., 235–6, 238, 242 Tomasello, M., 3 Tsoukas, H., 65 Turner, E.A., 62 Turner, M., 41, 42, 46, 52, 55–6 ‘two cities’, 219ff Van Dijk, T.A., 27, 29, 70, 117, 120, 146 Van Oostendorp, H., 27 Velasco-Sacristán, M., 144 Vertessen, D., 62, 66, 70, 71, 73
Vilarroya, Ó., 181 Virtanen, T., 106 Voss, A., 166 Wallington, A., 93 Wallis, P., 155, 156 war Iraq War, 66, 108 Vietnam War, 108 Washer, P., 154, 156 Weber, M., 117 Weingart, P., 173, 177 Weiss, G., 115 Wertheimer, M., 120 White, M., 144, 147–9, 229 Wilkins, J.S., 178, 182 Wilks, Y., 80 Williams, G.W., 247 Williams, R., 169 Wilson, A., 133 Wilson, D., 81 Windt, T., 64, 74 Wodak, R. 42, 61, 155 Wood, N., 224 Woodman, R., 64 Worth, V., 197 Zavadil, J., 174, 205, 234 Zinken, J., 3, 61, 65, 67–8, 78, 174–5, 181, 186, 244
