Gibbs & Steen (Eds.) - Metaphor in Cognitive Linguistics PDF

METAPHOR IN COGNITIVE LINGUISTICS
AMSTERDAM STUDIES IN THE THEORY A N D

HISTORY OF LINGUISTIC SCIENCE
General Editor
E. F. KONRAD KOERNER
(University of Ottawa)
Series IV - CURRENT ISSUES IN LINGUISTIC THEORY
Advisory Editorial Board
Raimo Anttila (Los Angeles); Lyle Campbell (Christchurch, N.Z.)

John E. Joseph (Edinburgh); Manfred krifka (Austin, Tex.)
Hans-Heinrich Lieb (Berlin); Ernst Pulgram (Ann Arbor, Mich.)
E. Wyn Roberts (Vancouver, B.C.); Hans-Jrgen Sasse (Kln)
Volume I73
Raymond W. Gibbs, Jr. and Gerard J. Steen (eds)
Metaphor in Cognitive Linguistics

Selected papers from the 5th International Cognitive Linguistics Conference,
Amsterdam, July ggy
METAPHOR IN
COGNITIVE LINGUISTICS
SELECTED PAPERS FROM THE FIFTH INTERNATIONAL
COGNITIVE LINGUISTICS CONFERENCE
Amsterdam, July 1997
Edited by
RAYMOND W. GIBBS, JR.

University of California, Santa Cruz
GERARD }. STEEN
Tilburg University Free University Amsterdam
JOHN BENJAMINS PUBLISHING COMPANY

AMSTERDAM/PHILADELPHIA
The paper used in this publication meets the minimum requirements of American
National Standard for Information Sciences Permanence of Paper for Printed
Library Materials, ANSI Z39.48-1984.
Library of Congress Cataloging-in-Publication Data

International Cognitive Linguistics Conference (5th : 1997 : Amsterdam, Netherlands)
Metaphor in cognitive lingusitics : selected papers from the fifth International Cognitive Linguistics
Conference, Amsterdam, UI / edited by Raymond W. Gibbs Jr., Gerard J. Steen.
p. cm. -- (Amsterdam studies in the theory and history of linguistic science. Series IV,
Current issues in linguistic theory, ISSN 0304-0763 ; v. 175)
Includes bibliographical references and index.
I. Metaphor Congresses. 2. Cognitive grammar Congresses. 3. Language and culture Congresses. I.
Gibbs, Raymonc V. II. Steen, Gerard. III. Title. IV. Series: Amsterdam studies in the theory and
history of linguistic science. Series IV. Current issues in linguistic theory ; v. I75.
P30I.5.M48I58 1999
4I5"dc2i 99-I558
ISBN 90 272 3681 X (Eur.) / I 556I9 892 2 (US) (Hb; alk. paper) CIP
I999 - John Benjamins B.V.
No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other
means, without written permission from the publisher.
John Benjamins Publishing Co. P.O.Box 75577 I070 AN Amsterdam The Netherlands
John Benjamins North America P.O.Box 275I9 Philadelphia PA I9118-0519 USA
ACKNOWLEDGEMENTS
This book contains a selection of specially revised and refereed papers originally
presented at the Fifth International Conference in Cognitive Linguistics,
Amsterdam 1997.
The chapter by Steen was not presented at the conference, and the chapter
by Grady, Oakley, and Coulson was specially commissioned for this volume.
The editors wish to thank the following colleagues who acted as anonymous
referees in the selection and editing process: Lynne Cameron, Herb Colston,
Jennifer Hamblin, Peter Harder, Lachlan Mackenzie, Teenie Matlock, Susanne
Niemeier, Jennifer O'Brien, Elena Semino, and Wilbert Spooren. Their help has
been essential.
The editors are also grateful to Gwen Perret of Tilburg University for her
expert assistance in the preparation of the manuscript.
We would also like to thank Anke de Looper of John Benjamins for her
seeing the manuscript through the printing process.
Raymond W. Gibbs, Jr, University of California, Santa Cruz

Gerard Steen, Tilburg University and Free University Amsterdam
CONTENTS
Introduction
Gerard Steen and Raymond W. Gibbs, Jr
Kant, Blumenberg, Weinrich: Some Forgotten Contributions
to the Cognitive Theory of Metaphor
Olaf Jkel
Metaphorical Mappings in the Sense of Smell
Iraide Ibarretxe-Antufiano
When a Bodily Source Domain Becomes Prominent: The Joy of
Counting Metaphors in the Socio-Economic Domain
Frank Boers
From Linguistic to Conceptual Metaphor in Five Steps
Gerard Steen
A Typology of Motivation for Conceptual Metaphor: Correlation
Resemblance
Joseph Grady
Blending and Metaphor
Joseph Grady, Todd Oakley, and Seana Coulson
Self and Agency in Religious Discourse: Perceptual Metaphors
for Knowledge at a Marian Apparition Site
Victor Balaban
Taking Metaphor Out Of our Heads and Putting It Into the
Cultural World
Raymond W. Gibbs, Jr.
Metaphor: Does It Constitute or Reflect Cultural Models?
Zoltn Kvecses
Metaphors and Cultural Models as Profiles and Bases
Alan Cienki
Viii CONTENTS
Congruence by Degree: On the Relation between Metaphor

and Cultural Models
Michele Emanatian 205
Subject Index 219
Name Index 222
INTRODUCTION
GERARD STEEN & RAYMOND W. GIBBS, JR.

Tilburg University/Free University Amsterdam & University of California,
SantaCruz
Metaphor in cognitive linguistics is a two-way affair: it can go from linguistic

metaphor to conceptual metaphor, or from conceptual metaphor to linguistic
metaphor. For instance, cognitive linguists have used the abundant and systematic
presence of metaphors in language as a basis for postulating the existence of
conceptual metaphors, which illustrates the move from language to thought. Thus,
Lakoff & Johnson (1980:46ff.) have presented a whole series of conceptual
metaphors which capture our thinking about the nature of 'ideas':
IDEAS ARE FOOD

IDEAS ARE PEOPLE
IDEAS ARE PLANTS
IDEAS ARE PRODUCTS
IDEAS ARE COMMODITIES
IDEAS ARE RESOURCES
IDEAS ARE MONEY
IDEAS ARE CUTTING INSTRUMENTS
IDEAS ARE FASHIONS
These are conceptual metaphors to the extent that they are abstractions of the ideas
lying behind the common usage of such expressions as the following:
IDEAS ARE MONEY

Let me put in my two cents' worth. He's rich in ideas. That book is a treasure
trove of ideas. He has a wealth of ideas.
Conventional linguistic metaphors reflect pervasive conceptual metaphors and are

perhaps the best source for discovering these metaphoric schemes of thought.
However, the very fact that there are diverging conceptual metaphors for the
same concept may also be used to explain some facts about language. Thus, Gibbs
compared people's responses to idiomatic phrases, such as "John blew his stack"
or "John bit her head off," that arise from different conceptual metaphors for
2 GERARD STEEN & RAYMOND W. GIBBS, JR.
"anger," such as ANGER IS HEATED FLUID IN A CONTAINER and ANGER IS ANIMAL

BEHAVIOR. When an idiom was presented in a context that described the person's
anger in metaphorical terms that matched that idiom's underlying conceptual
metaphor, people judged the idiom to be more appropriate than they did for
similar idioms that arose from a different conceptual metaphor. This evidence
illustrates the move from conceptual metaphor to linguistic metaphor by showing
how different metaphorical ways of conceptualizing the same concept (e.g.,
"anger") give rise to different linguistic expressions related to that concept (e.g.,
"blew his stack" and "bit her head off).
The basic assumption behind this two-way traffic between language and
thought is the idea that metaphor needs a home base. This foundation for
metaphor does not lie in language as an abstract system of signs or symbols and
rules or conventions; instead, metaphor is a significant part of people's everyday
conceptual systems. Conceptual metaphor and its relation to linguistic expression
is the central area of investigation for cognitive linguists who are interested in
meta-phor.
Cognitive linguists most notably adhere to two important scientific
commitments:
(a) a commitment to seek general principles governing all aspects of human

language (the generalization commitment), and
(b) a commitment to make their accounts of human language consistent with
what is generally known about human cognition (the cognitive commitment)
(Lakoff 1990).
As a result, the cognitive linguistic study of metaphor is related to the study of

other aspects of language, such as polysemy and grammaticalization, and to
general aspects of cognition, such as categorization. This research program carries
with it the responsibility to explain exactly in what way metaphor is partly
constitutive of everyday cognition, as well as to connect these ideas about
metaphorical cognition to the description and explanation of metaphor in
language.
However, contemporary metaphor theory in cognitive linguistics appears to
rest on several questionable assumptions about the psychological processes
involved in metaphor use. Cognitive linguists are, at times, ambiguous about the
implications of their linguistic analyses of conceptual metaphor as psychological
models. For instance, a cognitive theory of metaphor is not necessarily an
adequate portrayal of what individual speakers/hearers do when they think
metaphorically or understand linguistic expressions motivated by enduring
conceptual metaphors in everyday speech and written texts. Not every person
INTRODUCTION 3
possesses the same conceptual metaphors to the same degree of detailed

elaboration as is suggested by linguistic analyses. Linguists' explorations of
conceptual metaphors and their resulting rich set of entailments are not
constrained by the limitations of ordinary human performance, one reason why
linguists are so proficient at detailing all the logical and pragmatic implications
of different metaphors (as seen in numerous chapters in this book). However,
ordinary speakers/listeners often make do with incomplete and partial
representations of linguistically and culturally shared metaphorical concepts. As
a result, there may be a social division of labour between ordinary speakers in a
specific community: every speaker may possess a partial, yet still coherent,
representation of what linguists have revealed to be a rich, complex conceptual
metaphor.
A complete conceptual metaphor may only emerge from examination of the
communication between, or across, participants in some community. This
examination yields a "supra-individual," cultural class of conceptual metaphors
(Gibbs this volume). These regularities capture part of the semantic repertoire of
the idealized native speaker (Steen 1994). But how real people correspond to the
idealized native speaker has been an important source of tension between
linguistics and psycholinguistics since the beginning of the sixties (Chomsky), and
in another respect since the beginning of this century (Saussure, Voloshinov).
A related issue concerns the total set of possible conceptual metaphors and
its relation to the individual mind. People may only have a subset of particular
metaphors for a particular target domain, but not others. For instance, someone
might have the metaphor LOVE IS A JOURNEY, but not LOVE IS A PLANT or LOVE IS
MAGIC. Such variation between people within a community may have important
effects on their experience of specific linguistic expressions as conventional or
new, easy or difficult, appropriate or inappropriate, and so on, and may influence
people's production and comprehension of specific linguistic expressions in
concomitantly varying ways.
Furthermore, individual, partial representations of cultural conceptual
metaphors may not always be pre-stored in their entirety in people's mental
lexicons and encylopedias. Parts of these conceptual metaphors may have to be
(re)constructed in different ways on different occasions. At the same time,
pre-stored conceptual metaphors may not always be activated when people
immediately comprehend metaphorical language. The metaphorical and literal
meanings of polysemous words, for example, may possibly receive independent
representations (but see Williams 1992). This may be so even though polysemous
lexical items may be related to conceptual metaphors when analyzed from a
diachronic perspective.
Some of these issues have been discussed, and various psychological

evidence in support of conceptual metaphor can be found in Katz, Cacciari, Gibbs,
and Turner (1998). This empirical work shows that conceptual metaphors may
have different effects on people's learning, conscious interpretation, and their
rapid, mostly unconscious comprehension of various kinds of conventional and
novel metaphoric language. An important claim of this psycholinguistic research
is that very different methods must be employed to assess the potential role that
conceptual metaphors may have in different aspects of language production and
understanding. Nonetheless, this work still provides strong evidence that
systematic patterns of metaphorical thought have an important role in ordinary
linguistic behavior.
Finally, children may acquire conceptual metaphors wholesale from their
learning language without necessarily having to re-experience all the cultural and
embodied events that originally gave rise to these conceptual metaphors, events
that also help keep these alive in human conceptual systems. It is not necessary for
every adult to have undergone the same set of cultural experiences motivating the
bulk of conventional conceptual metaphors for these metaphors to be a significant
part of people's personal conceptual and linguistic repertoires. Adults may have
simply learned how to use particular words in a conventionally metaphorical
fashion on suitable occasions.
We raise these alternative possibilities about the cognitive status of
conceptual metaphors because of frequent ambiguities in cognitive linguistics
about the relation between linguistic and conceptual metaphor. Simply put, there
may not be a direct mapping between linguistic metaphor and conceptual
metaphor, on the one hand, and between linguistic metaphor and individual
cognition, on the other. Cognitive linguists should be careful not to immediately
assume that the results of their systematic examination of language necessarily
implies that each individual person must have all the full-blown conceptual
metaphors uncovered by linguistic analysis.
In fact, this issue is a significant source of tension in the contemporary study
of metaphor in cognitive science. Many cognitive psychologists are skeptical
about trying to infer much about human conceptual systems from an analysis of
systematic patterns in language (Murphy 1996). These psychologists argue that the
great variability in people's introspections about their own mental processes
makes linguistic intuitions a poor choice of methods for examining issues of
mental representation (a point that some linguists now agree withsee Croft
1998). Furthermore, linguists' intuitions often differ from those of ordinary
individuals who have no pre-conceived notion about the phenomenon of interest.
As Sandra and Rice (1995) put it, whose mind are we trying to understand-the
linguist's or the language user's? For these reasons, cognitive psychologists seek
INTRODUCTION 5
to experimentally test falsifiable hypotheses using more objective methods under

controlled situations. Both of us have conducted psychological studies with this
aim in mind and believe, contrary to some psychologists, perhaps, that at least
certain elements of cognitive linguistics have much to contribute to understanding
metaphor in language, thought, and culture (Gibbs 1994; Steen 1994).
Despite these debates on the cognitive and psychological status of metaphor,
the interaction between cognitive scientists and linguists has led to a growing and
impressive body of research about metaphor in language and cognition, especially
since the publication of Lakoff and Johnson's (1980) pioneering book Metaphors
We Live By. The present collection of articles represents a part of the most recent
developments in metaphor studies currently carried out in relation to the
framework of cognitive linguistics.
There are several important trends about the study of metaphor within the
chapters that follow. Perhaps the most significant to point out is that each author's
work reflects different ways of adhering to the generalization commitment and/
or the cognitive commitment (noted above). For example, Frank Boers and Irade
Ibarretxe-Antunano report their respective detailed examinations of the
metaphorical character of specific target domains. These authors attempt through
their systematic analyses of different linguistic expressions to show how metaphor
shapes people's conceptualizations of abstract concepts. The difference between
them is also important: Ibarretxe-Antunano's contribution redresses the picture of
metaphorical extension in connection with the senses first presented in Sweetser
(1990), and can be seen as a correction of a hypothesis. Frank Boers, on the other
hand, takes the hypothesis of embodied experience underlying metaphor use as
given, and tests it in a highly specific context by means of quantitative corpus
analysis. In terms of trends, the former is a much more typical exercise in
cognitive linguistics than the latter, and we are happy that we have been able to
include both, for obviously different reasons.
Another author, Joe Grady, explores systematic analyses of conventional and
novel metaphorical expressions to discover deeper correlations in embodied
experience that form the bedrock of many aspects of metaphorical thought. His
paper is a contribution to the central concerns of the theory of conceptual
metaphor, and before long it may produce interesting hypotheses for testing. In
another paper, Joe Grady, Todd Oakley, and Seana Coulson propose that complex
mappings in both metaphorical and certain non-metaphorical language reveal how
conceptual metaphors are just part of a larger system of conceptual integration.
Theirs is another theoretical contribution, which was specially commissioned for
this volume: it reflects the new trend in cognitive linguistics which pays attention
to all kinds of conceptual blending, and can be seen as a manifestation of the
cognitive commitment mentioned above. If metaphor is just one specific case of

conceptual blending, it becomes even less special than people have long assumed.
Victor Balaban goes on to test specific hypotheses about the role of
embodiment in certain abstract concepts by conducting quantitative counts of
particular metaphors underlying certain religious discourse. His work can be
usefully compared to the study by Boers, although Balaban's method is the one of
field work rather than corpus linguistics. That field work does not preclude the use
of statistical testing is another important feature of his chapter. Moreover, his
work also points forward to the theme of the last group of chapters, in which
conceptual metaphor is closely related to the cultural context in which it is used.
From that perspective, too, it is unusual to find a statistically tested analysis of
data.
Ray Gibbs also employs ideas from cognitive linguistics to test hypotheses,
in laboratory experiments, in order to reveal aspects of how metaphorical concepts
are employed in linguistic processing. At the same time, however, he argues how
this empirical work can, and cannot, reveal important aspects of culture in mind.
His chapter is an effort to begin disentangling the knot between metaphor in
language, the individual mind, and culture. If we have emphasized that cognitive
linguistics ought to be careful about positing too direct a relation between
linguistic analysis and psychological modeling, it ought to be realized that the
situation is even further compounded when one takes the concrete social and
cultural context into consideration.
Other authors, such as Alan Cienki, Michele Emanation, and Zoltn
Kvecses, continue on this theme. They attempt through their respective analyses
of linguistic expressions to uncover larger cultural models, and to especially
demonstrate how metaphor constrains these cultural schemes of thought.
Kvecses' emphasis is relatively theoretical, addressing the relation between the
literal and nonliteral emergence of metaphorical cultural models, in a debate with
Quinn (1987). Cienki and Emanatian are relatively more empirical, in that they
describe culturally specific conceptual metaphors for specific domains. However,
their contributions also connect with more general theoretical discussions about
profile and base and about congruence.
Overall, these authors use different methods to achieve different kinds of
generalizations about language, thought, and culture. One of the future challenges
for cognitive linguistic studies of metaphor will be to explicitly acknowledge the
degree of commitment to drawing generalizations about thought and culture from
the systematic analysis of language. In this connection, Gerard Steen provides a
detailed set of steps to increase the reliability of conceptual metaphor
identification on the basis of linguistic metaphor analysis. And Olaf Jkel argues
in his chapter that certains aspects of the contemporary theory of metaphor have
INTRODUCTION 7
their roots in the work of earlier philosophers and linguists. This is a proposal that
reminds cognitive linguists that there is a wealth of ideas about the relation
between metaphor and language and metaphor and cognition readily available, and
that cognitive linguists could do worse than relate their theoretical framework to
this tradition as well as explore some of these ideas empirically.
Finally, the chapters in this book also differ in the extent to which the
empirical work reported relates to the concerns of cognitive psychologists,
philosophers, and anthropologists. Thus, the work in these chapters reflects the
growing influence of cognitive linguistic ideas and research on metaphor to
neighboring disciplines in the humanities and social sciences. As we note above,
and as is evident in several of the chapters in this volume, especially the ones on
culture, there remain real tensions between the aims and methodologies in these
differing disciplines. At the very least, however, no scholarly discipline can
capture significant theoretical generalizations about metaphor in language and
thought without paying close attention to the continuing fruits of the cognitive
linguistics tree.
References
Croft, W. 1998. "Linguistic Evidence and Mental Representations". Cognitive
Linguistics 9. 151-174.
Gibbs, R.W., Jr. This volume. "Taking Metaphor Out Of our Heads and Putting
It into the Cultural World"
1994. The Poetics of Mind: Figurative Thought, Language, and
Understanding. Cambridge: Cambridge University Press.
Katz, A., C. Cacciari, R.W. Gibbs, Jr., & M. Turner. 1998. Figurative Thought
and Language. New York: Oxford University Press.
Lakoff, G. 1990. "The Invariance Hypothesis: Is Abstract Reason Based on
Image-Schemas?" Cognitive Linguistics 1. 39-74.
, & M. Johnson. 1980. Metaphors We Live By. Chicago: University of
Chicago Press.
Murphy, G. 1996. "On Metaphoric Representations". Cognition 60. 173-204.
Quinn, N. 1987. "Convergent Evidence for a Cultural Model of American
Marriage". Cultural Models in Language and Thought, ed. by D. Holland,
& N. Quinn, 173-192. Cambridge: Cambridge University Press.
Sandra, D., & S. Rice. 1995. "Network Analyses of Prepositional Meaning:
Mirroring Whose MindThe Linguist's or the Language User's?" Cognitive
Steen, G.J. 1994. Understanding Metaphor in Literature: An Empirical Approach.
London: Longman.
Sweetser, E. 1990. From Etymology to Pragmatics: Metaphorical and Cultural

Aspects of Semantic Structure. Cambridge: Cambridge University Press.
Williams, J. 1992. "Processing Polysemous Words in Context: Evidence for
Interrelated Meanings". Journal of Psycholinguistic Research 21.193-218.
KANT, BLUMENBERG, WEINRICH
SOME FORGOTTEN CONTRIBUTIONS TO THE COGNITIVE THEORY OF
METAPHOR
OLAF JKEL
Hamburg/Halle
The contribution to metaphor theory by Lakoff & Johnson (Lakoff & Johnson
1980; Lakoff 1987, 1993) needs to be assessed by confronting their views with
some earlier approaches that they appear to have overlooked. For about threehun-
dred years now, various mostly European philosophers and linguists have been
anticipating the central tenets and findings of the cognitive theory of metaphor.
The first section of this paper presents a short overview of the extensive
"ancestry" of the cognitive approach. However, pointing to predecessors of the
cognitive theory of metaphor is neither an end in itself nor done for the sake of
historical justice alone. In particular those scholars presented in the following
sections, Kant, Blumenberg, and Weinrich, could make substantial contributions
and amendments to a cognitive theory of metaphor. The final section gives a
summary and conclusion.
1. Predecessors: An overview
We start then with an overview of predecessors whose contributions will not
be acknowledged in detail. Most of these maintain at least the ubiquity of
linguistic metaphor as well as its unidirectionality (for a comprehensive
exposition and discussion of the main tenets of the cognitive theory of metaphor,
see Jkel 1997, chapter 1, which features a summary in terms of nine hypotheses
in section 1.2.5.). In many cases, these tenets are accompanied by a claim of
"cognitive" domains or models. My inquiry into this matter yields the following
list:
John Locke (1689) Essay concerning Human Understanding
Giambattista Vico (1744) Principi di una scienza nuova
Johann Gottfried Herder (1770) Abhandlung ber den Ursprung der
Sprache
Franz Wllner (1827) Die Bedeutung der sprachlichen Casus und Modi:
Ein Versuch
10 OLAF JKEL
Johann Adam Hartung (1831) Ueber die Casus, ihre Bildung und Bedeu-
tung, in der griechischen und lateinischen Sprache
Hermann Paul (1880) Prinzipien der Sprachgeschichte
F. Max Mller (1888) Das Denken im Lichte der Sprache
Ernst Cassirer (1923) Philosophie der symbolischen Formen. Vol.I: Die
Sprache
Jos Ortega y Gasset (1925) "Las dos grandes metforas"
Karl Bhler (1934) Sprachtheorie
Jost Trier (1934) "Deutsche Bedeutungsforschung"
Benjamin Lee Whorf (1939) "The relation of habitual thought and behavior
to language"
Arnold Gehlen (1940) Der Mensch
Walter Porzig (1950) Das Wunder der Sprache
Franz Dornseiff (1955) Bezeichnungswandel unseres Wortschatzes: Ein
Blick in das Seelenleben der Sprechenden
Nelson Goodman (1968) Languages of Art
Hannah Arendt (1971) Vom Leben des Geistes. Vol.I: Das Denken
J.M. Anderson (1971) The Grammar of Case: Towards a Localistic Theory
G.A. Miller & P.N. Johnson-Laird (1976) Language and Perception
Julian Jaynes (1976) The Origin of Consciousness in the Breakdown of the
Bicameral Mind.
Every single one of these approaches would be worth of detailed exploration and
could be seen in comparison with the tenets of the cognitive theory of metaphor.
Thus it would be worthwhile, for example, to investigate the relationship between
the localist theory of grammar and Cognitive Linguistics. Localists like Anderson
(1971) and their German predecessors Wllner (1827) and Hartung (1831)
constitute a particular line of ancestors to the cognitive approach. They anticipate -
at least programmatically - crucial elements of the cognitive theory of metaphor,
as the following passage shows:1
Our perception proceeds in part by way of the senses, in part by means of

the mind. Sensual perception is always first: therefore language serves
sensual perception before it serves mental apperception. By means of the
analogy of the mental and the sensual, words are later transferred to mental
apperception. Like the poets, ordinary people refine language by means of
metaphors. Just as there is no sensual expression which could not be
transferred to mental apperception, we claim that there are no terms for
mental issues which are not taken from sensual things. Thus wherever we
have both sensual and metaphorical applications, we should without doubt
regard the former as basic (Hartung 1831:4f.).
SOME FORGOTTEN CONTRIBUTIONS 11
Probably the Americans Lakoff and Johnson, whose research is for the most part
unhistorical, are not to be reproached for ignoring the works of a German scholar
from the nineteenth century like Johann Adam Hartung. The same is likely to
apply with regard to German linguists such as Hermann Paul (1880), Karl Bhler
(1934), Jost Trier (1934), Walter Porzig (1950), and Franz Dornseiff (1955) as
well as German philosophers and anthropologists like Max Mller (1888), Ernst
Cassirer (1923), or Arnold Gehlen (1940). Things might look different, though,
with some English "classics", of which we will pick out the two best known
examples.
In his 1939 essay 'The Relation of Habitual Thought and Behavior to
Language", the American Benjamin Lee Whorf, one of the initiators of compara
tive linguistics, makes the following general claim (Whorf 1939:146 and 155):
... we can hardly refer to the simplest nonspatial situation without constant
resort to physical metaphors. [...] Our metaphorical system, by naming
nonspatial experiences after spatial ones, imputes to sounds, smells, tastes,
emotions, and thoughts qualities like the colors, luminosities, shapes,
angles, textures, and motions of spatial experience.
Though Whorf is given some credit in Lakoff s & Johnson's preface (1980:xi),
this is at the most aimed at his well known hypothesis concerning the role of
language in structuring world view, while at other times (cf. Lakoff 1987:304ff.)
he is vehemently attacked as a crass relativist. Nowhere do Lakoff and Johnson
give Whorf any credit for the hypothesis quoted above, though this could be seen
as anticipating the basic finding of their cognitive theory of metaphor.
Still worse treatment is given to another, and much older, "ancestor" of that
theory: the British philosopher John Locke is condemned by Johnson for his
rejection of metaphor as a rhetorical device unsuitable for philosophical discourse
(Johnson 1980:46; cf. Lakoff & Johnson 1980:190f.). What is overlooked is the
fact that in the passage criticized, Locke is only concerned with the artistic trope.
In the first chapter of his philosophy of language, though, as part of his Essay
concerning Human Understanding from 1689, Locke explains the central
"Cognitive Linguistic" tenet quite precisely:
It may also lead us a little towards the Original of all our Notions and
Knowledge, if we remark, how great a dependance our Words have on
common sensible Ideas; and how those, which are made use of to stand for
Actions and Notions quite removed from sense, have their rise from thence,
andfrom obvious sensible Ideas are transferred to more abstruse significa-
tions, and made to stand for Ideas that come not under the cognizance of
our senses (Locke 1689:403, with his original italics and capitalization).
12 OLAF JKEL
Put sarcastically, Locke's only failure would be not to have addressed these
"Words taken from the Operations of sensible Things, and applied to certain
Modes of Thinking" (ibid.) explicitly as conceptual metaphors.
2. KanVs concept of analogy and his "symbolical sensualisation"

The first predecessor of the cognitive theory of metaphor to be treated more
comprehensively is Immanuel Kant (1724 - 1804). In his Kritik der reinen
Vernunft (1781/87), which forms the epistemological part of his "Critical
Philosophy", Kant ascertains two roots of knowledge: conceptual understanding
and sensual intuition. Only the combination of these two constituents yields real
knowledge. The crucial point for us is that intuition (Anschauung) is a necessary
constituent of knowledge (1781/87: B 33): "All thought must, directly or
indirectly, by way of certain characteristics, relate ultimately to intuitions, and
therefore, with us, to sensuality, because in no other way can an object be given
to us." Now there are concepts without any directly corresponding sensual
intuition. Such concepts need to be "sensualised" indirectly, and according to Kant
this is the cognitive function of metaphor.
To appreciate the relevant passage from Kant's Kritik der Urteilskraft
(1790) quoted below we need to be reminded of one thing: Kant does not have a
special term metaphor, but speaks of symbols instead (for confirmation see
Blumenberg 1960:l0f. and Keil 1991:226). Thus when he explains "symbolical
sensualisation" (1790: 59), we are justified in translating "metaphorical
sensualisation". And this works as follows:
... by means of an analogy (for which empirical intuitions are made use of),
in which judgment does a twofold job: first, applying the concept to the
object of a sensual intuition, and then applying the rule for reflecting on that
intuition to a completely different object, of which the first is only the
symbol [i.e. metaphor]. Thus, a monarchic state may be conceptualized as
a living being if governed according to democratic laws, but as a mere
machine (like a hand mill) if governed by a single absolute ruler. In both
cases, though, it is conceptualized only symbolically [i.e. metaphorically].
There is no similarity between a despotic state and a hand mill, but between
the rule for reflecting on either of the two and their causality. This issue has
not been explained in detail, though it is worth of deeper investigation.
However, we cannot dwell on this here. Our language is full of such indirect
conceptualizations by means of analogy, in which the expression [...]
contains merely a symbol [i.e. metaphor] for reflection. Thus the words
Grund ['ground, reason'] (support, basis), abhngen ['depend'] (be held
from above), woraus flieen ['flow'] (follow), Substanz ['substance'] (as
Locke expresses himself: the bearer of qualities) as well as countless others
[...] are only symbolical sensualisations. They are expressions for concepts
not based on any direct intuition, but only by means of analogy with such
an intuition, i.e. the transfer of reflection on some object of intuition to a
completely different concept, may be one to which no intuition can ever
correspond directly (Kant 1790: 59, with his original italics).
Careful interpretation of this dense passage confirms that the philosopher, though
without an explicit term metaphor, is onto what two hundred years later Lakoff
and Johnson will dub conceptual metaphor. Kant speaks of analogy, construed as
"the transfer of reflection on some object of intuition to a completely different
concept, maybe one to which no intuition can ever correspond directly". This is
the equivalent of Lakoff's and Johnson's cognitive-conceptual definition of
metaphor, combined with a claim of necessity and an epistemological reason for
the unidirectionality of metaphor: concepts to which no intuition corresponds
directly are experientially grounded by means of analogical transfer. This can also
be seen in the following quote from Kant's treatise Was heit: Sich im Denken
orientieren? (1786:267): "We may fashion our concepts in the most abstract,
abstaining from sensuality as best we can, yet still they will be linked with images,
whose true purpose is to make those concepts fit for experiential use that have not
been derived from experience in the first place."
The issue can be made clearer with the help of Kant's examples. To reflect
on something as abstract (i.e. not open to direct intuition; for a discussion of
abstractness in concepts see Jkel 1997, especially section 1.3.3.) as the political
STATE we make use of various analogies that supply indirect sensualisation or
metaphorical grounding. The conceptualization of the STATE as a mere MACHINE
(like a HAND MILL) focuses on different aspects than would its personification as
a LIVING BEING. According to Kant, the first conceptualization highlights despotic
and absolute structures of a monarchic state, while the second highlights
democratic aspects.
In passing as it were, yet unmistakably, the Kantian passage ascertains the
ubiquity of metaphor in everyday language, with examples like "Substanz"
indicating the need for a diachronic and etymological approach. What seems most
commendable in the way of the theory of metaphor is Kant's statement that there
is no similarity "between a despotic state and a hand mill", i.e. "objectively"
between target domain and source domain, "but between the rule for reflecting on
either of the two and their causality": reflection by means of conceptual metaphor
only constitutes similarities in the sense of analogical relations between the
elements and their functional connections in both target domain and source
domain.
14 OLAF JKEL
This "analogy (in the qualitative sense)" (1790: 90), which appears quite
modern in our age of cognitive science, is exemplified further in various other
parts of Kant's works. Thus we conceptualize the BEAVER'S LODGE metaphorically
as HUMAN ARCHITECTURE (1790: 90), the LEGAL RELATIONSHIPS WITHIN A
COMMUNITY as MECHANICAL ATTRACTION AND REPULSION OF BODIES (1783: 58
and 1790: 90), TIME as a POINTED LINE (1781/87: B 50), or GOD'S CREATION as
a WORK OF ART (1781/87: B 655, 1783: 57, and 1790: 90). The last example in
particular stands for the finding that the whole realm of the metaphysical depends
completely on metaphorical conceptualization. Kant (1790: 59) states: "Thus all
our knowledge of god is only symbolical" (cf. the discussion of the necessity of
metaphor in Jkel 1997, section 1.2.4.).
The explanatory function of these conceptualizations is emphasized in
Kant's Prolegomena (1783: 58): "By means of such an analogy I can achieve a
relational conceptualization of things which are utterly unknown to me." In this
case, the critical philosopher, who condemns all unenlightened "dogmatic
anthropomorphism" (1783: 57), does not have any epistemological reservations:
"We ... take the liberty of a symbolical anthropomorphism" (ibid.), construing
conditions in the target domain only "as if' they were like those in the source
domain. To construe the world as if it were the work of a divine artist is utterly
permissible as long as that artist is not hypostatised; i.e. as long as this anthropo
morphic conceptualization of god remains symbolical, and in Kant this reads:
metaphorical.
To sum up: despite its conciseness, Kant's exposition of "symbolical
sensualisation" by means of analogy anticipates the most important tenets of the
cognitive theory of metaphor.2 In addition, that approach is given an explicit
epistemological grounding. That "deeper investigation" which the cognitive and
linguistic phenomenon of metaphor merits according to Kant (cf. the above quoted
passage 1790: 59) may be claimed by Lakoff and Johnson as their genuine
achievement.
Another predecessor can be treated here only as a sort of "appendage" to
Kant: in his Philosophie der symbolischen Formen (1923), the Neo-Kantian Ernst
Cassirer remodels Kant's insistence on the cognitive importance of sensual
intuition as constituent of knowledge to that of spatial intuition. This surfaces in
the "metaphorical representation of mental properties in terms of spatial
properties" (1923:150), with the ultimate motivation lying in the fundamental
bodily orientation of human beings (1923:159; cf. also ibid., p. 271f., as well as
Cassirer 1972:109-136; see the discussion of the direction of metaphorical transfer
in Jkel 1997, section 1.3.3.):
... it is our distinctive knowledge of the limbs of our body which serves as
the starting point of all further orientation in space. The distinctive image
of our own body, seen as a complete and structured organism, serves as a
sort of model for our construal of the world as a whole. Here we have our
primary level of coordination, to which we can later turn back and relate,
and from which we take the vocabulary to refer linguistically to this
progress.
3. Blumenberg's philosophical "metaphorology"

The German philosopher Hans Blumenberg (1920/1996) develops his theory
of metaphor and the method of metaphorology in two extensive essays:
"Paradigmen zu einer Metaphorologie" (1960), and "Beobachtungen an
Metaphern" (1971). Some of his later books (1979a, 1989) can be seen as
applications of this metaphorology.
Blumenberg speaks of absolute metaphors (1960:9) in cases where we are
not faced with rhetorical surplus phenomena whose expressions could be
paraphrased literally without a problem. He notes that metaphors can also be
"basic components of philosophical language, 'transfers' which cannot be brought
back to the real, into logicality" (Blumenberg 1960:9, with his original italics).
The philosopher also provides the following explanation (1960:11): "That these
metaphors are called absolute means only that they resist the terminological claim
of being dissolvable into literalness. It does not mean that one metaphor could not
be replaced or substituted or even corrected by another more precise one." It is
true that Blumenberg' s fundamental observations are concerned with the language
of philosophy fixed in writing. But we will soon see that his approach cannot be
confined to that special language, but applies to language in general.
We also find the essential idea of our Cognitive Linguistic method of
investigation in Blumenberg's exposition: "Metaphorology attempts to get at the
substructure of thought, at the subsoil, the nutrient solution of systematic
crystallization" (1960:11). This aim is achieved by way of collection and
comparative analysis of linguistic metaphors, because (1960:64): "Just comparing
metaphors ... will disclose parts of the ground structure of cognition." In that
cognitive substructure we find "orientations" (1960:13) that are "'read' from quite
elementary cognitive models [Modellvorstellungen], which in the shape of
metaphors surface in the sphere of language."
Exactly like in the cognitive theory of metaphor, linguistic metaphors are
regarded as expressions and symptoms of cognitive models, systematic structures
of thought that provide general orientation though they usually reside in the
speakers' subconscious. Thus, even for many utterances that at first glance may
appear unmetaphorical, the following applies (1960:16f.):
16 OLAF JKEL
... on closer inspection they are clearly oriented around a metaphorical

background, which we propose to call 'implicative model'. This means that
metaphors in their function discussed here do not have to emerge at all in
the sphere of linguistic expressions. But a context of utterances may
suddenly merge in a unity of sense if we can hypothetically reconstruct the
metaphorical model from which these utterances are 'read'.
In addition to such metaphorical models Blumenberg (1960:69) speaks of

"background metaphors [HintergrundmetaphorikY'\ explaining this notion as "the
implicit use of a metaphor" (1960:86). These background metaphors are mostly
equivalent to our conceptual metaphors. As these are also found in ordinary
everyday language, the project of a systematic metaphorology gains importance
outside the domain of philosophy as well:
It is not only language that thinks for us, 'standing behind us' as it were in
our world view. Still more compelling is the way we are determined by our
supply and choice of images, 'channeled' in what we will perceive and
understand. Here lies the importance of a systematic metaphorology
(1960:69).
Such a metaphorology can take linguistic metaphor "as a guiding line for
observing our everyday world" (1979b:83). It enables the socio-historical study
of those background metaphors known as cultural models in the cognitive theory
of metaphor. Their function is described by Blumenberg (1960:20) as follows:
Their content determines behavior by providing orientation. They give

structure to our world, representing the unassessable whole of reality which
can never be experienced. To the historically understanding observer they
indicate those fundamental certainties, conjectures, and evaluations which
regulated the attitudes, expectations, activities and inactivities, desires and
disappointments, interests and indifferences of an age.
Many of Blumenberg's examples look familiar, too, as they remind us of

conceptual metaphors analyzed by Cognitive Linguists as well: TRUTH as LIGHT
(1960:12f.) or as an ACTIVE PERSON (1960:14ff. passim); the WORLD as a LIVING
BEING (1960:21), as a CLOCKWORK (1960:21f. and 70ff.), as a SHIP (1960:23), as
aTHEATER (1960:21 and 1971:167f.), or as aBOOK (1960:77ff. and 1989); TIME
as SPACE (1971:166f. and 1979b:81f.), HISTORY as aSTORY (1971:166 and 168),
HISTORICAL DOCUMENTS as SOURCES ( 1 9 7 1 : 1 9 1 - 1 9 5 ) , o r LIFE as a SEA VOYAGE
(1971:171-190 and 1979a). At least implicitly, these examples also confirm the
directionality of metaphorical transfer from concrete source domains to abstract

target domains, even though an explicit unidirectionality hypothesis is missing.
The fact that Blumenberg exemplifies these metaphors primarily with
linguistic material taken from the classics of philosophy, science, and literature
should not prevent his recognition as accomplished predecessor of the cognitive
theory of metaphor. We recognize his exposition of the relationship between
linguistic metaphors and cultural models as well as the analysis of their cognitive
function. Moreover, even the focusing character of metaphor is known to
Blumenberg, who speaks of the "phenomenon of metaphorical definition of
perspective" (1960:75).
One obvious difference from the cognitive approach of Lakoff and Johnson
lies in the fact that Blumenberg' s analyses of metaphor are for the most part done
historio-philologically. Yet this is a rather peripheral difference in the approach
to the manifold of linguistic data, while there is overall and fundamental
agreement in the theory of metaphor. Moreover, the philosopher himself states
(1960:38) that to make the most of metaphorology, synchronic cross-sections are
needed to supplement his diachronic longitudinal sections.
As Cognitive Linguists and "metaphorologists" doing primarily synchronic
research we would certainly be well-advised to give more attention to the
diachronic dimension in our analyses of metaphor than do Lakoff and Johnson.3
Especially in this regard, Blumenberg's approach could provide a model for the
methodology of the cognitive theory of metaphor.
4. Weinrich's linguistic theory of metaphorical "image fields"

The German linguist Harald Weinrich (*1927) unfolds his theory of
metaphor in five essays published between 1958 and 1976: "Mnze und Wort:
Untersuchungen an einem Bildfeld" (1958), "Semantik der khnen Metapher"
(1963), "Metaphora memoriae" (1964), "Allgemeine Semantik der Metapher"
(1967), and "Streit um Metaphern" (1976b).4 Of all predecessors presented here,
Weinrich's approach is the one that comes closest to the cognitive theory of
metaphor, anticipating all its central tenets.5 Because of this fact, and because the
work of this scholar of Romance as well as Germanic languages and literature
seems to have gone almost unnoticed in the world of English linguistics, our
presentation and discussion of his approach to metaphor will be the most
comprehensive.
Right from the beginning of his theoretical discussion Weinrich displays his
conceptual understanding of metaphor, analyzing linguistic metaphors not in
isolation but situated within domains. His collection of examples (1958:278-82)
such as "Prgen von Wrtern" ('coining of phrases'), "Wortschatz" ('vocabulary')
or "Wortreichtum" ('abundance of words') makes it clear to a German reader what
18 OLAF JKEL
has to be stated of every single one of these expressions (1958:282): "... this
metaphor is not isolated. From the moment of its birth it is rooted in a firm image
field."
As will be seen in the following, the term image field (Bildfeld) is
Weinrich's equivalent to our conceptual metaphor. To account for the linguistic
examples quoted above, he formulates the image field WORD CURRENCY, where
according to Lakoff and Johnson we would have WORDS ARE COINS or LANGUAGE
AS FINANCE. The difference here is merely in notation, not in the theoretical
analysis. Weinrich's notation utilizes the tendency of the German language to
form the longest compound nouns without problems. In order to convey the
principle of his approach, my translation is literal in most cases, even if this yields
ungrammatical English compounds. In general, each of Weinrich's image fields
of the type "AB" can be translated into a conceptual metaphor of the pattern "A
is/as B", and vice versa. Thus a lot of the examples of image fields discussed by
Weinrich will strike a familiar note: LIFE JOURNEY, WORLD THEATER, LIGHT OF
REASON and MARRIAGE VEHICLE (1958:285), LOVE WAR (1963:313) as well as
WAR OF WORDS (1976b:329) and many more are among those conceptual
metaphors rediscovered by cognitive metaphor research.
Weinrich even formulates an explicit domain hypothesis (1958:283): "What
really takes place in the actual and apparently singular metaphor is the linkage of
two conceptual domains." Later he provides the following explanation (1967:326):
"... above the actual metaphor as a speech act, in our linguistic competence there
is an image field as a virtual structure. In most cases, this image field does not
need to be created, as it is known already from countless sources." Such
metaphorical domains (Sinnbezirke) are then (ibid.) - following the field semantic
tradition - identified as semantic fields: "Image fields ... can be construed as the
connection of two semantic fields." Of these two fields, one is the image donor,
the other the image recipient (1958:284), Weinrich's terminological equivalents
to the source domain and target domain of the cognitive approach (for an
overview of the terminological correspondences between Weinrich and
Lakoff/Johnson see table 1 below). According to his own words (ibid.), Weinrich
adopts these terms from Jost Trier (1934). In the example discussed above,
LANGUAGE would be the image recipient field, and FINANCE the image donor field.
The task Weinrich sets for the linguistic metaphorologist will also sound
quite familiar to Cognitive Linguistic ears - it is the systematic investigation of
image fields (1958:285f.): "Whoever wants to provide a comprehensive,
substantial metaphorology will have to list them [i.e. the image fields], supply
monographic descriptions of each, and explain how they are interrelated." And
there is the following noteworthy restriction, which in effect reinforces the
linguists' task (1958:286):
Is every metaphor rooted in an image field? That would be too much to

claim. In fact, every word can take on metaphorical meaning, every matter
can be addressed metaphorically, and imagination knows no bounds.
Arbitrary, isolated metaphors are always possible. But they are rarer than
some may think, and what is more important, usually they are not successful
in the linguistic community. The linguistic community favors integrated
metaphors, in particular (though not exclusively) for the domain of mental
experience. The metaphor that is integrated within an image field has the
best chance of being accepted by the linguistic community, and the masters
of language know this.
Again, Weinrich's exposition equals that of the cognitive theory of metaphor.

Linguistic metaphors which are isolated and idiosyncratic are a theoretical
possibility. But it can be predicted that only those expressions will prevail in the
linguistic community that come from conceptual metaphors already in existence
(cf. e.g. Lakoff & Johnson 1980:54f. passim). This point is corroborated by the
results of my empirical investigation of the understanding of metaphor (cf. Jkel
1993b and Jkel 1997, chapter 2).
The two tenets of metaphorical models and of the necessity of metaphor are
also central to Weinrich's approach. First he formulates a hypothesis of the
linguistic necessity, using the target domain of TIME as an example (1963:316):
"We cannot refer to time without speaking metaphorically." That the ultimate
reason for this linguistic finding lies in the cognitive necessity of metaphor is
stated as part of Weinrich's discussion of the target domain MEMORY: "We cannot
think of an object such as memory without recourse to metaphors. Metaphors,
especially if they appear in consistent image fields, must be estimated as
(hypothetical) cognitive models [Denkmodelle]" (Weinrich 1964:294, with my
italics; cf. ibid., pp. 291-94; see also the discussion in Weinrich et al. 1968:117
passim).
With this we have reached Weinrich's model hypothesis, and his metaphor-
based "(hypothetical) cognitive models" indisputably match Lakoff s (1987)
metaphorical ICMs exactly. In the role of cognitive models, conceptual metaphors
(alias "image fields") really determine our world view (1958:288): "Thus we see
now that our view of the world is essentially determined by our image fields,
indeed much more than by word fields." This "world making" function of
metaphor is emphasized even more strongly in the following statement, which
Weinrich makes against the comparison view of metaphor: "... our metaphors do
not, as the old theory of metaphor claimed, mirror real or imagined similarities,
but they only establish their analogies, they create their correspondences. Thus
they are demiurgical instruments" (Weinrich 1963:309, with my italics; cf. also
Weinrich et al. 1968:118f.). This argument, which is very similar to that of Max
20 OLAF JKEL
Black (1977), is a graphic confirmation of the basically constructivist view which

is one of the general characteristics of the cognitive theory of metaphor (see e.g.
Taylor 1995:4f.; cf. Jkel 1997, section 1.2.4.).
Even the tenet of the unidirectionality of metaphorical transfer is found in
the early works of Weinrich, where it concerns conceptual metaphors, or image
fields. In his argument against Aristotle's claim of a general reversibility of
metaphors, Weinrich (1963:315, with his original italics) focuses on
... image fields, which, being traditional and social products, as a rule are
unidirectional [einsinnig]. [...] Here, tradition has favored one direction of
metaphor, and with this directionality [Gerichtetheit], metaphor has
unfolded into an image field, thus becoming not just a stylistic, but a
linguistic reality.
This "linguistic reality" of conceptual metaphor is up for investigation by modern

linguistics. Weinrich's unidirectionality hypothesis is unambiguous and without
need of further comment.
What then is the status of conceptual metaphors or image fields? Weinrich's
answer to that question can be presented in three steps. First of all, most
conceptual metaphors are not a matter of the individual speaker or author. Instead,
there is "a supraindividual world of images as an objective, substantial possession
of metaphors within a community" (1958:277). Secondly, that community is not
necessarily confined to a single language (Weinrich 1958:287, with my italics):
Concrete image fields will hardly ever be common property of all mankind,
nor will they belong exclusively to one single language. They belong to the
linguistic world view of a whole culture. [...] There is a harmony of image
fields between the individual western languages. The West is an image field
community. [Das Abendland ist eine Bildfeldgemeinschaft.]
This is a point that had already been made by Bral (1900:132): "Among the old
nations of Europe there exists a common fund of Metaphor [sic] which arises from
a certain unity of culture." Thus, conceptual metaphors have the status of
"multiversals". According to Weinrich, this is the reason why linguistic metaphors
can quite often be translated without loss. And there is still a third extension of the
"dominion" of conceptual metaphors that the philologist thinks possible
(1976:335): "I cannot exclude that even between different cultures there may be
surprisingly similar image fields, which then would give voice to certain
anthropologically basic experiences of all mankind."
This is the suggestion that some conceptual metaphors, instead of being just
"multiversal", might even have the status of universals. This view is shared by the
proponents of the cognitive theory of metaphor. Probably because they do

represent "certain anthropologically basic experiences of all mankind", image
schemata, which underlie many metaphorical ICMs, are the most promising
candidates for the status of cognitive universals (see Jkel 1997, section 1.2.3.; cf.
Johnson 1992:354 as well as Lakoff 1987:302, 312, and 336).
So far we have inspected Weinrich's theoretical anticipations and parallels
to the cognitive approach to metaphor, which are quite remarkable in number and
degree of concurrence. In the following we will discuss two additional aspects in
which the cognitive theory of metaphor might well receive inspiration or
amendments from Weinrich's outline. The first point concerns the definition of
metaphor.
Right from the start in 1980, Lakoff and Johnson confine their definitional
efforts to conceptual metaphor and thus to the cognitive-conceptual level. As
regards the level of language, where lexemes are used metaphorically, they remain
quite vague (for a detailed discussion see Jkel 1997, ch. 1, in particular sections
1.2.2. and 1.3.1.). By contrast, Weinrich makes an effort to supplement his
conceptual-metaphorical image fields with a definition of metaphor on the
language level. In this he follows a text semantic approach, very rightly regarding
linguistic metaphor as a contextual phenomenon (cf. the discussion in Jkel 1997,
section 1.3.1., whose results are in complete agreement with Weinrich's theory):
A metaphor is never a simple word, but always a piece of text, however

small. And do not let orthography, that eternal temptress of linguistic
analysis, deceive you: Windrose ['compass rose'], though one word in
written German, is a piece of text in which the element Wind provides the
context for the element Rose, determining it to become metaphorical
(Weinrich 1967:319).
Thus by its use in a particular context, the original meaning of a lexeme is

converted to a metaphorical reading. Weinrich calls this process "counter-
determination", thus gaining his definition of linguistic metaphor (1967:320):
"With this concept, metaphor can be defined as a word in a counterdeterminating
context."
I do not mean to say that this attempt at a definition will do a lot. After all,
we still have to find out in every single case what the literal meaning or ordinary
determination of a lexeme is, and how exactly a particular context causes its
particular counterdetermination. Yet Weinrich's proposal of a text semantic
definition indicates an awareness of linguistic problems which the main
proponents of the cognitive theory of metaphor sometimes seem to lack (cf.
Lawler's 1983:202 review of Lakoff & Johnson 1980: "L&J [sic] do not even
22 OLAF JKEL
distinguish nouns from verbs, let alone selectional restrictions from idiomatic
usage"). After all, there is nothing to be said against Weinrich's definition of
metaphor, which is also completely compatible with the cognitive approach.
The second aspect in which Weinrich could contribute to the cognitive
theory of metaphor concerns the systematicity of method. Like Blumenberg (s.a.),
Weinrich starts his theory of metaphor with the aim of (1958:277) "developing the
outline of a methodology for metaphor research." In this connection, and
exploiting two traditional terms from semantic field analysis, Weinrich (1958:284)
draws the useful distinction between a "semasiological approach" and an
"onomasiological approach" in the systematic investigation of metaphors. Applied
to the initial example of the image field WORD CURRENCY, the semasiological
approach would start from the source domain, collating all FINANCIAL metaphors.
The onomasiological approach, on the other hand, would start from the target
domain and investigate all metaphors for issues of LANGUAGE.
These two general options, which complement one another, are also open
to the metaphor researcher in the tradition of Lakoff and Johnson. And in fact,
both approaches are applied within the framework of Cognitive Semantics.6
5. Summary and conclusion

The aim of this paper was to show that the cognitive theory of metaphor has
a very presentable line of ancestors. After an overview of some possible lines of
tradition that could only be sketched roughly, we presented more comprehensive
portraits of three of the most distinguished ancestors of the cognitive approach.
The accounts of Kant's, Blumenberg's, and Weinrich's contributions to the theory
of metaphor were meant to give an idea of the extensive similarities and
concurrences. The following table 1 provides a final overview of the central
terminological correspondences between the cognitive approach and its most
important predecessors.
Kant analogy symbol 0
Blumenberg background metaphor cognitive model 0 0

metaphor
Weinrich image field metaphor (hypothetical) image donor image recipient

cognitive model field field
Lakoff/ conceptual metaphor. cognitive model source domain target domain
|Johnson metaphor expression (ICM)
Table 1: The central terminological correspondences between the cognitive theory of

metaphor and its predecessors (0 indicates lack of corresponding term)
In all its conciseness, this tabular overview gives a graphic impression of one of
the conclusions of this paper: Weinrich's approach represents in many aspects a
European anticipation of Lakoff s and Johnson's theory of metaphor.
Maybe this will lessen the originality of the cognitive approach a little. But
certainly we gain more than we lose, as the central tenets of the cognitive theory
of metaphor are confirmed by the fact that scholars of completely different
backgrounds have reached the same or very similar results independently of each
other. Thus the epistemologist Kant hits on metaphor in the course of his critical
stocktaking of human understanding. Blumenberg, the historian of philosophy,
discovers metaphor while reconstructing the history of central philosophical and
scientific concepts. The linguist Weinrich resembles the cognitive researchers of
metaphor most closely also as regards his own heuristics, with his theory of
metaphor resulting from the philological-linguistic observation of everyday
language.
Thus it has been shown that the works of Kant, Blumenberg, and Weinrich
have more to offer to the cognitive approach than mere anticipations. Their
genuine contributions to methodology as well as to the epistemological framework
should not be ignored by a cognitive theory of metaphor that can still be amended.
Notes
1. This and all remaining quotations from German sources are given in translation by the
author. I thank Roger Bhm for drawing my attention to Wllner (1827) and Hartung
(1831).
2. In addition, the concept of schema, which is so important for the cognitive theory of
metaphor, also goes back to Kant (1781/87). See Jkel (1997), sections 1.2.3. and 9.2.
3. Cf. Jkel (1997), section 1.3.2. My case study on metaphors of science (Jkel 1996) takes
up Blumenberg's impulse in the attempt to integrate a historical-diachronical investigation
of metaphors into the Cognitive Linguistic paradigm.
4. All reprinted (in revised versions) in Weinrich (1976a), Sprache in Texten. While these
essays have been translated into French (Weinrich is professor of Romance languages and
literature), there is no English translation whatsoever (Harald Weinrich, p.c.). It would
certainly be worthwile to remedy this shortcoming for the benefit of American recipients.
5. To this it can be added that Weinrich himself is in the German tradition of linguistic content
research (Sprachinhaltforschung). Thus he adopts certain aspects of his theory of metaphor
from the field semanticist Jost Trier (1934: cf. also 1931); in addition, there are influences
by Franz Dornseiff (1955). See the meritorious exposition of this line of tradition by Liebert
(1992:83-85 and 90-93). A full-blown theory of metaphor, though, can only be acknowl
edged for Weinrich. Note that at least since his later works (cf. Weinrich 1967:325) he is
also aware of Blumenberg's approach to metaphor.
6. For an account of Cognitive Semantic research along these lines, see Jkel (1997), section
5.1. That book also features three comprehensive case studies in "onomasiological cognitive
metaphorology". See also Jkel (1993a, 1994, 1995 and 1996) for examples of this
approach.
24 OLAF JKEL
References
Anderson, John M. 1971. The Grammar of Case: Towards a Localistic Theory.
Cambridge: Cambridge University Press.
Arendt, Hannah. 1971. Vom Leben des Geistes, Voll: Das Denken [The mind's
life, Vol.1: Thinking']. Mnchen/Zrich: Piper (19892).
Black, Max. 1977. "More about Metaphor". Andrew Ortony. 1979. 19-43.
Blumenberg, Hans. 1960. "Paradigmen zu einer Metaphorologie" ['Paradigms for
a metaphorology']. Archiv fr Begriffsgeschichte, Vol.6. Rothacker, ed. by
Erich, 7-142. Bonn: Bouvier.
1971. "Beobachtungen an Metaphern" ['Observations on metaphors'].
Archiv fr Be griffs geschicnte, Vol. 15.Grnder, ed. by Karlfried. Bonn:
Bouvier. 161-214.
1979a. Schiffbruch mit Zuschauer: Paradigma einer Daseinsmetapher
['Wreckage with spectators: Paradigm of an existential metaphor']. Frankfurt
am Main: Suhrkamp.
1979b. "Ausblick auf eine Theorie der Unbegrifflichkeit" ['Perspective on
a theory of inexplicability']. Blumenberg (1979a), 75-93.
19892. Die Lesbarkeit der Welt ['The readability of the world']. Frankfurt
am Main: Suhrkamp.
Bral, Michel. 1900. Semantics: Studies in the Science of Meaning. Transl. by
Mrs. Henry Cust. London: Heinemann (Original: 1897 Essai de smantique).
Bhler, Karl. 1934. Sprachtheorie: Die Darstellungsfunktion der Sprache
['Theory of language: The symbolic function of language']. Stuttgart/New
York: Fischer UTB (1982).
Cassirer, Ernst. 1923. Philosophie der symbolischen Formen, Vol. 1 : Die Sprache
['Philosophy of symbolic forms, Vol.1: Language']. Darmstadt: Wissen
schaftliche Buchgesellschaft (1973).
1972. An Essay on Man. New Haven/London: Yale University Press.
Dornseiff, Franz. 1955. Bezeichnungswandel unseres Wortschatzes: Ein Blick in
das Seelenleben der Sprechenden ['Denotational change in our vocabulary:
Looking into the mental life of speakers'] . Lahr in Baden: Moritz
Schauenburg.
Gehlen, Arnold. 1940. Der Mensch: Seine Natur und seine Stellung in der Welt
['Man: His nature and position in the world']. Wiesbaden: Aula (198613).
Goodman, Nelson. 1968. Languages ofArt: An Approach to a Theory of Symbols.
Indianapolis: Hackett (19882).
Hartung, Johann Adam. 1831. Ueber die Casus, ihre Bildung und Bedeutung, in
der griechischen und lateinischen Sprache ['On cases, their formation and
meaning, in Greek and Latin']. Erlangen: Palm & Enke.
Herder, Johann Gottfried. 1770. Abhandlung ber den Ursprung der Sprache
[Treatise on the origin of language']. Stuttgart: Reclam (1981).
Honeck, Richard P., & Robert R. Hoffman, eds. 1980. Cognition and Figurative
Language. Hillsdale, New Jersey: Erlbaum.
Jkel, Olaf. 1993a. 'Economic Growth' versus 'Pushing up the GNP': Metaphors
of Quantity from the Economic Domain. -L.A.U.D. 1993, Series C, Paper No
24. Duisburg University.
1993b. Is Metaphor Really a One-way Street? One of the Basic Tenets of
the Cognitive Theory of Metaphor Put to the Test. C.L.E.A.R. Cognitive
Linguistics: Explorations, Applications, Research 1:1993. Hamburg
University [in: Leon G. de Stadler, ed. (1998) Issues in Cognitive Linguistics:
1993 Proceedings of the International Cognitive Linguistics Conference].
1994. "'Wirtschaftswachstum' oder 'Wir steigern das Bruttosozialprodukt:
Quantitts-Metaphern aus der konomie-Domne". Unternehmenskom-
munikation: Linguistische Analysen und Beschreibungen, ed. by Bungarten,
Theo, 84-101. Tostedt: Attikon.
1995. "The Metaphorical Concept of Mind". Language and the Cognitive
Construal of the World, ed. by John, R. Taylor, & Robert E. MacLaury, 197-
229. Berlin/New York: de Gruyter.
1996. "Metaphorical Scenarios of Science". The Construal of Space in
Language and Thought, ed. by Martin Ptz, & Ren Dirven, 649-678.
Berlin/New York: de Gruyter.
1997. Metaphern in abstrakten Diskurs-Domnen: Eine kognitiv-linguis-
tische Untersuchung anhand der Bereiche Geistesttigkeit, Wirtschaft und
Wissenschaft ['Metaphors in abstract domains of discourse: A cognitive
linguistic investigation of the domains of mental activity, economy, and
science']. Frankfurt am Main/Berlin/Bern etc.: Lang.
Jaynes, Julian. 1976. The Origin of Consciousness in the Breakdown of the
Bicameral Mind. Boston: Houghton Mifflin (19902).
Johnson, Mark. 1980. "A Philosophical Perspective on the Problems of Meta
phor". Richard P. Honeck, & Robert R. Hoffman, 1980. 47-67.
1992. "Philosophical Implications of Cognitive Semantics". Cognitive
Linguistics 3-4. 345-366.
Kant, Immanuel. 1781. Kritik der reinen Vernunft ['Critique of pure reason'].
Hamburg: Meiner (1986).
1783. Prolegomena zu einer jeden knftigen Metaphysik, die als Wissen-
schaft wird auftreten knnen ['Prolegomena to any future metaphysics'].
Hamburg: Meiner (1976).
26 OLAF JKEL
1786. Was heit: Sich im Denken orientieren? ['What is the meaning of:
Orientation in thinking?'] Werkausgabe, Vol. 5. Frankfurt am Main:
Suhrkamp (1968).
1790. Kritik der Urteilskraft ['Critique of Judgement']. Hamburg: Meiner
(1990).
Keil, Geert. 1991. Kritik des Naturalismus ['Critique of Naturalism']. Dissertation
(ms.), Hamburg University.
Lakoff, George. 1987. Women, Fire, and Dangerous Things: What Categories
Reveal about the Mind. Chicago/London: University of Chicago Press.
1993. "The Contemporary Theory of Metaphor". Andrew Ortony, ed.
19932. 202-251.
,& Mark Johnson. 1980. Metaphors We Live By. Chicago/London:
University of Chicago Press.
Lawler, John M. 1983. "Metaphors We Live By, by George Lakoff, & Mark
Johnson." Review Article. Language 59. 201-207.
Liebert, Wolf-Andreas. 1992. Metaphernbereiche der deutschen Alltagssprache:
Kognitive Linguistik und die Perspektiven einer Kognitiven Lexikographie
['Metaphorical domains of everyday German: Cognitive linguistics and the
perspectives of a cognitive lexicography']. Frankfurt am Main/Berlin/Bern
etc.: Lang.
Locke, John. 1689. An Essay concerning Human Understanding. Oxford: Oxford
University Press (1988).
Miller, George A., & Philip N. Johnson-Laird. 1976. Language and Perception.
Cambridge, Mass.: Cambridge University Press.
Mller, F. Max. 1888. Das Denken im Lichte der Sprache [ 'Thinking viewed from
a linguistic perspective']. Leipzig: Engelmann.
Ortega y Gasset, Jos. 1925. "Die beiden groen Metaphern" [The two big
metaphors']. Gesammelte Werke, Voll. Stuttgart: Deutsche Verlags-Anstalt
(1978). 249-265.
Ortony, Andrew, ed. 1979. Metaphor and Thought. Cambridge: Cambridge
University Press (1986).
, ed. 19932. Metaphor and Thought. (2nd edition, substantially revised)
Paul, Hermann. 1880. Prinzipien der Sprachgeschichte ['Principles of the history
of language']. Tbingen: Niemeyer (19759).
Porzig, Walter. 1950. Das Wunder der Sprache: Probleme, Methoden und
Ergebnisse der Sprachwissenschaft ['The miracle of language: Problems,
methods, and results of linguistics']. Tbingen/Basel: Francke UTB (19939).
Taylor, John R. 1995. "On Construing the World". John R. Taylor, & Robert E.
MacLaury, 1995. 1-21.
, & Robert E. MacLaury, eds. 1995. Language and the Cognitive Construal
of the World. Berlin and New York: de Gruyter.
Trier, Jost. 1931. Der deutsche Wortschatz im Sinnbezirk des Verstandes: Die
Geschichte eines sprachlichen Feldes [The German vocabulary in the
domain of the mind: The history of a semantic field']. Heidelberg: Winter
1934. "Deutsche Bedeutungsforschung" ['German semantic research'].
Germanische Philologie: Ergebnisse und Aufgaben. Festschrift fr Otto
Behaghel, ed. by Alfred Gtze, Wilhelm Horn, & Friedrich Maurer.
Heidelberg: Winter. 173-200.
Vico, Giovanni Battista. 1744. Prinzipien einer neuen Wissenschaft ber die
gemeinsame Natur der Vlker ['Principles of a new science of the common
nature of nations']. (2 vols.) Hamburg: Meiner (1990).
Weinrich, Harald. 1958. "Mnze und Wort: Untersuchungen an einem Bildfeld"
['Coin and word: Investigations of an image field']. Weinrich 1976a, 276-
290.
1963. "Semantik der khnen Metapher" [The semantics of bold meta
phors']. Weinrich 1976a, 295-316.
1964. "Metaphora memoriae" ['Metaphors of memory']. Harald Weinrich
1976a, 291-294.
1967. "Allgemeine Semantik der Metapher" ['A comprehensive semantics
of metaphor']. Harald Weinrich 1976a, 317-327.
1976a. Sprache in Texten ['Language in texts']. Stuttgart: Klett
1976b. "Streit um Metaphern" ['Contested metaphors']. Weinrich 1976a,
328-341.
et al. 1968. "Die Metapher (Bochumer Diskussion)". Poetica. Zeitschrift
fr Sprach- und Literaturwissenschaft 2:1968, ed. by Karl Maurer, 100-130.
Mnchen.
Whorf, Benjamin Lee. 1939. "The Relation of Habitual Thought and Behavior to
Language". Whorf 1966. 134-159.
19665. Language, Thought, and Reality, ed. by John B. Carroll. Cam
bridge, Mass.: MIT Press
Wllner, Franz. 1827. Die Bedeutung der sprachlichen Casus und Modi: Ein
Versuch ['The meaning of linguistic cases and modes: An essay']. Mnster:
Coppenrathsche Buch- und Kunsthandlung.
METAPHORICAL MAPPINGS IN THE SENSE OF SMELL
IRAIDE IBARRETXE-ANTUNANO
University of Edinburgh
1. Introduction
A fundamental principle of cognitive semantics (Johnson 1987; Lakoff
1987; Lakoff and Johnson 1980, and forthcoming) is that we have no access to a
reality independent of human categorization, and that the structure of reality as
reflected in language is a product of the human mind. The most fundamental tenet
in this model is embodiment. Human conceptual categories, the meaning of words
and sentences and the meaning of linguistic structures at any level are not a set of
universal abstract features or uninterpreted symbols but motivated1 and grounded
more or less directly in experience, in our bodily, physical and social / cultural
experiences, because after all, "we are beings of the flesh" (Johnson 1992:347).
A consequence of this primacy of general cognitive abilities is the essential role
of imagination. As Johnson (1987:172) explains, the way we reason and what we
can experience as meaningful are both based on structures of imagination that
make our experience what it is. Metaphor is a basic imaginative cognitive
mechanism. It is not a figure of speech (as it was considered by many objectivist2
approaches) but the means by which it is possible "to ground our conceptual
systems experientially and to reason in a constrained but creative fashion"
(Johnson 1992:351).
This view of metaphor as a largely automatic correspondence between
experiential domains can be applied to the study of polysemy and semantic
change. Sweetser (1990) studies the semantic changes in the field of English sense
perception verbs. She claims that the paths of semantic change are one-way and
lead from the external (socio-physical) domain to our internal (emotional,
psychological) domain and that these two domains are linked by means of
metaphor. In the case of perception verbs, the source domain is the vocabulary of
physical perception, whereas the target domain is the vocabulary of external self
and sensations. Thus, in the particular case of English perception verbs, Sweetser
establishes the following metaphorical mappings:
30 IRAIDE IBARRETXE-ANTUNANO
VISION KNOWLEDGE
HEARING HEED OBEY
TASTE LIKES / DISLIKES
TOUCH FEELINGS
SMELL 'DISLIKEABLE'3 FEELINGS
In the explanation of the structure of these metaphors of perception, Sweetser

distributes these senses into two groups: the former comprises vision and hearing
and the latter touch and taste. The focusing ability of vision and hearing, i.e., their
ability to pick up one stimulus more or less consciously is what makes them be
connected to objectivity and intellect. Subjectivity, intimacy and emotion, on the
other hand, are linked to touch and taste, due to their associated entailments of
physical contact with the thing sensed. Other authors have divided the senses in
a similar way. Viberg (1984:148), for instance, establishes a similar dichotomy:
taste and touch as opposed to hearing and smell.
Sweetser does not mention where the sense of smell should be placed in her
dichotomy and takes the view that this sense "has fewer and less deep metaphori
cal connections with the mental domain than the other senses" (1990:43). The
only two mappings she establishes are the following:
- Bad smell to indicate bad character or 'dislikeable' characteristics
(1) He is a stinker (Sweetser 1990:37)
- Detection of such characteristics
(2) I smell something fishy about this deal (Sweetser 1990:37)
With only these two abstract meanings, it can be understood that Sweetser regards
this sense as less salient than the rest of the senses. However, a closer look at the
different meanings that these verbs can convey proves this claim to be overstated:
what Kvecses (1995; in press) would call the 'metaphorical scope'4 of the sense
of smell is not weaker than that of other perception domains like sight or hearing.
2. The sense of smell: Extended meanings

One of Sweetser's main claims is that these metaphorical mappings between
different conceptual domains are not specific to one language but cross-linguistic.
In order to test this statement, I have chosen three languages from different
families, English, Spanish (both IE, Germanic and Romance respectively) and
Basque (non-IE). The linguistic material used in this study, both meanings and
examples, comes from bilingual and monolingual dictionaries (see bibliography).
METAPHORICAL MAPPINGS IN THE SENSE OF SMELL 31
The meanings discussed in this study do not represent a comprehensive typology

of all the meanings conveyed by these verbs in each language; only those
meanings5 present in all the three languages of the sample are included.
Although the sense of smell in human beings is not as developed as other
senses such as vision, there is a great number of verbs connected to the sense of
smell in one way or another. In this study, I focus only on verbs denoting neutral
perception and emission of smells. Consequently, verbs like Eng. stink and stench
(as well as their cognates in the other languages, cf. Bq. ufaztu, kiraztu; Sp.
apestar, are not included, despite the fact that in earlier stages of the
language, they were used for indicating neutral smell (see, for instance, OE stinc).
Within the physical domain of smell, it is necessary to distinguish two different
kinds of smell: the emission of odours6 and the perception of odours, and within
the latter, when the subject is an active or a passive one7. In some of the other
senses this distinction is overtly expressed by the choice of a different verb. For
example, in the sense of hearing, we have in English the verbs sound, listen and
hear respectively. As Lehrer (1990:223) points out, in the case of smell these
different kinds of smell are covered by the same polysemous lexical item smell.
Apart from these physical meanings, smell verbs can have additional senses. For
instance, smell can also mean 'to suspect' as in (3) and 'to guess, to sense
something intuitively' as in (4):
(3) Things... wouldn't always get past the sharp-eyed QC. If a case smelt, he would
smell it (OED-1973)
(4) Mary can smell trouble a mile off (OSD)
In (3), smell is used in two different ways. The meaning of the former is what
Sweetser defines as the indication of bad characteristics and it corresponds to
example (1). It can be easily replaced by the verb stink. The second smell is the
one that interests us, because it means 'to suspect'. (3) could be paraphrased as 'if
there was something wrong in the case, the QC would suspect it'. It has been
suggested to me by an anonymous referee that the second smell could be
paraphrased as 'to know' instead of 'to suspect'; although it is true that there is a
great deal of variability in the interpretation of this example, it is important to take
into account that the information that we get when we use our sense of smell is not
as reliable as that we have if we use another sense, such as vision. In (3), the QC
did not know for sure that there was something wrong with the case and that is
why the verb smell is used instead of see, in which cases the sentence would not
offer any doubt in respect to its meaning.
The meaning of (4) is rather different from (3); what it says is not that Mary
suspects that there is going to be trouble, but that in case there was, she would
sense it, she would guess it beforehand. Although sometimes guess and suspect
can be taken as synonyms, in these two examples they appear to be different.
Suspect always carries a negative meaning; if we suspect something or from
somebody, there are always negative connotations implied. This is not the case of
guess', what is guessed might be a negative or a positive thing. Its quality is not
implied by the verb itself, which by contrast only signals the fact that it is
foreseen.
Both meanings are not restricted to English, they are also possible in
Basque. In (5), smell with the meaning 'to suspect':
(5) Sailburuaren kontuakzuzenak ez zirela

erraz usain zitekeen
minister. GEN account.ABS.PL right ABS.PL neg were.COMP
easily smell could
"It was easy to suspect that the minister's accounts were not clear"
As was the case for (3), in this example the verb of smell comes to mean 'to
suspect' ; we suspected that the Minister's accounts were not clear, that there was
something wrong with them. This verb is always connected to negative connota
tions. And (6), smell with the meaning 'to guess':
(6) Kanturako haren zera ikusiz, mutrikuarra zela

usaindu nuen
song.ALL.ADN he. GEN way.ABS seeing mutriku.GEN was.COMP
smell TRANS: lSG.PAST
"From his way of singing, I guessed he was from Mutriku" (HM)
(6) might be a more illustrative example than (4) above. In (6), there are not
negative connotations or bad characteristics to be discovered, but only the fact that
this person was from that particular place called Mutriku. In his particular way of
singing, there were some hints that make us guess where he was coming from.
And finally, these same meanings in Spanish, 'to suspect' in (7) and 'to
guess' in (8):
(7) Me huele que ella est detrs de todo sto

refl.l SG smells that she is behind of all this
"I suspect she is behind all this" (OSD)
As in the previous examples in English (3) and Basque (5), there are negative
connotations implied in (7). These negative characteristic are not present in (8)
below:
(8) Juan ya se ha olido la broma

John already refl.3SG has smelt the joke
"(I think) John might have guessed that it's a joke" (RCD)
T o guess' and 'to suspect' are not the only two possible extensions in the domain
of smell. Smell verbs can also mean 'to trail something' as in (9) and 'to
investigate' as in (10).
(9) The dog was sniffing the ground looking for the hare
(10) The police have been sniffing around here again (RCD)
In (9), the meaning of the verb of smell is still physical, whereas in (10), it is
abstract. In (9), the dog was actually physically smelling the ground and following
the trail (i.e. smell) left by the hare. On the other hand, in (10), the police are not
using their noses to physically smell; although the same kind of action as in (9) is
implied, in this latter example, it should be understood in a different manner, not
in a physical but in a metaphorical way.
Once again, as predicted by Sweetser, these meanings are not specific to
English but also possible in Basque 'to trail something' in (11) and 'to investigate'
in (12):
(11) Txakurra usnaka zebilen erbiaren bila

dog.ABS smelling was hare.GEN search.ABS
"The dog was sniffing around looking for the hare"
(12) Bere gauzetan usnaka ibili ondoren, bera

hiltzailea izan zitekeela usaindu nuen
he.GEN things.INE smelling be after he.ABS
murderer.ABS be could.COMP smell TRANS:lSG.PAST
"After I sniffed around, I suspected he could be the murderer"
And also in Spanish, 'to trail something' in (13) and 'to investigate' in (14):
(13) El perro estaba olfateando el suelo en busca de

la liebre
the dog was smelling the ground in search of
the hare
"The dog was sniffing the ground looking for the hare"
(14) Le han ordenado que husmee las

cuentas
he.DAT have.3rd.PL ordered that sniff:3rd.SG.SUBJ the
accounts
"They have ordered him to investigate the office accounts" (HM)
From the above discussion, it can be concluded that Sweetser's claim that the
verbs of smell are connected to only two types of perceptual development is not
correct. Not only is the metaphorical scope of these verbs larger - meanings such
as 'to guess', 'to suspect' and 'to investigate' are possible- but also some of these
extensions of meaning remain physical like 'to trail something'. Sweetser's
analysis offers us an explanation of how concrete meanings map onto abstract
metaphorical meanings, but she does not observe that some semantic extensions
are not abstract but remain physical, and therefore, cannot be accounted for by
means of metaphor. Another point that remains unanswered is why certain source
domains - sense of smell - get mapped onto certain target domains - 'to suspect',
'to guess', 'to investigate' ; the reason for saying Mary smelt danger and not Mary
touched danger.
In this chapter, I claim that the solution for these shortcomings is to be found
in 'Property Selection Processes' (PSPs), i.e. the selection in the target domain
of only some of those prototypical properties that characterize the physical source
domain. PSPs also represent a formalization of the metonymicai character of
metaphorical mappings, the so-called "used" part of metaphor (Lakoff and
Johnson 1980:52; Johnson 1987:106), the fact that only part of the structure of the
source domain is projected onto the target domain.
3. The sense of smell: Property selection process

In the introduction, I presented as the most central tenet of cognitive
semantics the idea of embodiment, i.e. how meaning is grounded in the nature of
our bodies and perception, in our interaction with the physical, social, and cultural
environment that surrounds us. The fact that concepts are grounded in our bodily
experience and then elaborated by structures of imagination - metaphor - implies
that if we are able to characterize the domain of experience that constitutes the
source domain, then it will be possible to constrain the semantic extensions that
occur in the corresponding target domain. In other words, the reason why it is
possible to use these verbs of smell to express other meanings - apart from the
physical sense perception- must lie in the way we perceive and experience the
sense of smell. Therefore, if we can provide an account of those properties that
characterize the physical sense of smell (the source domain), then we can offer a
motivated explanation of why these extensions of meaning are possible in this
sense. These properties are to be considered as constraints for the metaphorical
mappings that can take place in this conceptual domain. What follows is a
typology of the properties that characterize smell perception. This type of analysis
should not be confused with Componential Analysis (Katz & Fodor 1963; Katz
& Postal 1964; Katz 1972). The properties listed below are not to be understood
as semantic primitives, i.e. smallest basic components of meaning that are part of
our psychological architecture. These properties are not considered to be innate
atomic conceptual units that when combined differently form the meanings of
different words, but as shorthand ways of referring to the defining properties used
to describe how we perceive through the sense of smell, which is the bodily basis
for the physical prototypical meaning of smell verbs.
3.1 Properties of the sense of smell

One of the basic properties of the sense of smell is that it is <internal>. In
order to perceive smells, it is necessary to inhale air into our nostrils, from the
outside to our inside. When we breathe, we usually inhale air into our nostrils, we
take a breath and let the air come inside us, towards our lungs. Each breath passes
air over our olfactory sites; when we inhale or exhale, we smell odours, these
odours enter our bodies (Sekuler & Blake 1994).
Unless we suffer from any kind of olfactory disorder or eventually, if we
just close our nose, we are smelling all the time since we breathe all the time, but
unconsciously. Studies (Badia 1995) indicate that we smell a wide variety of
odours throughout every day and night of our lives, but without being aware of
them at all. Only when a smell pleases, annoys, warns or brings a memory do we
stop to take notice of it. Thus, this property we can represent as <voluntary no>. A
further explanation should be made in this case because - as pointed out before -
apart from the emission of smells which we do not study in this chapter, two kinds
of smell perception must be distinguished: perception with an active subject and
perception with a passive subject. These two types of perception are characterized
differently in respect to this property. Thus, in the case of perception with a
passive subject, smell is indeed involuntary, as we are not consciously controlling
it; so, we have <voluntary no>. Whereas in the perception with an active subject,
the smell is voluntary and the property is <voluntary yes>.
As we have said before, we are smelling all the time but we only become
aware of it either if we lose our olfactory faculty or if we detect a new, good or
bad, smell. For instance, if we are in a room without any particular smell and a
person starts preparing some coffee, we immediately smell the new odour, we
detect that new smell, which later we recognize as coffee. After a while, we
become used to the smell of coffee and no longer smell it consciously. But if
somebody enters the room, that person will detect the smell of coffee straight
away. This well-documented phenomenon (cf. Ahlstrom et al. 1986) is called
odour adaptation, i.e. the decrease of sensitivity to an odour after a prolonged
exposure to it, and some people believe it is due to the unique capability of
olfactory cells to die and reproduce themselves. So, another property is <detec-
tion>.
Another characteristic is that one is very rarely sure of what is smelling.
That is to say, smells are difficult to identify immediately. When we use the sense
of vision, for instance, if we see a dog, unless we have sight problems or we have
never seen a dog before, we immediately recognize that entity as a dog. This does
not happen with smell. The reason why we are never a hundred per cent sure that
what we are smelling is one specific thing or another, lies in the fact that olfactory
fibres individually can detect that some odorous substance is present, but they are
unable to provide unequivocal information about the identity of that substance;
consequently, people can smell an odour, but cannot tell what odour they are
smelling (Engen 1960). Furthermore, smells are difficult to name. Aristotle
already pointed out the fact that the sense of smell lacks an independent
classification similar to that of other senses such as taste (sweet, bitter...), and in
fact, the situation nowadays has not changed. There have been various attempts8
such as Henning's (1916) Smell Prism and Schiffman's (1974) Multidimensional
Scaling, but as Buck (1949:1024) remarks, "the only widespread popular
distinction is that of pleasant and unpleasant smells- good and bad smells [... ] this
is linguistically more important than any similar distinction, that is, of good and
bad, in the case of the other senses". Otherwise, the terms used for defining a
smell are taken either from other senses, primarily from taste (cf. sweet) and touch
(cf. pungent, originally 'pricking') or by naming the object that emits the smell,
as the smell of an apple. So, we have the property represented by <identifica
tion no>.
Smells are different for people: what for one person is a nice smell could be
bad or simply neutral for another. Smell is also cultural (Classen et al. 1994). Our
reactions to smell also vary depending on our personal and unique odour/memory
association. Smells are context-dependent, that is to say, the same substance can
be perceived in different ways depending on the smells that may be in the same
environment; a property widely used in the art of cuisine. How familiar a person
is in respect to a smell is also important. Experiments (Cain 1982) show that
people can become familiar, can learn smells and as a consequence, can easily
identify them. That is why it can be said that smell is <subjective>.
The connection between smell and memory is very strong. Herz (1995) has
found that memories evoked by the sense of smell are more emotional than those
evoked by other senses, including sight, hearing and touch. This seems to be due
to the connection between the olfactory and the limbic systems, the latter involved
in emotional responses. So another property is <emotional>.
All these characteristics are present in physical smell perception. If we
accept that semantic changes take place from the concrete domain to an abstract
domain, it can be said that these characteristics are the first properties that the
sense of smell had, before extending its meanings to a wider scope; therefore, we
will call these characteristics 'prototypical properties' as they are the properties
of the first prototypical meaning of the sense of smell9.
The prototypical properties for smell that we have identified so far are:
<internal>
<voluntary yes> / <voluntary no>
<detection>
<identification no>
<subjective>
<emotional>
3.2 Distribution of properties in the target domain:

Property Selection Processes.
In the previous section, I have presented a set of six properties that
characterize the sense of smell. As has been noted before, metaphorical mappings
are grounded in the body and in every-day experience, therefore it follows that the
metaphorical mappings in smell verbs must be grounded in these properties,
because they characterize how we perceive through the sense of smell (the source
domain). The independence of description of the source domain is guaranteed by
the fact that these properties are based on psychological and physiological studies
on the sense of smell10. They are not simply the result of a post hoc analysis of the
metaphorical mappings found in this domain (see Keysar and Bly 1995; Murphy
1996, for a discussion of this issue).
It is commonly agreed among metaphor researchers that not everything from
the source domain gets mapped onto the target domain. Lakoff and Johnson
(1980:52) call this partial map of the structure of the source domain the 'used' part
of metaphor. In the case of smell perception, which is characterized by the set of
properties defined above, this statement means that not all the properties are
'used' in the target domain but only a selection of them. However, in order to
constrain metaphorical mappings it is not enough to say that there is a selection

of the source domain, it is necessary to show exactly what it is that is partially
mapped and what constraints are applied to that selection. Attempts to constrain
the mapping process in metaphorical production and comprehension can be found
in Lakoff's (1990; 1993) 'Invariance Principle'11, i.e. "metaphorical mappings
preserve the cognitive topology of the source domain in a way consistent with the
inherent structure of the target domain" (1993: 215). The Invariance Principle is
useful in order to constrain the nature of those mappings: that is to say, it is not
possible to map from the source domain structure that does not preserve the
inherent structure of the target domain. The only problem with this principle is
that it does not show exactly what part of the source domain is the one that must
be consistent with the structure of the target domain.
As a solution, I propose the processes called 'Property Selection'. These
processes will show not only how some of the set of properties that characterize
the source domain are mapped onto the target domain, but also what properties are
mapped. It is precisely by this selection of properties from the source domain in
the target domain that metaphorical mappings are constrained. The properties
selected in the target domain must be part of the properties identified in the source
domain and no others. The number of properties from the source domain
preserved in the extended meanings is not an issue, as this is not the same in each
extension (see Ibarretxe-Antunano forthcoming). What is important is the fact that
there is a transfer of only some properties from the source to the target domain.
As these processes take place prior to metaphor, they can also account for
semantic extensions that remain physical, as was the case of 'to trail something'.
For example, in the case of the meanings 'to trail something' and 'to investigate'
illustrated below in (15) and (16), the properties selected are <detection> and
<voluntary yes>.
(15) The dog was sniffing the ground looking for the hare
(16) The police have been sniffing around here again (RCD)
These meanings select the property <detection> because the dog in (15) and the
police in (16) are trying to detect those hints that would lead them to find what
they are looking for; the property <voluntary yes> is selected because this search
is carried out consciously, both the dog and the police are active subjects of the
action of smell. The only difference in these two sentences is that in (15) the
action of smell is a physical one, where the dog is actually using its nose in order
to follow the trail left by the hare, whereas in (16), the police are not smelling
physically, but in a metaphorical way.
Therefore, my hypothesis is that from the original prototypical meaning, i.e.

the six properties, through a property selection process, there is a choice of
properties; this process occurs in both examples as shown in Figure 1 for (15) 'to
trail something' and in Figure 2 for (16) 'to investigate'.
Prototypical Property Selected

physical physical
meanings meaning
< internal <voluntary yes>

<voluntary yes>/ . <detection>
<vomnrary no >
<detection>
<identificationno>
<subjective>
<emotional>
Figure 1: Property selection processes in (15) 'to trail something'.
However, in the case of (16) 'to investigate', a further step takes place: that of
metaphor; and that is why the meaning is no longer concrete but abstract.
It is important to bear in mind that the extended metaphorical meaning 'to
investigate' comes from the first prototypical meaning 'to perceive by smell' and
not from the extended physical meaning 'to trail something'. Otherwise it will be
implied that every metaphorical meaning needs to have a physical counterpart.
This is not true. For example, another of the metaphorical extensions discussed
in Section 2 - 'to suspect'- does not have an extended physical meaning
counterpart. Nevertheless, it can be accounted for by these property selection
processes. The only difference lies in the properties selected for these meanings.
The properties that can explain the extended meaning 'to suspect' are <voluntary
no> and <detection>. We do not consciously look for hints that would lead us to
form a suspicion, as was the case in 'to investigate' ; we detect that something
happens, but we are passive perceivers of those hints that lead us to suspect. And
<identification no> also works like this, because when we suspect something, all
we know is that something is going on but we cannot tell for sure whether what
we suspect is true or not.
40 IRAIDE IBARRETXE-ANTUANO
Prototypical Property Selected Property Selected,

physical physical metaphorical
meanings meaning abstract
meaning
<internal <voluntary yes> <voluntary yes>

<voluntary yes>/ <detection> . <detection>
<volunrary no >
<detection>
<identificationno>
<subjective>
<emotional>
Property Selection
Metaphor
Figure 2: Property selection and metaphorical processes in (16) 'to investigate'.
One of the aims of the Property Selection model proposed in this chapter is to
show which aspects of the source domain are mapped onto the target domain.
There are many different theories and experimental studies that have explored
what gets mapped in metaphor understanding (see Gibbs 1994: ch.5). Models such
as the 'salience imbalance' (Ortony 1979) model, the 'domains interaction'
(Tourangeau and Sternberg 1981; 1982) model, the 'structure mapping' model
(Gentner 1983; Gentner and Clements 1988) and the 'class inclusion' model
(Glucksberg and Keysar 1990) have put forward constraints on the nature of the
interaction between source and target domains in metaphorical expressions. These
models assume that the process of metaphor understanding depends on some
novel act of mapping information from a source domain to a target domain. This
assumption is a major difference with the model proposed here: our understanding
of metaphor is inherently constrained by our conceptualization of experience.
Despite these theoretical differences, the ideas presented by these other theories
as applied to the Property Selection model may yield interesting results, but must
be reserved for further research.
4. Conclusions
In this chapter, I have shown that the metaphorical scope of the sense of
smell is not weaker than that of other perception domains like hearing or vision.
Apart from the two mappings proposed by Sweetser, namely the indication and
detection of dislikeable characteristics, I identify other metaphorical extensions
like 'to guess', 'to suspect' and 'to investigate', as well as physical semantic
extensions like 'to trail something'. These meaning extensions are not restricted
to one language but common in the three languages examined, namely English,
Basque and Spanish, which corroborates Sweetser's claim for the cross-linguistic
nature of such semantic extensions.
Although I agree with the usage of metaphor as a primary cognitive function
by which the structure of human experience and understanding is created and
extended (Johnson 1987), these metaphorical mappings proposed by Sweetser
represent neither an explanation for the reasons why these particular domains
(source and target) are linked; nor an exhaustive account of all the possible
extended meanings that these verbs can convey.
As a solution I propose the 'property selection process', i.e. the selection of
some properties from the set of prototypical properties that characterize the sense
of smell. This set of properties is drawn from the physical experience that human
beings have when they perceive through this sense and constitutes the bodily basis
that grounds these metaphorical mappings. This selection of properties occurs in
all the extended meanings of these verbs and it is only in those cases when the
meaning is abstract that metaphorical processes take place.
Although in this chapter, these 'property selection processes' have only been
applied to the sense of smell, they can be expanded to other sense perceptions as
well as to other semantic fields. For instance, the reason why the sense of vision
is linked to the objective side of our mental life (Sweetser 1990:37) lies in the very
nature of visual perception which is characterized by an accurate and reliable
reception and manipulation of data. Therefore, instead of having <identification
no> as in the case of smell, this property would be<identificationyes> in the case
of vision.
In short, 'property selection processes' do not only explain physical and
metaphorical semantic extensions but they are also a formalization of what Lakoff
and Johnson (1980) called the 'used' part of metaphor.
Acknowledgment
I would like to thank Jon Altuna, Ronnie Cann, Raymond Gibbs, Susanne Schle, Gerard
Steen, and the anonymous reviewers for their comments on earlier versions of this chapter.
Preparation for this article was partly supported by Grant BFl98.71 from the Basque Country
Government's Department of Education, Universities and Research. The author can be contacted
at iraide@ling.ed.ac.uk.
Notes
1. This idea contradicts the traditional Saussurian principle of the arbitrariness of the sign.
2. The term 'objectivism' is used by Lakoff (1987) and Johnson (1987) to refer to those
theories of linguistic meaning that understand objective reality as independent from human
cognition, such as Frege (1952), Montague's model-theoretical semantics (Dowty, Wall
& Peters 1981; Cann 1993) and Barwise and Perry's (1983) situation semantics.
3. After Sweetser (1990:37).
4. "The range of the application of particular source domains to particular target domains"
(Kvecses 1995:316).
5. The main aim of this study is not to show how often or salient the meanings presented here
are in each language, but just the fact that they are possible to infer; therefore, I have not
included any data on frequencies.
6. In this chapter, I focus only on the semantic extensions resulting from the perception of
smells.
7. For more information on the classification of the physical meanings of perception verbs,
see Leech (1971), Rogers (1971; 1972), Kryk (1979), Viberg (1984) among others.
8. For a complete discussion on the topic of the classification of odours, see Sekuler and
Blake (1994:414-418). Gamble (1921) is a good review and critique on Henning' s method.
9. All these ideas about prototypicality are based on Eleanor Rosch's (1978) work on
categorization and prototypes.
10. See Ibarretxe-Antunano (in preparation) for a more detailed account of these properties in
perception verbs.
11. See also Lakoff and Turner (1989:82), Brugman (1990), Turner (1987:143-148; 1990;
1991:172-182; 1996)
References
Ahlstrom, R., R. Berglund, U. Berglund, T. Lindvall, & A. Wennberg. 1986.
"Impaired Odour Perception in Tank Cleaners". Scandinavian Journal of
Work and Environmental Health 12. 574-581.
Badia, P. 1995. "Olfactory Sensitivity in Sleep and the Effects of Fragrances on
the Quality of Sleep". Compendium of Olfactory Reseach 1982 -1994, ed.
by A. Gilbert. New York: Kendall and Hunt Publishers.
Barwise, J., & J. Perry. 1983. Situations and Attitudes. Cambridge, Mass.: MIT
Press.
Brugman, C. 1990. "What is the Invariance Hypothesis?" Cognitive Linguistics
1:2.257-266.
Buck, C D . 1949. A Dictionary of Selected Synonyms in the Principal Indo-
European Languages. Chicago: Chicago University Press.
Cain, W.S. 1982. "Odour Identification by Males andFemales: Predictions versus
Performance". Chemical Senses 7. 129-142.
Cann, R. 1993. Formal Semantics. Cambridge: Cambridge University Press.
Classen, C , D. Howes, & A. Synnott. 1994. Aroma: The Cultural History of
Smell. London and New York: Routledge.
Diccionario 3000 Hiztegia. Adorez 7. 1996. Bilbo: Bostak Bat. (HM)
Dowty, D. R., R.E. Wall, & S. Peters. 1981. Introduction to Montague Semantics.
Dordrecht: D. Reidel.
Engen, T. 1960. "Effects of Practice and Instruction on Olfactory Thresholds".
Perceptual and Motor Skills 10. 195-198.
Frege, G. 1952. "On Sense and Reference". Translations from the Philosophical
Writings of Gottlob Frege, ed. by P. Geach, & M. Black, 56-78. Oxford:
Basil Blackwell.
Gamble, E.A.M.C. 1921. "Review of Der Geruch by Hans Henning". American
Journal of Psychology 32. 290-295.
Gentner, D. 1983. "Structure-Mapping: A Theoretical Framework for Analogy.
Cognitive Science 7. 155-170.
, & c . Clements. 1988. "Evidence for Relational Selectivity in the
Interpretation of Analogy and Metaphor". The Psychology of Learning and
Motivation, ed. by G. Bower, vol. 22, 307-358. Orlando, Fla: Academic
Press.
Gibbs, R. 1994. The Poetics of the Mind: Figurative Thought, Language, and
Glucksberg, S., & B. Keysar. 1990. "Understanding Metaphorical Comparisons:
Beyond Similarity". Psychological Review 97. 3-18.
Herz, R. 1995. "The Emotional Distinctiveness of Odour-Evoked Memories".
Compendium of Olfactory Reseach 1982 - 1994 ed. by A. Gilbert. New
York: Kendall / Hunt Publishers.
Henning, H. 1916. Der Geruch. Leizpig: Barth.
Ibarretxe-Antunano, I. Forthcoming. "An Inside Look at the Semantic Extensions
in Tactile Verbs". Proceedings of the 26th Congress of the Spanish Applied
Linguistics Association.
In Preparation. Semantic Change and Polysemy in Perception Verbs: A
Cross-Linguistic Study (Working Title). PhD Thesis. University of
Edinburgh.
Johnson, M. 1987. The Body in the Mind: The Bodily Basis of Meaning, Reason
and Imagination. Chicago: Chicago University Press.
1992. "Philosophical Implications of Cognitive Semantics". Cognitive
Linguistics 3:4. 345-366.
Katz, J.J. 1972. Semantic Theory. New York: Harper and Row.
, & J. Fodor. 1963. "The Structure of Semantic Theory". Language 39.170-
210.
, & P.M. Postal. 1964. An Integrated Theory of Linguistic Descriptions.
Cambridge, Mass.: MIT Press.
Keysar, B., & B. Bly. 1995. "Intuitions of the Transparency of Idioms: Can One
Keep a Secret by Spilling the Beans?" Journal of Memory and Language
34. 89-109.
Kvecses, Z. 1995. "American Friendship and the Scope of Metaphor". Cognitive
Linguistics 6. 4, 315-346.
In Press. "The Scope of Metaphor". Metaphor and Metonymy at the
Crossroads: Cognitive Approaches, (Provisional Title) ed. by A. Barcelona-
Sanchez. Berlin and New York: Mouton de Gruyter.
Kryk, B. 1979. "Some Remarks on the Verbs of Perception in English and
Polish". Papers and Studies in Contrastive Linguistics, ed. by J. Fisiak, vol.
VIII, 113-231. Poznan and Arlington, Va.: Center for Applied Linguistics.
Lakoff, G. 1987. Women, Fire and Dangerous Things: What Categories Reveal
about the Mind. Chicago and London: Chicago University Press.
1990. "The Invariance Hypothesis: Is Abstract Reason Based on Image
schemas?" Cognitive Linguistics 1:1. 39-74.
1993. "The Contemporary Theory of Metaphor". Metaphor and Thought,
ed. by A. Ortony, 202-251. Cambridge: Cambridge University Press.
, & M . Johnson. 1980. Metaphors We Live by. Chicago and London:
Chicago University Press.
, & M. Johnson. 1999. Philosophy in the Flesh. New York: Basic Books.
, & M Turner. 1989. More Than Cool Reason: A Field Guide to Poetic
Metaphor. Chicago: Chicago University Press.
Leech, G. 1971. Meaning and the English Verb. London: Longman.
Lehrer, A. 1990. "Polysemy, Conventionality, and the Structure of the Lexicon".
Cognitive Linguistics 1:2. 207-246.
Murphy, G.L. 1996. "On Metaphoric Representation". Cognition 60. 173-204.
Ortony, A. 1979. "Beyond Literal Similarity". Psychological Review 86.161-180.
Oxford English Dictionary. 1986. Oxford: Oxford University Press. (OED).
Oxford Spanish Dictionary. 1994. Oxford, New York, Madrid: Oxford University
Press (OSP).
Richmond Compact Dictionary. 1995. Madrid: Santillana (RCD).
Rogers, A. 1971. "Three Kinds of Physical Perception Verbs". CLS 7. 206-223.
1972. "Another Look at Flip Perception Verbs". CLS 8. 303-316.
Rosch, E. 1978. "Principles of Categorization". Cognition and categorization, ed.
by E. Rosch, & B. B. Lloyd, 1-49. Hillsdale, N.J.: Erlbaum.
Schiffman, S. 1974. "Physiochemical Correlates of Olfactory Quality". Science
185.112-117.
Sekuler, R., & R. Blake. 1994. Perception. New York: McGraw-Hill.
Sweetser, E. 1990. From Etymology to Pragmatics. Cambridge: Cambridge
University Press.
Tourangeau, R., & R. Sternberg. 1981. "Aptness in Metaphor". Cognitive

Psychology 13. 27-55.
, & R. Sternberg. 1982. "Understanding and Appreciating Metaphors".
Cognition 11. 203-244.
Turner, M. 1987. Death is the Mother of Beauty: Mind, Metaphor, Criticism.
Chicago: Chicago University Press.
1990. "Aspects of the Invariance Hypothesis". Cognitive Linguistics 1:2.
247-255.
1991. Reading Minds: The Study of English in the Age of Cognitive
Science. Princeton: Princeton University Press.
1996. The Literary Mind. Oxford: Oxford University Press.
Viberg, A. 1984. "The Verbs of Perception: A typological Study". Explanations
for Language Universals, ed. by B. Butterworth, B. Comrie, & O. Dahl.
123-162. Berlin: Mouton de Gruyter.
WHEN A BODILY SOURCE DOMAIN BECOMES PROMINENT
THE JOY OF COUNTING METAPHORS IN THE SOCIO-ECONOMIC DOMAIN
FRANK BOERS
Universit Libre de Bruxelles
1. Introduction
According to Cognitive Semantics (Johnson 1987; Lakoff 1987) abstract
reasoning depends largely on the use of conceptual metaphors. These conceptual
metaphors can roughly be divided into two categories.
The first category maps image schemas (CONTAINER, PATH, UP-DOWN, etc.)
onto abstract experiential domains: THE BODY IS A CONTAINER FOR THE EMOTIONS,
e.g. She was filled with hatred; TIME IS A PATH, e.g. Leave the past behind you;
GOOD IS UP, e.g. Living up to high expectations; etc. One single image schema can
be used to structure various abstract domains. The UP-DOWN schema, for example,
is commonly used to conceive of abstract quantities (MORE IS UP, e.g., An IQ of
over 150), social hierarchy (HIGH STATUS IS UP, e.g., Climbing the academic
ladder), mental states (HAPPY IS UP, e.g., Cheer up!), and so on. Conversely, one
single experiential domain can be conceived in terms of different image schemas.
Time, for instance, is commonly conceived as aPATH(e.g., Leave the past behind
you) or as an OBJECT (that moves, e.g., Time flies, or that is valuable, e.g., Time
is money). The inference patterns and value judgements that are associated with
the image schema are generally preserved in the metaphorical mapping onto the
abstract target domain (Lakoff 1990). Since conceptual metaphors that map image
schemas are motivated by everyday bodily experience, abstract thought is
fundamentally 'embodied' (Johnson 1987).
The second category of conceptual metaphors builds on more specific or
more elaborate source domains. Abstract competition, for example, is often
structured in terms of RACING (e.g., Clinton was ahead of Dole in the polls) or in
terms of a FIGHT (e.g., Bush was defeated by Clinton). Again, different metaphors
are usually available to conceive of an abstract phenomenon. The human mind, for
instance, is commonly understood as an EDIFICE, a piece of MACHINERY, a
COMPUTER, an ORGANISM, and so on (e.g., Roediger 1980). As each metaphor
maps its proper inference patterns, they may guide one's reasoning about abstract
phenomena (Gentner & Gentner 1983).
48 FRANK BOERS
2. Parameters of variation
If abstract thought is largely metaphorical, and if different metaphorical
perspectives on a single domain exist, then we may wonder whether different
communities conceive of certain abstract phenomena differently because of the
conceptual metaphors that are most readily available according to cultural or
linguistic conventions (Lakoff 1987:295).
Since bodily experience is pretty much alike across the globe, image
schemas are likely to be universal, and their associated conceptual metaphors will
probably be shared by many different cultures. THE BODYISA CONTAINER FOR THE
EMOTIONS metaphor, for example, can be found in languages as remote as English,
Hungarian, Chinese and Japanese (Kvecses 1995). Nevertheless, cultures may
differ by virtue of the more specific imagery that is often added to the general
image schema. For instance, while English takes the body as a whole as the
CONTAINER for anger (the latter being conceived as a hot fluid), other languages
may show a preference for locating this emotion in specific parts of the body, like
the head in Hungarian, or the stomach in Japanese (Kvecses 1995). PATH
metaphors (e.g., Is this article leading anywhere?) abound in both English and
French, but (probably due to historical reasons) the additional imagery in English
is more often that of ships and sailing (Boers &Demecheleer 1997). When various
metaphors are available to structure a single abstract domain, they may have
different degrees of popularity or conventionality. In other words, a given
metaphor may be more typical of the discourse of one community than that of
another. One way of measuring the degree of popularity or conventionality of a
certain metaphor in a community is counting its frequency of occurrence. Using
this frequency principle may point to different preferences of image-schema based
metaphors, even at the level of closely related languages or language varieties. For
instance, the TIME IS A MOVING OBJECT metaphor appears to be more common in
British than American English (Boers 1996).
Cross-cultural variation is likely to be more pronounced when metaphors of
the second category are concerned, i.e. metaphors that build on more specific or
elaborate source domains. Corpus-based studies of socio-economic discourse
indicate, for example, that HEALTH metaphors (e.g., The market cure), are more
productive in English than in Dutch. GARDENING metaphors (e.g., Pruning costs)
are more frequent in English than in French, where FOOD metaphors (e.g.,
Gobbling up small companies) are more common (Boers & Demecheleer 1997).
These observations are not confined to linguistic communities. Metaphorical
models and their relative popularity may also vary across communities defined in
ideological terms. ARCHITECTURAL metaphors, for instance, are especially typical
of the discourse of free-masons (Rigotti 1995). In the socio-economic domain,
FITNESS metaphors (e.g., Slimming down) and RACING metaphors (e.g., Staying
WHEN A BODILY SOURCE DOMAIN BECOMES PROMINENT 49
ahead of the competition) appear to be more common in freemarketeer rhetoric

than in socialist oriented discourse (Boers 1997a).
The range of available conceptual metaphors is also subject to change over
time. Metaphors of the human mind, for example, tend to draw analogies from
developments in the domain of technology. Over time, as new machines have been
invented, their characteristics have often been selected to metaphorically structure
and describe mental processes (Draaisma 1995). In fact, most scientific revolu
tions correspond to novel metaphorical perspectives on a given subject (Johnson
1987:129; Miller 1995). The popularity of existing metaphors may grow or
decline over time, along with scientific evolution (Gentner & Grudin 1985) or
along with ideological changes (Rigotti 1995). An example of the latter can also
be found in socio-economic discourse about the notion of social security (welfare).
Up to the early nineties, the standard metaphor for social security used to be a
shelter or a cover (Living under the umbrella of the welfare state). As the
influence of free-market ideology has grown, that metaphor has been replaced by
the image of the safety net, which feeds different inference patterns (Boers 1997a).
To sum up, variation with respect to the relative popularity of conceptual
metaphors occurs along multiple parameters. In the following section we shall
take a closer look at yet another type of variation over time.
3. The HEALTH metaphor: A case of seasonal adjustment?

Various metaphors are usually available to conceive a single abstract
domain. In general, the likelihood of a given source domain being used for
metaphorical mapping may be enhanced when it becomes more salient in everyday
experience. Although bodily experience is probably the most basic source domain
for metaphor, people's awareness of their bodies may vary as well. One
circumstance under which the awareness of one's bodily functioning is enhanced
is when it starts malfunctioning, i.e. when one gets ill. When a range of source
domains, including the bodily source domain, is available for the metaphorical
conception of a certain abstract target domain, are we then more likely to 'choose'
the bodily source domain when it becomes more salient in our everyday
experience?
Weather conditions, along with a variety of other factors, can affect one's
health, and weather conditions usually cause most hardship in winter (at least in
the northern hemisphere). From experience and observation we know that
illnesses like colds, bronchitis, laryngitis, influenza, pneumonia, etc. occur most
often in wintertime. If people are more often confronted with health problems in
winter, then they may also be more likely to use this bodily experience as a source
domain for metaphorical mapping in that season. The abstract domain we shall be
looking at to test this hypothesis is socio-economics again.
50 FRANK BOERS
Evidently, the HEALTH metaphor is just one of many conceptual metaphors

that are used to structure economic phenomena and policies. Other source domains
include WARFARE (e.g., Invading foreign markets), RACING (e.g., Lagging behind
in economic development), MACHINERY (e.g., The exchange rate mechanism) and
GARDENING (e.g., Pruning costs). Each maps its own inference patterns and
associated value judgements. Experimental research has shown that exposure to
these metaphors can indeed have a profound effect on people's reasoning about
socio-economic issues (Boers 1997b). So, could one's reasoning about economic
processes and policies be guided more often by the HEALTH metaphor in winter
than in the other seasons?
In order to try and answer this question, we systematically counted the
instances of the HEALTH metaphor in the "leaders" (i.e., the editorials) of all the
weekly issues of The Economist over a ten-year period, from April 1986 to March
1996. In these articles, the editors of The Economist give their analysis of and
opinion about a variety of political and economic topics. Only articles that clearly
dealt with economic subjects were selected (usually 3 articles per issue). This
yielded a total sample of about 1,137,000 words.
The instances of the HEALTH metaphor in our sample show a wide lexical
variety : thriving industries, Vibrant enterprises, Economic paralysis, A crippling
strike, Sclerotic industries, Arthritic markets, A healthy economic climate, Healthy
companies, Sickly firms, Symptoms of a corporate disease, A chronic deficit,
Diagnosing a shortage, A financial haemorrhage, Anaemic industries, The right
economic remedy, Prescribing the best economic medicine, The market cure, A
financial injection, Surgery that costs jobs, Amputating unprofitable departments,
Economic recovery, Economic revival, Subsidies become a palliative, and so on.
When the instantiations consisted of several words (e.g., Prescribing the right
cure), then these were counted as single occurrences, unless they stretched over
clause boundaries.
Some instances in the sample were 'creative' (e.g., Arthritic labour
markets), while others were highly conventional (e.g., A healthy economy). Within
the set of conventional expressions, some were clear instances of the HEALTH
metaphor (e.g., The market cure), while some others were vague or ambiguous
(e.g., Economic recovery, Economic remedy). Sometimes the context clearly
pointed to the source domain of health, but in a number of cases the association
with health was hard to determine with any certainty. In those cases, a distinction
was made between 'clear' instances and 'vague or ambiguous' ones. To make this
distinction the Collins Cobuild English Language Dictionary was referred to. The
entries in Cobuild are organized on the basis of a frequency principle, with the
most common usage of a lexical item mentioned first. When the dictionary
explanation of this first sense mentioned the domain of physical health, then its
figurative use in the sample was taken as a 'clear' instance of the HEALTH meta
phor (e.g., cure). Otherwise, for instance when only the second entry mentioned
physical health (e.g., remedy), the items were counted as 'vague or ambiguous'.
The quantitative data were analyzed twice: once for the 'clear' instances only, and
a second time for all the counted instances, including the 'vague or ambiguous'
ones.
It needs to be acknowledged at this point that the adoption of a frequency
principle to make such a distinction does not necessarily reflect the intuitions (or
semantic priming effects) at the level of individual language users. The frequency
principle can be taken as valid only with respect to trends in a linguistic
community as a whole. The inherent limitation of corpus-based research is that it
offers no clear evidence of what actually goes on in an individual language user's
mind. This calls for complementary experimental research. Nevertheless, the
corpus-based analysis may suffice for the present purposes, especially since the
quantitative data concerning the 'clear' instances and those concerning the
'ambiguous and vague' ones turned out to be mutually supportive.
Figures 1 and 2 represent the average frequency of occurrence of the
HEALTH metaphor per month over stretches of 1,000 words. Frequencies are
clearly highest from December to March. The average in that period for the 'clear'
instances is 1.12, compared to an average of 0.60 for the other months (including
the 'vague or ambiguous' category, the ratio is 1.74 versus 1.09, respectively). 22
out of 40 winter months show a frequency of 'clear' instances of over 1.00 per
1,000 words. For the other months this ratio is only 7 out of 80. (A chi-square test
yields a significance level of p < 0.001.)
Figure 1: Average frequencies per 1,000 words per month : "clear" instances only
52 FRANK BOERS
Figure 2: Average frequencies per 1,000 words per month : all instances
Figures 3 and 4 trace the average frequency of occurrence of the HEALTH metaphor
over the ten-year period, contrasting December-March with April-November.
While the charts show varying degrees of popularity of the HEALTH metaphor over
time in general, they clearly show a systematic pattern with the highest peaks in
winter times. Only one year (1987) appears to deviate from that pattern. This,
however, is due to an unusually high number of instances in April 1987.
Coincidentally, that month happened to be unusually cold (according to the
weather reports we consulted). In fact, both November and April occasionally
show high frequencies of occurrence of the HEALTH metaphor. Since both these
months are adjacent to winter, this need not undermine our hypothesis, however.
Figure 3: Average frequencies per 1,000 words per period: "clear" instances only
Figure 4: Average frequencies per 1,000 words per period: all instances
A more flexible approach to tracing fluctuations in relative frequencies is shown

in figures 5 and 6. If we allow for this kind of flexibility in the analysis, the cyclic
trend shows up even more.
Figure 5: Average frequencies per 1,000 words per period (alternative):

"clear" instances only
In short, the quantitative data drawn from our corpus corroborate the hypothesis:
the domain of physical health is more likely to be used for metaphorical mapping
in winter, i.e. as it tends to become more salient in people's everyday experience.
One might argue against the above interpretation of the data by referring to
the specific rhetorical nature of the discourse analyzed. The editorials of The
Economist constitute argumentative discourse, abounding with rhetorical devices
such as the deliberate use of metaphor. In a fair number of articles, the authors are
clearly using metaphors consciously and creatively. They may decide to choose
HEALTH metaphors especially frequently in winter, simply because of their
54 FRANK BOERS
intuition that readers may be more susceptible to them in that period. This line of
argument suggests that the authors are not influenced by a growing awareness of
their own bodily existence. Nonetheless, the influence would still exist, albeit
indirectly, due to the observed (or assumed) growing prominence of the domain
of physical health within the community as a whole.
Figure 6: Average frequencies per 1,000 words per period (alternative): all instances
The relatively low frequency of HEALTH metaphors in summer time could also be
the result of a general trend of 'poor' metaphor use in this period. After all,
summer tends to be a slack period for politics and journalism in general. Under
that explanation, however, one would expect the frequency of other conceptual
metaphors to be relatively low in summer time, too. In order to examine this
possibility, we reconsidered part of our sample (from December 1994 to January
1996, i.e. approximately 110,000 words) and we counted two other common
metaphors: WARFARE and RACING. The average frequency of occurrence of the
WARFARE metaphor per 1,000 words was 1.54, with an average of 1.33 in winter
and a slightly higher average of 1.76 in summer. The RACING metaphor occurred
on average 0.41 times per 1,000 words, with a relatively low frequency of 0.33 in
winter and a slightly higher frequency of 0.44 in summer. In short, the distribution
over time of these metaphors does not at all show the same seasonal fluctuations
as those noted for the HEALTH metaphor. Hence, the additional exercise does not
give any evidence of a general pattern of 'poor' metaphor use in summer time.
Socio-economic issues belong to the realm of the abstract throughout the year and
so it is not surprising that their conception requires the use of metaphor throughout
the year as well.
4. Conclusion
It has been argued in this paper that, in general, a source domain is more
likely to be used for metaphorical mapping as it becomes more salient in everyday
experience. This increase or decrease of salience may also be noted for the source
domain of the human body. One circumstance under which the body becomes
more salient (i.e., when the awareness of one's bodily existence is enhanced) is
when it starts malfunctioning, like in cases of illness. Since many common
illnesses are related to bad weather conditions, this type of growing bodily
awareness will occur most typically in winter (in the northern hemisphere). Our
corpus-based quantitative analysis of 10 years of opinion articles in The
Economist reveals that the authors' use of the HEALTH metaphor, rather than other
available conceptual metaphors, is indeed especially frequent in winter times.
If - as suggested by experimental research - these metaphors guide people's
abstract reasoning, then the observed seasonal fluctuation may be taken as indirect
evidence of the connection between bodily experience and abstract thought.
References
Boers, Frank. 1996. Spatial Prepositions and Metaphor: A Cognitive Semantic
Journey along the UP-DOWN and the FRONT-BACK dimensions. Tbingen:
Gunter Narr.
1997a. "Health, Fitness and Mobility in a Free-Market Ideology". Voices
of Power: Co-operation and Conflict in English Language and Literatures,
ed. by Jean-Pierre van Noppen, & Mark Maufort, 89-96. Lige: Language
and Literature.
1997b. "No Pain, No Gain in a Free-Market Rhetoric: A Test for Cognitive
Semantics?" Metaphor & Symbol 12. 231-241.
, & Murielle Demecheleer. 1997. "A Few Metaphorical Models in
(Western) Economic Discourse". Discourse and Perspective in Cognitive
Linguistics, ed. by Wolf-Andreas Liebert, Gisela Redeker, & Linda Waugh,
115-129. Amsterdam/Philadelphia: John Benjamins.
Draaisma, Douwe. 1995. De Metaforenmachine : Een Geschiedenis van het
Geheugen. (The Metaphor Machine: A History of Memory). Groningen:
Historische Uitgeverij.
Gentner, Dedre,, & D. Gentner. 1983. "Flowing Water or Teeming Crowds:
Mental Models of Electricity". Mental Models, ed. by Dedre Gentner, & A.
Stevens, 99-129, Hillsdale: Lawrence Erlbaum.
, & Jonathan Grudin. 1985. "The Evolution of Mental Metaphors in
Psychology: A 90-year Retrospective". American Psychologist 40. 181-192.
Johnson, Mark. 1987. The Body in the Mind: The Bodily Basis of Meaning,
Imagination and Reason. Chicago and London: University of Chicago Press.
56 FRANK BOERS
Kvecses, Zoltan. 1995. "The Container Metaphor of Anger in English, Chinese,

Japanese and Hungarian". From a Metaphorical Point of View: A Multi-
disciplinary Approach to the Cognitive Content of Metaphor, ed. by
Zdravdo Radman, 117-147, Berlin and New York: Walter de Gruyter.
Lakoff, George. 1987. Women, Fire and Dangerous Things: What Categories
Reveal about the Mind. Chicago and London: University of Chicago Press.
1990. "The Invariance Hypothesis: Is Abstract Reasoning Based on Image-
Schemas?" Cognitive Linguistics 1. 5-38.
Miller, Arthur I. 1995. Imagery and Metaphor: The Cognitive Science Con-
nection, ed. by Zdravko Radman, 1995. 199-224.
Rigotti, Francesca. 1995. The House as Metaphor, ed. by Zdravko Radman, 419-
446. Berlin and New York: Walter de Gruyter.
Roediger, Henry L. 1980. "Memory Metaphors in Cognitive Psychology".
Memory & Cognition 8. 231-246.
FROM LINGUISTIC TO CONCEPTUAL METAPHOR IN FIVE STEPS
GERARD STEEN
Tilburg University and Free University Amsterdam
How does the cognitive linguist get from linguistic metaphor to conceptual
metaphor? Is there a procedure for the determination of conceptual metaphor when
metaphorical language has been encountered? These are the questions that are
addressed in this chapter, which aims to build a bridge between linguistic and
conceptual metaphor by proposing a series of five analytical steps. Together they
may form the beginning of a procedure for conceptual metaphor identification in
discourse.
The procedure is meant to constrain the relation between linguistic and
conceptual metaphor. It has sometimes remained an act of faith that particular
metaphors in language reflect particular metaphors in thought. This does not mean
that there is no linguistic support for the existence of conceptual metaphors. And
indeed, there are many clear cases in which the name of a particular conceptual
metaphor is used in a linguistic expression, as can be demonstrated by a brief
glance at the by now classic list of references Lakoff & Johnson (1980), Johnson
(1987),Lakoff (1987),Turner(1987), Lakoff & Turner (1989), and Lakoff (1993).
However, these clear cases serve the purpose of demonstration; they have not been
systematically and exhaustively collected from large stretches of discourse, but
they have been selected for their persuasive power. Now that the theory of
conceptual metaphor has been firmly established as one important component of
a general theory of metaphor, providing one of the main inspirations to cognitive
linguistics as a general approach to language, it is time to reverse the perspective.
Thence the question arises how stretches of discourse can be said to express
particular conceptual metaphors as opposed to others, and this is a difficult issue.
It presupposes a generally accepted procedure of deriving conceptual metaphors
from linguistic metaphors encountered in on-going discourse, and that is currently
not available.
Most readers will be familiar with some of the examples of metaphorical
correspondences between conceptual domains such as the following:
58 GERARD STEEN
THE LOVE-AS-A-JOURNEY MAPPING

The lovers correspond to travelers.
The love relationship corresponds to the vehicle.
The lovers' common goals correspond to their common destinations one the
journey.
Difficulties in the relationship correspond to impediments to travel.
(Lakoff, 1993:207)
But from the present perspective, these are at best the output of the last step of the
envisaged procedure, and this would probably only hold in ideal cases. What I am
interested in is to explicate the assumptions that lead linguists to arrive at such
conceptual mappings in departing from metaphorical expressions in discourse.
This chapter is a logical reconstruction of these assumptions in an attempt to reach
agreement about the steps that are inevitable when one goes from linguistic to
conceptual metaphor identification.
It is noteworthy that this explication can be related to a number of
theoretical issues which were previously discussed in the seventies, before the
advent of conceptual metaphor theory as we now know it (Cohen & Margalith,
1972; Van Dijk, 1975; Reinhart, 1976; Cohen 1993; Miller 1993). In retrospect,
most of these references can be seen as attempts to make the jump from linguistic
to conceptual metaphor in one way or another, but they failed to do so in an
optimal manner because of the lack of a well-developed conceptual theory of
metaphor. The time is now ripe to return to these issues in order to put conceptual
metaphor theory on a firmer linguistic footing. It is ironic that cognitive linguists
are going out of their way to show that linguistic metaphor is fundamentally
conceptual, but that in doing so, they have neglected the method for showing how
they get from linguistic metaphor to conceptual metaphor in the first place.
My recourse to these sources has one consequence which may be misleading
and which has to be circumvented from the beginning. Some or most of the
examples discussed by theorists in the seventies were not of the conventional kind
that have since become popular in the literature. In present-day terms they might
be seen as one-shot and often poetic metaphors rather than systematic conceptual
metaphors. Moreover, I do not address the question whether my illustrations of
metaphor are actually found in other expressions of a similar kind, which is the
generally accepted approach to establishing conceptual metaphors in cognitive
linguistics. These may be surprising features of a chapter titled "From linguistic
to conceptual metaphor in five steps." However, I believe that they are actually
immaterial to the purpose of this particular contribution, which is to reconstruct
how the linguist gets from linguistic metaphor to conceptual metaphor. For
methodologically speaking, the linguist has no a priori knowledge whether a
FROM LINGUISTIC TO CONCEPTUAL METAPHOR IN FIVE STEPS 59
particular expression is to be counted as a one-shot metaphor or as a systematic

metaphor: he or she first has to identify metaphorical expressions and determine
what the conceptual nature of the metaphorical expression in question is. Only
once this has been achieved, can the metaphorical concept be examined as to its
possible relation(s) with other metaphorical concepts, which then leads to a
decision about one-shot conceptual metaphoricity versus systematic conceptual
metaphoricity. Such a comparison across metaphors presupposes that the other
metaphorical concepts have also been collected from discourse analysis in the
same fashion. What I am focusing on, then, is the procedure for collecting such
metaphorical concepts with the purpose of examining their systematic relations.
If one insists on regarding as conceptual metaphors only those metaphors which
are systematic (as opposed to one-shot metaphors), which I do not, then a sixth
step will have to be added to the procedure, saying that the output of the first five
steps is to be compared across large numbers of metaphors in order to establish
more or less systematic groups of metaphorical concepts, labeling the largest
systematic groups as conceptual metaphors.
I have to add one more caveat from the beginning. I wish to emphasize that
I am dealing with metaphor analysis, not metaphor understanding. Metaphor
analysis is a task for the linguist who wishes to describe and explain the structure
and function of language. Metaphor understanding is a cognitive process which
is the object of investigation of psycholinguists and discourse psychologists who
are conducting behavioral research. This chapter does not deal with behavior. This
does not mean that it cannot make use of theories of metaphor understanding for
the identification of specific stages in the analytical procedure; on the contrary, it
would be odd if there were no connection between understanding and analysis.
However, metaphor analysis is a goal- and norm-related activity in the pursuit of
data collection. It is the intentional technical identification of conceptual
metaphors from metaphorical language in discourse. This chapter is concerned
with a logical reconstruction of the discrete steps involved in that activity.
The logical reconstruction may then be transformed into a procedure for
practical use in linguistic research. From the perspective of the cognitive linguist,
who is interested in the analysis of discourse and the way it reflects concepts and
cognition, it is essential that there is such a procedure for relating linguistic
metaphor to conceptual metaphor in a reliable fashion. Ultimately, the cognitive
linguist has to begin with stretches of discourse and determine which linguistic
expressions are metaphorical and related to which conceptual metaphors, and this
is no trivial matter. However, such a procedure is also important for constructing
the link between cognitive linguistics and psycholinguistics, in which precise
descriptions of literal and nonliteral materials are needed for the development of
well-controlled linguistic stimulus materials. Manipulating texts and expressions
60 GERARD STEEN
with a view to activating particular conceptual metaphors requires the same solid
foundation in linguistic methodology.
1. Metaphor focus identification

The first step of the procedure involves the identification of metaphorical
expressions in discourse. This naturally involves the theoretically thorny issue of
the definition of metaphor itself. As it is the purpose of this contribution to present
a procedure for conceptual metaphor identification, which in itself more or less
presupposes that we know what a metaphor is, I will cut a long story short and
make the following assumption. It seems best to adopt the most widely accepted
definition of metaphor that is currently available, the Lakoffian one of metaphor
as a set of correspondences between two conceptual domains, with linguistic
metaphor deriving from conceptual structures. The presence of two domains is
intended to capture the fact that we are dealing with nonliteral similarities between
entities and relations at some level of the analysis, which rules out other types of
mappings like metonymies.
The first step consists of identifying metaphorical expressions in discourse,
and we now need to become more precise about the nature of this operation. For
it is not true that identifying metaphorical expressions is tantamount to identifying
linguistic metaphors. I will now show that when expressions are identified as
metaphorical, it is the focus of the metaphor that we are dealing with, and this is
only one part of the complete metaphor. In effect, it depends on a number of
factors what one will call the complete metaphor in the first place. This observa
tion needs to be explained with reference to a number of familiar but sometimes
elusive concepts in metaphor theory.
One of the more interesting discussions in this respect is Reinhart (1976),
who has aligned some of the classic theories of metaphor such as Richards (1936),
Beardsley (1958), and Black (1962). The main point of interest at present is best
introduced by comparing the following metaphors, discussed by Reinhart (1976):
(1) I have seen the mermaids riding seawards on the waves (T.S. Eliot, The love song
of J. Alfred Prufrock')
(2) The royal court is going to hunt
As an aside, the actual line from Eliot does not contain 'the mermaids' but 'them',
but Reinhart has explicated the anaphoric pronoun for expository purposes.
Reinhart explains that the focus of the metaphor in (1) is riding on. The
focus of the metaphor in (2), which is about 'lions stalking their prey' (1976:391,
fn. 8), is the royal court. Together, (1) and (2) illustrate that the focus is the
linguistic expression used nonliterally in the discourse. This means that the focus
expression activates a concept which cannot be literally applied to the referents in
the world evoked by the text. The concept RIDING-ON cannot be literally applied
to the relation projected between the entities referred to by MERMAIDS and WAVES,
and the concept ROYAL COURT cannot be literally applied to the entity referred to
by LIONS. It will be noted that referents can be entities, relations, and attributes in
some situation evoked by the discourse.
The twin concept related to focus is frame, which I will define, with
Reinhart and others, as the immediate linguistic environment of the metaphor
focus. In (1), the frame is waves, according to Reinhart (1976:385). However,
Reinhart does not explicate the linguistic frame of (2). This is probably because
(2) is a special class of metaphor: the focus is not non-literal in relation to the rest
of the linguistic expression, is going to hunt; there is no semantic tension between
focus and frame. The kind of metaphor exemplified by (2) is purposefully left
aside in Reinhart's attempt at ordering the theoretical concepts for the analysis of
metaphor. As we shall see below, a cognitive linguistic approach which includes
linguistic and conceptual as well as other discourse aspects of analysis is better
equipped to handle these issues.
There are some important issues in Reinhart's discussion of focus and
frame, as well as of other notions like tenor and vehicle, and some of them will
return later. But our present concern is the first step in the conceptual metaphor
identification procedure. It may now be appreciated more fully that the first step
is largely concerned with metaphor focus identification, not linguistic metaphor
identification as a whole. The reason is this. Many metaphor foci may be related
to a literally used concept which is explicitly expressed in the metaphor frame, as
in the case of (1); however, there is also a good number of metaphor foci located
in metaphor frames without a linguistic expression of the literal concept of the
metaphor, as in (2). When both literal and nonliteral concept are present in the
frame, it is possible to identify a complete linguistic metaphor in the first step of
the analysis. However, when the literal concept is not expressed in the frame, the
linguistic metaphor cannot be identified in the first step; then it is only the focus
that is identified in the first step. These are implicit metaphors (Steen 1999) and
they need explication through propositional analysis, which clarifies to which
concept the nonliteral concept expressed by the focus is applied. For instance,
words like scene, shit, heat, and so on are often used metaphorically without the
literal concept to which they are applied being expressed explicitly in the
discourse. As it is the metaphor focus that can be identified throughout all classes
of metaphor in the first step, this is why step 1 of the procedure is called metaphor
focus identification.
62 GERARD STEEN
2. Metaphorical idea identification

Now that we have a metaphor focus as a result of step 1, the question arises
to what other kind of element the focus applies. As a matter of principle it cannot
be the linguistic frame, as the linguistic frame merely provides a linguistic
background in which the focus is a nonliteral expression which may or may not
stand out as the case may be. In other words, the focus is not a focus on account
of its relation to the linguistic frame, as we have observed, for it may even exhibit
a literal relation to the frame, as was shown by (2). What does make a focus into
a focus is the fact that it expresses a concept which is to be related to another
concept to which it cannot be applied in a literal fashion: 'riding on' cannot be
literally applied to 'mermaids' doing something to 'waves', and 'the royal court'
cannot be literally applied to 'lions'.
The other, literal, concept has been variously referred to as the tenor or the
topic or the principal subject of the metaphor, but I will call it the literal part of the
metaphorical idea. As not all literal parts of metaphors are explicitly expressed in
discourse, as is the case for (2), they sometimes have to be inferred. This is why
metaphor identification is a matter of concepts, propositions, and reference. As
these are general aspects of discourse analysis which are not limited to metaphor,
and propositional analysis was specially designed to cater for them, it is now time
to turn to propositional analysis. Propositional analysis was developed independ
ently of the study of metaphor and also aims to bridge the gap between discourse
and conceptualization (see the contributions to Britton & Black, 1985; in
particular Bovair & Kieras, 1985).
Consider the following propositional analysis of (1), a personal, notational
variant of mine of the technique used by Bovair & Kieras (1985):
(3) I have seen the mermaids riding seawards on the waves

P1 (SEE I P2)
P 2 (RIDE-ON MERMAIDS WAVES)
P 3 (DIRECTION P2 SEAWARDS)
This analysis of (1) is maximally consistent with the one of Reinhart (1976), even
though I can see that at least one other analysis is possible, namely one in which
P2 merely consists of (RIDE MERMAIDS) and an additional P4 is needed to capture
that (ON P2 WAVES). However, since this difference is immaterial to my argument,
I will not unnecessarily increase the complexity of the exposition and leave this
point aside.
Irrelevant details aside, (3) is a linearly and hierarchically ordered list of
propositions, each consisting of a predicate and one or more arguments. The list
captures the meaning of (1) as a series of minimal idea units, or propositions, and
presents the structural relations between the concepts contained by the idea units.
All of this is relatively uncontroversial in discourse psychology (Britton & Black
1985; Perfetti & Britt 1995; but cf. Garnham 1996).
The metaphorical idea that is present in T.S. Eliot's line is found in
proposition P2. The nonliteral part of the metaphorical proposition is the concept
RIDE-ON, which functions as a predicate, and it can be seen to apply to two other
concepts, MERMAIDS and WAVES. They are conceptualizations of the literal
referents in the projected text world about which something is said in a metaphori
cal manner, namely that they are in a relation of the one 'riding on' the other. This
relation could also have been conceptualized in a literal manner, for instance by
means of FLOATING, which would have yielded a literal expression. This is also
accounted for by Reinhart's definition of focus:
Given a metaphorical expression Fi[Ei]

Ei is the focus if it is possible to substitute Ei for Ei, so that Fi[Ej] is a literal
expression and Fi[Ej] is similar in meaning to Fi[Ei]. (1976:391)
Whereas RIDING-ON does not refer literally to the presumed relation between the
entities of mermaids and waves in the projected textworld, FLOATING does and
produces a similar meaning.
The conceptual basis of this approach is clarified by considering the other
metaphor and its propositional analysis:
(4) The royal court is going to hunt

P1 (REF COURT LIONS)
P2 (HUNT COURT)
P3 (MOD COURT ROYAL)
The crucial proposition is P1, where it is clarified that one concept in S2, COURT,
refers to another concept, LIONS, which is available from the previous (or
following) discourse, be it co-text or context. This explicit and immediate form
of reference assignment between arguments is standard procedure in Bovair &
Kieras' (1985) method of propositionalization, and is aimed at ensuring referential
coherence between concepts in consecutive sentences, especially when different
expressions are used about the same entities in a projected text world. Of course
this is precisely what is needed to solve the problem of cases like (2) for the theory
of metaphor: in order to be able to interpret the main idea unit P2 correctly, it is
first necessary to clarify potentially confusing elements such as 'court' when there
is no literal entity of that kind in the projected text world. This is the function of
P1, which thereby automatically identifies a complete metaphor: it relates the
64 GERARD STEEN
nonliteral concept COURT to the literal concept it is supposed to apply to, LIONS,
by means of the specially designed predicate 'REF' , and reveals that the metaphor
of (2) is P1. All of these observations are intended to capture the propositional
structure of the language: they are not meant to be read as claims about compre
hension or other forms of processing.
Propositional metaphor analysis can apply to all kinds of metaphors. It also
lays bare how metaphors can differ from each other with respect to important
dimensions of conceptual structure. For instance, Miller (1993:387) discusses the
metaphor a watchdog committee. I did not find it coincidental that he presented
the metaphor in precisely this form, and ran an automatic Microconcord search of
the Times corpus of approximately 1,000,000 words coming with that program.
There were 10 occurrences of watchdog in all, and they were all metaphorical in
the above sense. What is interesting is the formal variation between the instances.
There was one case of Miller's structure, a watchdog organisation. It exemplifies
an explicit metaphor, because it contains both the literal and the nonliteral concept
of the underlying proposition. It is also a reduced metaphor, in that the linguistic
structure is not equivalent to a proposition itself, but is a nominal phrase. (A full
version of this metaphor would be something like: the committee is a watchdog,
which is not likely to be found in genuine discourse, as is shown by the results of
this search.) The metaphor is also simple, in that there is no additional material
attached to the nonliteral concept itself. This should be contrasted to the following
class: the National River Authority, the new watchdog for the water industry, and
the Audit Commission, the local authority financial watchdog. In the latter case,
the concept WATCHDOG is modified by additional concepts, such as NEW, turning
the metaphorical focus into a complex structure needing more than one proposi
tion. Another class of metaphor may be illustrated by a national watchdog and an
independent watchdog', these are implicit metaphors, because the literal concept
to which WATCHDOG is applied is not expressed in the text; and they are complex
metaphors, because WATCHDOG does not stand by itself. Half of the ten watchdog
metaphors were of this kind. The only example of an implicit and simple
metaphor was found in a headline: watchdog may be too fierce; contextual
assumptions about the use of words in headlines apparently have relaxed the need
for adding information to the nonliterally used concept WATCHDOG which was
exemplified by all of the other cases. Propositional analysis is hence a valuable
tool in metaphor classification and raises consequential questions for interpreta
tion and processing (Steen to appear).
Propositional metaphor analysis also clarifies some interesting aspects of the
previous discussion of Reinhart (1976). In accordance with Reinhart and the
theorists she discusses, the use of propositional analysis emphasizes that the
nature of metaphor is conceptual and that the analyst can only access it through
the idea unit of the proposition rather than some rank of linguistic form, be it the
word, the phrase, the clause, or the sentence. The complete metaphor is only
identified when the appropriate literal and nonliteral concepts in the proposition
have been identitied. Metaphor identification, even linguistic metaphor identifica
tion, is fundamentally a matter of conceptual analysis. It deals with the concepts
referred to by the words and is at the core of a functional approach to language.
What is more, the complete metaphor is not always expressed as a complete
metaphor in the surface of the discourse: there are metaphors which are only
signalled by means of their focus expression, and their literal part has to be
inferred by means of propositionalization (implicit metaphors). Propositional
metaphor analysis hence also throws into relief the nature as well as the limited
role of the notion of frame: frames are nothing but the immediate linguistic
environment of the focus and hence do not always help in setting off the focus as
a nonliteral expression, as in (2). Frames have to receive a semantic and pragmatic
interpretation in terms of reference and intentions before metaphorical idea
identification can succeed, and this requires going beyond the surface of the
frame: metaphorical idea identification (step 2) has to follow after metaphor focus
identification (step 1). If many linguists were still uncomfortable with such an
approach in the seventies, the advent of discourse analysis, cognitive linguistics,
and related functional studies of language use is now alleviating this problem.
Propositional analysis also raises an interesting question about the scope of
the frame: in Reinhart' s analysis of (1), the frame is riding X, which in effect is the
next grammatical level up in comparison to the focus (the waves). However, as is
shown by the propositional analysis, other reaches of the frame may be imagined
as well, ranging from the clause through the utterance to the sentence:
(5) a. riding on the waves

b. riding seawards on the waves
c. the mermaids riding on the waves
d. the mermaids riding seawards on the waves
e. I saw the mermaids riding seawards on the waves
This is no mere terminological or technical matter, for it determines whether one

says that (1) contains the metaphor 'riding on the waves' or, for instance,
'mermaids riding on the waves'. I would prefer the metaphorical proposition as
the indicator of the complete metaphor, as it embodies the turning point between
language and conceptualization, as we shall see in a moment. What is important
to point out here is that such decisions have effects on the coding and counting of
metaphors of varying linguistic structures in corpus analysis, precise operational
66 GERARD STEEN
definitions being required for the reliable annotation of data. This is precisely one
of the issues that a procedure for metaphor identification needs to resolve.
3. Nonliteral comparison identification

The output of step 2 is a proposition in which we have a nonliterally used
concept (expressed by the linguistic focus identified in step 1) that is related to one
or more literally used concepts identified in step 2 which evoke the relevant literal
referent(s). The literally used concept(s) may be explicitly expressed in the frame
or may have been inferred by the analyst from co-text or context. The input of step
3 is hence a metaphorical idea in the form of a proposition with literal and
nonliteral concepts.
We have assumed that metaphors are sets of correspondences between
conceptual domains in which nonliteral similarity or comparison plays a pivotal
role. Therefore we need to work from the metaphorical proposition towards a
conceptual representation of the mapping between the two conceptual domains
involved. The next step in our procedure is hence to begin to set up the compara
tive structure that is implicit in the nonliteral mapping between domains for every
conceptual metaphor. There is an excellent source for this objective in Miller
(1993), who presents a sophisticated view of the comparison theory and has been
unjustly neglected in conceptual metaphor theory.
The essential point of Miller's contribution is that '[R]econstruction of the
implied comparison is a critical step in understanding a metaphor' (1993:381).
That metaphors imply comparisons is shown by the postulation of conceptual
metaphors like LOVE IS A JOURNEY, LIFE IS A GAME, and so on. Metaphors in
discourse hence require 'reconstructing the conceptual basis of the comparison'
(1993:382) in order to be interpreted. This statement begs some fundamental
questions about on-line comprehension behavior, but it certainly holds true for the
off-line analysis of the relation between linguistic and conceptual metaphor, with
which we are concerned here.
When an author uses a metaphor, Miller (1993:384-5) says, 'The claim is
[...] that he had a general conceptresemblance, comparison, analogy--that we are
trying to appreciate. Such concepts have a structure, and S1 makes that structure
explicit.' S1 is the general form of all comparison statements, and encompasses
nonliteral and metaphorical comparison statements:
(6) Sl.SIM[F(x), G(y)]
This is a conceptual analysis of similarity and comparison, as Miller repeatedly

emphasizes (1993:377; 381; 382; 385; 389). Miller shows that it allows for an
analysis of metaphorical propositions as nonliteral analogies with conceptual
elements left out, and he proposes three specific rewrite rules which transform
metaphorical propositions into nonliteral comparisons. That is also why he
requires propositional analysis as providing the input for his analysis of the
underlying comparisons (1993:375-6): metaphorical idea identification (step 2)
has to precede nonliteral comparison identification (step 3). Let us examine how
this works in practice.
The metaphorical proposition of (1), (RIDE-ON MERMAIDS WAVES), may be
rewritten as a nonliteral comparison with the following structure:
(7) (RIDE-ON MERMAIDS WAVES) -> ( 3 F ) (By,y ') { S I M [ F ( M E R M A I D S , WAVES), RIDE-ON

(y,y')]}
A paraphrase of this formal notation of the conceptual structure of the implied

comparison would be the following: there is an activity (or relation) F and two
entities y and y ' such that there is a similarity between mermaids and waves ' doing
F' on the one hand and y riding on y' on the other. The input of the rule is the
output of our step 2, while the output of the rule is the automatic product of
Miller's rewrite rules.
Miller adds, 'The first step in interpreting [such a] comparison would be to
find appropriate values for the missing terms' (1993:384). This will be our step
4, nonliteral analogy identification. To jump ahead, in the present case we may fall
back on the analysis suggested by Reinhart (1976:388-90):
(8) (RIDE-ON MERMAIDS WAVES) SIM[FL0AT(MERMAIDS, WAVES), RIDE-ON (JOCKEY,

HORSE)]
Filling in the missing terms in this manner will be step 4 of the procedure, and
fleshing out the resulting analogy into a full-blown nonliteral mapping, step 5. For
now, we have to concentrate on step 3, identifying the underlying comparison.
The derivation of the comparison statements from the metaphorical
propositions created by step 2 is highly mechanical. Miller introduces three
general rules which automatically create comparison statements from proposi
tional input, for three classes of metaphors: nominal, verbal, and sentential.
(Actually Miller uses the term 'predicative', but to the linguist this is misleading,
for the nominal part of a metaphor like 'Man is a wolf' is also used predicatively.)
The rules are as follows:
(9)
68 GERARD STEEN
Although the structure of the comparison statements is complete, not all of the
concepts of the comparisons are known: the unknown ones are thematized, as it
were, by the two existential propositions preceding the comparison structures. For
instance, in the nominal metaphor 'Man is a wolf, also discussed by Miller
(1993:382), the underlying comparison suggests that there is a property of men
such that it is similar to some other property of wolves. It is the task of step 4,
nonliteral analogy identification, to attend to these unexpressed properties and fill
them in at the appropriate slots of the incomplete comparison statement generated
by rule M l . However, the derivation of the incomplete nonliteral comparison
statements themselves is much less interpretive, for it follows from the analysis
of a stretch of discourse as containing a metaphorical proposition of a particular
type, nominal, verbal, or sentential.
4. Nonliteral analogy identification

The fourth step handles the reconstruction of the complete nonliteral
comparison statement by inferring the implied concepts for the empty slots. Miller
calls this the reconstruction of the comparison, but I would like to call this
nonliteral analogy identification. It would serve well to suggest what step 4 adds
to the previous steps of (1) metaphor focus identification, (2) metaphorical idea
identification, and (3) nonliteral comparison identification: by filling in the empty
slots of the comparative structure produced by step 3, the incomplete nonliteral
comparison statement is turned into a full-blown nonliteral analogy in step 4. If
step 3 turns complete propositions into comparisons between two incomplete
propositions by means of rules Ml through M3, step 4 fills out each of these
incomplete propositions into complete ones. Nonliteral analogy identification
(step 4) has to follow after nonliteral comparison identification (step 3).
The term nonliteral analogy identification also suggests that the aim of the
entire undertaking is still to obtain reliable analyses of metaphorical language. Our
frame of research is the kind of claim that a particular stretch of language is an
expression of a specific conceptual metaphor, and if such claims are to be upheld,
we need to be able to support them with reliable identification procedures at every
stage of the analysis. In other words, the step from incomplete nonliteral
comparison statements to complete nonliteral analogies is supposed to be one of
identification, in which interpretation is to be kept under firm control: we are to
pick out the underlying analogy that may plausibly be conjectured to play a
guiding role in interpreting the incomplete comparative structure produced in step
3. In this connection it may be disheartening to find Miller issuing the warning
that 'there can be no uniquely correct comparison statement' (1993:384).
Somewhat later, he writes, '[T]he search for suitable values to convert [an
incomplete comparison statement to a nonliteral analogy, GS] is, strictly speaking,
a matter of interpretation' (1993:384). But Miller is right, and, what is more, there
is no contradiction between these warnings and our undertaking. Even if our goal
is one of analysis and identification, it will be good to remember that there are
objective limitations imposed on our pursuit and that interpretation needs to be
kept on the leash.
It is fortunately possible to say a little more about the actual procedure of
nonliteral analogy identification. For this purpose it is useful to return to Reinhart
(1976), whose main point was to introduce a distinction between focus and vehicle
identification and interpretation as two distinct aspects of understanding metaphor.
These two aspects may now be put to use in the identification of the nonliteral
analogy. I will begin with focus interpretation, but this does not mean that this
reflects the actual order of proceeding.
Let us examine another of Reinhart's examples. She discusses another
metaphor from T.S. Eliot's 'Love Song of J. Alfred Prufrock', reproduced under
(10):
(10) The yellow fog that rubs its back upon the window panes
Reinhart offers the following analysis of the focus:
In this metaphor, the focus is rubs its back upon, since we can substitute
another expression for it, e.g. touches, swirls against, or comes up against,
to yield a literal expression, such as The yellow fog that touches the window
panes. (1976:391)
Let us examine how this analysis fares in our procedure and how it can be inserted
into step 4. If we follow Reinhart in identifying the focus as she does, we
rediscover this in the following list of propositions produced by step 2:
(11) The yellow fog that rubs its back upon the window panes
P1 (RUB FOG BACK)
P2 (UPON Pl PANES)
P3 (MOD PANES WINDOW)
If we combine all of these propositions and take 'rub its back upon' as a complex
verbal group used in a nonliteral fashion, we need Miller's rule M2 to create the
following incomplete comparison statement as step 3:
(12) Rub back upon (fog, window panes) -> (E3F) (Ey,y') {SIM [F(fog, window panes),
rub back(y,v')]}
70 GERARD STEEN
To fill this in, as part of step 4, we require the operation of Reinhart's focus
interpretation. It is nothing but the search for the predicate F which is needed to
cement the relation between the fog and the window panes in the first incomplete
proposition of the comparison. The choice of 'touch' by Reinhart is one possible
option (although it is metaphorical, too). This would fill out the empty slot on the
literal side of the equation quoted as (12). Reinhart comments:
Focus-interpetation assigns a reading to the focus expression which is a

matter of selecting those properties associated with the focus expression
which are relevant to the context. Thus among the properties of rubbing
one's back, the properties of physical contact and of being in movement are
consistent with the context of Eliot's metaphor, hence they can be selected.
This procedure provides a rough understanding of what the metaphor is
about, or what the actual situation which is being depicted is (the fog
swirling against the window panes), and how it ties in with the wider
context of the metaphor. (1976:391-2)
Focus interpretation produces a partial but basic understanding of the metaphor.

In Miller's terms, it deals with 'the cast of characters in the reader's concept of the
text' (1993:382). Note the comparable referential style in Reinhart's passage,
when she uses 'what the metaphor is about, or what the actual situation which is
being depicted is.'
However, the focus interpretation of back rubbing can only take place if it
is also at least partly interpreted as rubbing one's back (see Reinhart's selection
of 'the properties of physical contact and of being in movement' in the above
quotation). In other words, the second incomplete proposition of (12) also has to
be part of the equation. This is the beginning of vehicle identification. And to
construct the full set of conceptual correspondences between the two conceptual
domains involved in the metaphor, the second incomplete proposition of (12)
leads us into a relatively independent consideration of the source domain. The
second incomplete proposition needs to be fleshed out itself, too, primarily by
filling in the empty slots of the arguments relating to rubbing one's back. This is
not just the beginning but nothing less than Reinhart's vehicle identification.
Reinhart's definition of the notion of vehicle squarely falls within our
operation of filling in the missing terms of the nonliteral side of the analogy:
The vehicle is the frame Fj[ ] in which the occurrence of Ej results in a

literal expression, Fj[Ei], where Fj[Ei] is not similar in meaning to Fi[Ei].
(1976:391)
Reinhart's vehicle is hence identical to the second existential proposition of the

incomplete comparison statement quoted above, (Ey,y'): we need to find a frame
consisting of two literal entities which may be related to each other by means of
the function 'rub its back upon' so that they form one coherent state of affairs. The
resulting proposition is supposed to be similar in a nonliteral fashion to the fog
touching the window panes. Reinhart suggests that 'cat' would be a good
candidate for the vehicle thus defined, and the poetic context of the line in
question provides corroboratory evidence.
There are again many aspects which would deserve further treatment.
However, let us conclude this section by pointing out that there is an important
difference between the two aspects of step 4. Focus interpretation is usually more
richly constrained by the context of the metaphor, in that it involves the
construction of a literal proposition against the background of the topic and
content of the previous discourse (see Reinhart's remarks above, and compare
Miller 1993:394). Vehicle identification does not have this rich contextual
constaint-it has to activate prototypical or default knowledge about the source
domain. Miller writes: 'usually it is sufficient to take as y whatever the most
generic argument G is conventionally predicated of (1993:393). Indeed, part of
the difficulty of vehicle identification is precisely that there may be more than one
source domain which can be involved in the interpretation of second completed
proposition (cf. Vosniadou and Ortony 1989). This is clearly not the case for focus
interpretation, in which the conceptual domain is identical to or part of the
conceptual domain of the stretch of discourse in which the metaphor is located.
Another crucial aspect is the relation between the two sides of the analogy.
Part of the meaning of y may have to be filtered out by its lack of relevance to the
literal topic of the metaphor and the discourse. However, many possible
assumptions about the source domain can usually be maintained with varying
degrees of strength in the context of the target domain: a range of possible values
for the empty slots is usually possible, that is, compatible with the textual world,
which makes it rather difficult to hit upon the best vehicle for the eventual
analogy. That focus interpretation and vehicle identification thus exert a mutual
influence on each other when one attempts to align the two parts of the analogy
is self-evident and the cornerstone of the interaction theory (cf. Miller 1993:394
and Reinhart 1976:389). This does make step 4 highly dependent on the specifics
of every particular metaphor.
5. Nonliteral mapping identification

The last step in the procedure is to identify the complete nonliteral mapping.
This is done by filling out the conceptual structure of the two sides of the
nonliteral analogy, the source and the target domain. Other concepts and the
72 GERARD STEEN
relations between them are to be listed for each domain, and their interdomain
relations are to be projected. Some of these additions may be motivated
semantically whereas others are based on our knowledge of the world in particular
domains (cf. Miller 1993:382 andReinhart 1976:388). The result of this operation
is a conceptual network from which the analyst may derive sets of correspon
dences such as those illustrated at the beginning of this chapter (Lakoff 1993:207).
The transition from analogies to mappings is not given much principled
attention by Reinhart and Miller: they seem to relegate it to the domain of
interpretation about which they do not offer systematic observations. It is a theme
which, to my knowledge, has not been discussed by Lakoff either, except in a
negative fashion, as in (1993:210): 'Mappings should not be thought of as
processes, or as algorithms that mechanically take source domain inputs and
produce target domain outputs.' One of the more prolific writers in this area,
however, is Dedre Gentner (1982; 1983; 1989; Gentner & Jeziorski 1993), who
has done experimental research on behavior as well as AI modeling of knowledge
structures. She has listed six principles of analogical reasoning which may be used
to constrain the production of mappings from analogies. However, Gentner
(personal communication) has also admitted that the technical analysis of such
mappings is still basically an art. I cannot escape the impression that another
alternative in this area, Turner and Fauconnier's (1995) conceptual blending
theory, fares no better.
Thus the question arises how step 5 of the procedure may be better
constrained. And indeed, asking this question invites reviewing the other steps and
inspecting the relation between step 5 and step 4 in particular. For identifying the
analogy in step 4 seems to require at least a partial identification of the nonliteral
mapping, which is step 5, in order for the proposed analogy to be plausible. To
take this further, it might be suggested that identifying the analogy is a kind of
summary or abstraction of the nonliteral mapping identified in step 5, and this
could lead to the query why step 4 and 5 are not reversed. In other words, one
might also entertain an order in which the incomplete comparison statements of
step 3 are first fleshed out into conceptual domains with conceptual correspon
dences and are only afterwards condensed, as it were, into neat explicit analogies.
However, to raise the question is to answer it. For one needs a kind of
searchlight in the construction of the conceptual domains and their relations,
which involves a kind of propositional interpretation of the implied comparison.
Step 4 provides such a provisional interpretation in the form of an analogy, and
this analogy acts as a target for the construction of the more complex mapping (or
mappings, if you follow conceptual blending theory). There is hence a special
relation between steps 4 and 5, in which step 4 provides a tentative analogy which
is rejected or retained depending on the success of the ensuing step 5. This does
not happen anywhere between steps 1 and 4. Step 4 is crucial, because that is
where interpretation comes in, albeit as tightly controlled by the aims discussed
above as possible, and step 5 acts as a verification of step 4 in spelling out its
consequences in a more complex and explicit manner.
It is clear that the last two steps of the procedure form the weakest parts of
the chain, with step 5 being the weakest of all. Future research will have to
concentrate on strengthening these stages of the analysis. Otherwise, the gap
between metaphorical language and conceptual metaphor may never be bridged
in a reliable fashion. I will now move on to a brief illustration of the complete
sequence of steps.
6. Conclusion
I have suggested that the identification of conceptual metaphor in discourse
requires five steps:
(1) metaphor focus identification

(2) metaphorical idea identification
(3) nonliteral comparison identification
(4) nonliteral analogy identification
(5) nonliteral mapping identification
These steps are all called identification, because they give an answer to the
question 'What is ...?' The first three questions are easiest to answer: what is the
metaphor focus, what is the metaphorical proposition, and what is the metaphori
cal comparison? Question 4 is more difficult to handle, because it involves filling
in empty slots in an analogy on the basis of focus interpretation and vehicle
identification in mutual interaction. However, context and default language use,
respectively, act as guides to provide an answer to question 4. The answer to this
question has to be seen as a searchlight for constructing a nonliteral mapping,
which then has to be checked against the discourse regarding its appropriateness.
This step is the least reliable step in the procedure, while it is essential in order to
arrive at metaphors as sets of conceptual correspondences.
Let us return to the watchdog metaphor discussed above. Identifying the
focus of the metaphor involved checking that the word watchdog was not used in
its literal sense, referring to an animal guarding some property or people. This was
not difficult on account of the accompanying information which explicitly
signalled that the watchdog was T h e National River Authority' or 'the local
authority financial watchdog'. Such explicit lexical signals make focus identifica
tion an easy task, but they are often unnecessary given the overt contrast between
the domains of the metaphor focus and the topic of the text.
74 GERARD STEEN
The identification of the metaphorical proposition is not problematic either.

It involves constructing the proposition (BE ORGANISATION WATCHDOG) or some
variant thereof. As was shown above, there is some additional conceptual
variation between the individual cases of the metaphor.
The identification of the nonliteral comparison is done by feeding the
proposition into rule M l . This yields the paraphrase "Some property of a
committee is like some property of a watchdog."
Identifying the nonliteral analogy next concerns the finding of the properties
left open in the nonliteral comparison. I will begin with vehicle interpretation for
the sake of showing that there is no predetermined order between the two aspects.
Taking the canonical values of what watchdogs usually do, we can suggest that the
relevant property is 'guarding property or people'. The property of the committees
in question is similar, in that they have to guard the public interest in the economic
domain. One case has the following additional information: 'To protect the
interests of non-Treasury use of economic data'.
Fleshing out the analogy into a complete mapping involves listing attributes
of both committees and watchdogs and attempting to match them. The following
provisional list may be entertained:
THE COMMITTEE-AS-WATCHDOG MAPPING

The committee corresponds to the watchdog.
The organizational domain corresponds to the yard.
The interest or activity at risk corresponds to the property.
Malpractice corresponds to trespassing.
Monitoring corresponds to watching.
Warning the public corresponds to barking.
Other aspects can be included, and the list may be adjusted according to context.
As I have acknowledged at the beginning of this paper, there may be another
step, which is to compare the analysis of one metaphor with those of others. This
would be the last step in determining whether a metaphor is part of a systemati
cally organized set of metaphorical concepts (conventional conceptual metaphors)
or not. In our case, the analysis would not be dramatically different for every case
of the ten instances of the watchdog metaphor, and we might come to the
conclusion that the metaphor is relatively conventional, depending on other
frequencies.
There are many other issues which have also had to be left aside. For
instance, etymological and sociolinguistically restricted metaphor may pose
special problems in step 1. Implicit, reduced, complex, multiple, extended, and
mixed metaphors provide special challenges to steps 2 and 3 (Steen 1999 and to
appear). Differences between nominal, verbal, and sentential metaphor may cause
different situations for step 4. And so on. These and other issues are on the agenda
for future research. One specific item is the testing of the reliability of the various
steps: specialist informants may be given tasks to carry out steps 1 through 5,
either separately or in a row, with diverging sets of materials in order to reveal
more specific difficulties of applying the procedure to real discourse data.
The general conclusion, however, is that we have offered a logical
reconstruction of the analytical process, and that it holds a promise for practical
use as a descriptive tool in semantics. Again, the procedure does not pretend to
model the comprehension process. Moreover, as was suggested above, some steps
in the analysis make greater interpretative jumps than others, and this is where
cognitive linguists should be aware of alternative explanations of their data and
their beliefs in conceptual metaphors. It is hoped that the procedure can pinpoint
some of the more risky moments of analysis and that it can help in recording
experiences in negotiating these moments.
Author's note
I wish to thank Ray Gibbs, Rachel Giora, Lachlan Mackenzie, and Wilbert Spooren for
their acute observations and comments on an earlier version of this paper.
References
Beardsley, M. 1958. Aesthetics: Problems in the Philosophy of Criticism. New
York: Harcourt Brace Jovonovich.
Black, M. 1962. Models and Metaphors. Ithaca, N.Y.: Cornell University Press.
Bovair, S., & D.E. Kieras. 1985. "A Guide to Propositional Analysis for Research
on Technical Prose". Britton & Black 1985. 315-362.
Britton, B.K., & J.B. Black, ed. 1985. Understanding Expository Text. Hillsdale,
N.J.: Lawrence Erlbaum Associates.
Cohen, L. J. 1993. 'The Semantics of Metaphor". Metaphor and Thought, second
edition, ed. by A. Ortony, 58-70. Cambridge: Cambridge University Press.
-, & A. Margalith. 1972. "The Role of Inductive Reasoning in the Interpreta
tion of Metaphor". Semantics of Natural Language, ed. by D. Davidson, &
G. Harman, 722-744. Dordrecht: Reidel.
Dijk, T.A. van. 1975. "Formal Semantics of metaphorical Discourse". Poetics 4.
173-198.
Garnham, A. 1996. "Discourse Comprehension Models". Computational
Psycholinguistics: AI and Connectionist Models of Human Language
Processing ed. by T. Dijkstra, & K. de Smedt, 221-244. London: Taylor &
Francis.
76 GERARD STEEN
Gentner, D. 1982. "Are Scientific Analogies Metaphors?" Metaphor: Problems

and Perspectives, ed. by D. Miall, 106-132. Brighton: Harvester.
1983. "Structure-Mapping: A Theoretical Framework for Analogy".
Cognitive Science 7. 155-170.
1989. "The Mechanisms of Analogical Learning". Similarity and
Analogical Reasoning ed. by S. Vosniadou, & A. Ortony, 199-241.
, & M. Jeziorski. 1993. "The Shift from Metaphor to Analogy in Western
Science". Metaphor and Thought, second edition, ed. by A. Ortony, 447-
480. Cambridge: Cambridge University Press.
Johnson, M. 1987. The Body in the Mind. Chicago: University of Chicago Press.
Lakoff, G. 1987. Women, Fire, and Dangerous Things: What Categories Reveal
about The Mind. Chicago: University of Chicago Press.
1993. "The Contemporary Theory of Metaphor. Metaphor and Thought,
second edition, ed. by A. Ortony, 202-251. Cambridge: Cambridge
University Press.
Chicago Press.
, & M. Turner. 1989. More Than Cool Reason: A Field Guide to Poetic
Metaphor. Chicago: University of Chicago Press.
Miller, G.A. 1993. "Images and Models, Similes and Metaphors". Metaphor and
Thought, second edition, ed. by A. Ortony, 357-400. Cambridge: Cambridge
University Press.
Perfetti, CA., & M.A. Britt. 1995. "Where Do Propositions Come From?"
Discourse Comprehension: Essays in Honor of Walter Kintsch, ed. by C.A.
Weaver, HI, S. Mannes, & C.A. Fletcher, 11-34. Hillsdale, N.J.: Lawrence
Erlbaum Associates.
Reinhart, T. 1976. "On Understanding Poetic Metaphor". Poetics 5. 383-402.
Richards, I.A. 1936. The Philosophy of Rhetoric. London: Oxford University
Press.
Steen, G.J. 1999. "Metaphor and Discourse: Towards a Linguistic Checklist for
Metaphor Analysis". Researching and Applying Metaphor, ed. by L.
Cameron, & G. Low, 81-104. Cambridge: Cambridge University Press.
To appear. "Analyzing Metaphor in Literature: With Examples from
William Wordsworth's 'I wandered Lonely as aCloud'". Poetics Today 20.
Turner, M. 1987. Death Is the Mother of Beauty: Mind, Metaphor, Criticism.
Chicago: University of Chicago Press.
, & G. Fauconnier. 1995. "Conceptual Integration and Formal Expression".
Metaphor and Symbolic Activity 10. 183-204.
Vosniadou, S., & A. Ortony. 1989. "Similarity and Analogical Reasoning: A

Synthesis". Similarity and Analogical Reasoning, ed. by S. Vosniadou, &
A. Ortony, 1-17. Cambridge: Cambridge University Press.
A TYPOLOGY OF MOTIVATION FOR CONCEPTUAL METAPHOR
CORRELATION VS. RESEMBLANCE1
JOSEPH. E. GRADY
George Mason University
1. Conceptual metaphors and experiential motivation

One of the most basic principles of the theory of metaphor outlined by
Lakoff & Johnson in Metaphors we live by (1980) is that there are conventional
metaphoric associations, or mappings, between some concepts, but not between
others. For instance, emotional unresponsiveness is mapped onto coldness in the
domain of temperatures, but not onto warmth, and not onto any number of
properties in other domains, such as width, monetary value, or innateness.
So, for instance, we say that an unsympathetic person has a cold and
unfeeling demeanor, but not a wide and unfeeling demeanor or a precious and
unfeeling demeanor.
(1) a. He has a very cold and unfeeling demeanor.

b. ? He has a very warm and unfeeling demeanor.
c. ? He has a very wide and unfeeling demeanor.
d. ? He has a very precious and unfeeling demeanor.
Unlike la, examples l b-d are difficult to interpret as statements about an

individual's temperament. Speaking more generally, it is often the case that a
given pairing of concepts invokes no conventional or easily determined meta
phoric mapping. The following example from Mark Turner's Reading Minds
(1991:154) illustrates this point.
(2) ? The moon is a monkey wrench.
Although we could probably find a way to impose meaning on this statement (as
Turner does), the interpretation would hardly be predictable, and would demand
a good deal of creativity.
If there is one set of figurative correspondences which is conventional, and
is manifested in numerous linguistic expressions, and another set of pairings
which do not have this status, then cognitive linguists should be concerned with
80 JOSEPH GRADY
finding the principles which cause some metaphors to be in the conceptual

repertoire, and others not to be. Unless this distinction is arbitrary, it should be
possible to account for it.
Table 1 lists some examples of conventional metaphors, which underlie
various linguistic examples, and also some unconventional pairings of concepts,
which are difficult to make sense of.
Conventional Unconventional
NORMAL IS STRAIGHT NORMAL IS LARGE

SIGNIFICANT IS LARGE SIGNIFICANT IS TASTY
PLEASING IS TASTY PLEASING IS HEAVY
DIFFICULT IS HEAVY DIFFICULT IS STRAIGHT
Table 1 : Conventional and unconventional metaphoric pairings
There are various kinds of evidence which show that the distinction is not entirely
arbitrary - that is, not simply a matter of historical accident. For instance, the
recurrence of the same metaphoric patterns across broad samples of unrelated
languages argues against the view that the conventionality of particular metaphors
is arbitrary. The examples in (3) illustrate one such pattern:
(3) Zulu -khulu "big; important"

Hawaiian nui "big; important"
Turkish byk "big; important"
Malay besar "big; important"
Russian krupnij "big; important"
In each of these languages, a term which literally refers to physical size may also
refer metaphorically to degree of importance. This is a conceptualization with very
wide cross-linguistic distribution.
If there is a principle behind the conventionalization of certain metaphors,
that principle must logically be due to either something about the human organism
- for instance, the metaphoric correspondences are innate and hardwired into our
cerebral structure - or the patterns must arise from something about our experi
ences, or possibly both. Presumably, even a cause which lies in the realm of our
experiences must relate in some way to the structures of our brains and bodies,
since these structures constrain our experiences so definitively. The recurrence of
particular metaphorical patterns across cultures is so striking that any experiences
which could give rise to these metaphors must be fundamental to human life in
A TYPOLOGY OF MOTIVATION FOR CONCEPTUAL METAPHOR 81
general, rather than based on any particular, local, culturally bound types of
experience.
Metaphor researchers such as Lakoff, Johnson, Sweetser, Turner, Gibbs and
Brugman have been unanimous in invoking the principle of 'experiential
motivation' rather than arguing that metaphors are arbitrary or innate. This
position contrasts sharply with other theories of metaphor. A striking statement
of an opposing viewpoint is Searle's argument that coldness "just is" associated
with being unemotional:
(4) I think the only answer to the question, "what is the relation between cold things
and unemotional people that would justify the use of 'cold' as a metaphor for lack
of emotion?" is simply that as a matter of perceptions, sensibilities, and linguistic
practices, people find the notion of coldness associated in their minds with lack
of emotion. The notion of being cold just is associated with being unemotional.
(Searle, 1979:267)
Searle appears to dismiss the enterprise of trying to find motivations for

metaphors, and even the idea that there is any principled reason why some
metaphorical conceptualizations arise and make sense to us while others do not.
By contrast, I cite the following passage from Lakoff & Johnson (1980) as
a concise statement of the position on experiential motivation within the theory
of conceptual metaphor: "We feel that no metaphor can ever be comprehended or
even adequately represented independently of its experiential basis ..." (Lakoff &
Johnson 1980:19). The typical example of experiential motivation referred to in
conceptual metaphor literature is the grounding for MORE IS UP (e.g., "Drunk
driving arrests are up this year"). In this case, the motivation is a straightforward
correlation between the two concepts: as objects or substances accumulate in
greater quantities, their level often rises - consider for instance the top of a stack
or the level of fluid in a container.
2. Death is a thief
Many other metaphors cited in the literature, however, are much harder to
account for in terms of such simple correlations as the one between quantity and
vertical elevation. This observation has been one of the motivations for the
development of the theory of 'primary metaphors ', which holds that certain basic,
low-level metaphorical correspondences have a privileged status, and are the bases
for other metaphorical expressions and conceptualizations (Grady et al. 1996;
Grady 1997a, etc.)2. These basic conceptual associations, which are excellent
predictors of how and whether linguistic data may be interpreted, are also the
82 JOSEPH GRADY
metaphors which are most clearly grounded in aspects of our experience (Grady
1997b).
As an example of a metaphor which is not primary in this sense, consider
DEATH IS A THIEF, discussed by Mark Turner in Reading Minds.
(5) [D]eath robbed him of his life. (Turner 1991:174)
This metaphor, unlike MORE IS UP, is not based on any common aspect of our
experience. There is certainly no recurring scene we all experience involving both
death and a thief. In fact, many of us who understand the metaphor may have no
direct experience whatsoever with thieves, and even our indirect experience of
them, e.g. in books and films, would not motivate a tight association between
thieves and death - they are most closely associated with stealing, not murder.
An additional fact which will be relevant to us about this metaphor is that
the sentences and expressions which it appears to motivate are extremely similar
to expressions about target concepts other than death. In fact, anything which we
greatly appreciate - such as our own experiences of hope, happiness, comfort, etc.
- can be metaphorically "robbed" from us, just as life can. Whatever causes us to
metaphorically lose these valued elements of our experience can be cast as a thief:
(6) a. All hope and comfort have been robbed from me in this awful place.
b. He broke her heart and stole her happiness.
In each of these cases it seems that the thief as an entire person is not relevant -
instead it is merely the thief as the entity responsible for our loss that figures in the
mapping. For this reason, expressions like the following are very difficult to
interpret.
(7) a. ? Worry has robbed me of my piece of mind, and he [i.e. Worry] is tall.
b. ? A lifetime of poverty has robbed her of her hopes and dreams, and it [i.e. the
lifetime of poverty] is fatigued.
Clearly, the metaphorical conceptualization of these situations which allows us to

speak of "thieves" and "robbing" does not involve a rich understanding of the
thief as an individual with physical characteristics, feelings, and so forth. Of
course we are capable of enriching our figurative image of Death-as-Thief with
any degree of detail, including his physical appearance; but the conventional
mapping does not include such elements.3 In the simpler, conventional mapping
the thief is merely the agent of loss. Note that primary metaphors like DIFFICULT
IS HEAVY and PLEASING IS TASTY do not involve this sort of very partial
projection. In these cases virtually any lexical item which refers to the source
concept can refer metaphorically to the target concept, yielding an expression
which is interpretable according to the given conceptual pattern even if it is
lexically unconventional - a delicious idea, a succulent idea, a weighty task, a ten-
ton task. That is, while these mappings are very schematic, they are relatively
exhaustive within their limited range. As 7 illustrates, the same cannot be said of
DEATH IS A THIEF. Since the mapping between thieves and agents of unwanted
change is so selective, or to put it more strongly, so narrowly restricted, we are
better justified in stating the mapping at the level of the projections which actually
occur.
From the considerations above we can conclude that possession is the key
metaphorical concept here - abstract entities which we value are understood as
prized possessions, and whatever causes us to lose those things fits into this
conceptualization as a thief, or other individual who takes away from us what we
hold dear. The expressions mapping valued experiences onto possessions need not
refer to thieves, however, as we see in 8:
(8) a. My most precious possession is my health.

b. I still treasure those memories.
Thus the linguistic evidence suggests that a metaphor along the lines of VALUED
ASPECTS OFEXPERIENCE ARE PRECIOUS POSSESSIONS lies behind the conceptualiza
tion of death as a thief. This less specific mapping is also much easier to account
for than DEATH IS A THIEF in terms of a plausible, direct association between the
source and target concepts. There are strong, recurrent correlations between
physical and emotional aspects of our experience as we interact with objects. For
instance, we may feel a strong sense of pleasure when we acquire certain objects,
satisfaction as we hold them, and loss or grief when they are taken away. As long
as life, hope, happiness, and the love of others are appreciated on an emotional
level they can be construed as metaphorical treasures, vulnerable to theft by
metaphorical thieves.
Contrast this more schematic metaphor with DEATH IS A THIEF: it would be
unsatisfying (and, as we have seen, unnecessary) to invoke our few experiences
with actual thieves as motivations for the metaphor, especially since so little of
those experiences is relevant to the expressions. Fear, for instance, might be a
typical reaction to an encounter with a thief, but does not figure in expressions like
6 and 7. The recurring experience types mentioned above, however, are plausible
motivations for the VALUED OBJECTS METAPHOR, and instances of what have been
termed 'primary scenes' (see Grady 1997 and Grady & Johnson, to appear).
84 JOSEPH GRADY
The types of interactions with objects which ultimately motivate the

conceptualization of death as a thief occur in all sorts of settings and transcend
particular experiential frames such as restaurant dining or highway driving - the
precious object could be a wedding ring, a pen we like to write with, a favorite
book, a photograph, a seashell, a toy, etc. And, like other types of experience
which appear to underlie conventional metaphoric mappings, our experiences with
valued objects correlate some aspect of our perception of or interaction with the
world with some aspect of our cognitive response to the world. Other such
experience types include lifting a heavy object and experiencing strain, tasting a
sweet object and experiencing pleasure, judging an object to be flawed due to
irregularities in its shape, and paying particular attention to a larger object
(because of its potentially greater significance as an obstacle, threat, reward, etc.).
These are recurring experiences which plausibly motivate the conventional
metaphors listed in Table 1.
Experiences with valued objects constitute plausible bases for a metaphor
like VALUED ASPECTS OF EXPERIENCE ARE PRECIOUS POSSESSIONS since they
involve tight, recurring correlations between the emotional and physical
dimensions of our interactions with possessions.
(9) Experience types which motivate VALUED ASPECTS OF EXPERIENCE ARE PRECIOUS
POSSESSIONS
a. Gaining a possession and feeling happy
b. Holding a possession and feeling content
c. Losing a possession and feeling sad
This basic metaphor, in turn, licenses more particular conceptualizations such as

DEATH IS A THIEF.
The reason I have taken the time to discuss this particular case is that it
illustrates the more general principle that specific or complex metaphors, or ones
which have been elaborated with rich detail, often owe their existence to mappings
at a more fundamental level, and this is where we should look for experiential
motivation.
3. Experiential correlation as a motivation for primary metaphors

When we investigate the apparent experiential bases of a number of primary
metaphors we find that the experiences which could plausibly give rise to them are
similar in scale and structure to the experience types just discussed. Namely, a
recurring 'primary scene', which can be characterized at a very local and
schematic level, involves a tight correlation between two dimensions of
experience - typically with one more directly related to sensory input than the
other. Typical of these scenes is that they are elements of universal human
experience - basic sensori-motor, emotional and cognitive experiences which do
not depend on the particulars of culture.
Some other metaphors at the primary level include:
(10) [CAUSAL] ORGANIZATION IS PHYSICAL [PART-WHOLE] STRUCTURE

STRONG DESIRE IS HUNGER
ACHIEVING A PURPOSE IS REACHING A DESTINATION
ORGANIZATION IS PHYSICAL STRUCTURE (e.g., "Our monitoring system has

unraveled over the past week"), which involves a mapping between physical part-
whole structure and the logical and causal relationships which we refer to very
generally as organization, is plausibly accounted for in terms of a correlation
between physical interaction with complex objects and the formation of cognitive
representations of their causal structure. For instance, our experience of table legs
includes both perceptual information about shape and conceptual information,
informed by our understanding of gravity, regarding their causal role in supporting
tabletops. As motivation for STRONG DESIRE IS HUNGER - e.g., "Our team is very
hungry for a victory" - we can point to the correlation between the physical
sensation of hunger and the focused conscious desire (for food) which accompa
nies it. Finally, ACHIEVING A PURPOSE IS REACHING A DESTINATION appears to be
based on the correlation between arriving at physical landmarks and achieving
purposes (see Johnson 1987 for a discussion of this metaphor and its basis in
experience.)
The kinds of motivations I have just mentioned sound strikingly like
metonymic relationships between aspects of experienced scenes (which could be
represented in schema theory by close network links between schemas embedded
within larger scenarios). In fact primary metaphors and their motivations should
be especially fertile material for research on the relationship between metaphor
and metonymy. A number of researchers (e.g., Goossens 1995) have explored this
relationship and pointed out that the two often appear to be closely related, or even
hard to distinguish, both in the kinds of conceptual relationships they comprise
and especially in the kinds of situational relationships that motivate them. While
it is clearly metaphoric to cast difficulty as heaviness in cases where no physical
burdens or physical weights are relevant, the origins of the conceptual association
between the physical assessment of weight and the affective experience of
exertion certainly bear comparison to the type of frame-internal relations typical
of metonymy.
Chris Johnson and I have argued (Grady & Johnson 1998; Grady & Johnson
to appear) that the characterization of the primary scenes which underlie primary
86 JOSEPH GRADY
metaphors can also help account for other aspects of language, including the
organization of basic semantic fields and patterns in children's acquisition of
semantic and grammatical forms. For instance, there is evidence to suggest that
children acquire the Instrumental sense of with later than other senses, and that
this pattern is due to the relative complexity of the semantics of Instrumentality.
In particular, Instrumentality cannot be defined with respect to an individual
primary scene - unlike the Accompaniment sense of with, for instance, which may
refer simply to co-location with another person (as in, "He's with Paul"). Instead,
Instrumentality must involve at least possession plus the performance of a
particular action. The possible role of primary scenes in motivating acquisition
patterns is a fascinating topic in itself, but not one we can explore here.
The title of this paper refers to a typology of motivations for conceptual
metaphors, and the cases we have considered so far fall into a category which I
will call 'correlation metaphors'. As we have seen, each case, when examined at
the appropriate level of locality and specificity, involves a correlation between
distinct dimensions of experience. Metaphors at this level, which arise from
primary scenes, are characterized by a number of interesting features; in the
present context I will mention only one of these. Much of the recent literature on
conceptual metaphor has suggested that target concepts are abstract in the sense
of being sophisticated or complex intellectual constructs - e.g., "Conceptual
metaphors arise when we try to understand difficult, complex, abstract, or less
delineated concepts, such as love, in terms of familiar ideas, such as different
kinds of nutrients" (Gibbs 1994:6). I have found instead that the target concepts
of primary metaphors refer to basic cognitive processes, and are typically no more
sophisticated or distant from our direct experience than corresponding source
concepts.
For instance, the metaphor MORE IS UP has as its target the concept of
quantity, which we judge instantaneously in many situations. If quantity is judged
instantly and perceived as a simple, scalar parameter, then it is not a complex
concept (whatever the complexities of the neural mechanisms needed to calculate
it, or of giving it a satisfactory definition). The primary metaphor DESIRE IS
HUNGER maps hunger onto desire, a basic cognitive-emotional state which again
we do not conceive as having a complex internal structure, and which we need no
help to recognize or understand. In a similar manner the other primary metaphors
mentioned in (10) refer to fundamental sorts of cognitive experience, such as the
(in many cases automatic and unconscious) inference that some events are
causally connected with others, and the immediate feeling of satisfaction
(probably the result of hormonal activity) when we have achieved a simple goal.
If there is an advantage to be gained from entertaining metaphorical
conceptualizations of some of the simplest elements of conscious experience, one
possibility is that we are more efficient at the conscious manipulation of images -

i.e. mental representations of sensori-motor experience, not necessarily visual -
than at dealing with such notions as Quantity or Desire per se. Even though the
ability to attend to and judge quantity and the tendency to experience desire seem
to be fundamental aspects of cognitive function, these functions may take place
at a level of cognition whose operation is not directly accessible to consciousness.
In order to manipulate them at the conscious level it may be necessary to tie these
elements of mental experience to specific sensory images. This idea fits well with
findings about basic level categorization, for instance - the types of concepts
which people find easiest to store, describe, and name.
4. Resemblance metaphors
Despite the value of experiential correlation in accounting for many basic
metaphors, not all metaphors are plausibly motivated in this way.
4.1 "Achilles is a lion "

Consider a statement like "Achilles is a lion," a classic example of a type
cited regularly in philosophical and psychological studies of metaphor. It is
difficult to imagine how the conceptualization underlying this statement could
arise from recurring correlations in experience. Most obviously, many of us who
might use and understand such an expression have no personal experience with
lions. But even if we allow for the importance of indirect experience in forming
schemas, e.g. learning about lions by reading about them, it is still problematic to
identify correlation as a motivating factor for the metaphor.
For a start, it would be difficult to name any concepts that are correlated
here, in a way that could give rise to the metaphor. Is bravery correlated with
"lionhood"? If so, what content does the concept of lionhood contain? Presum
ably, it includes all the information in our shared schema for lions (cf. Lakoff &
Turner 89:195), including their appearance, the fact that they live in prides, the
fact that they sleep much of the day, and so forth. None of these features, however,
is relevant to the metaphorical lionization of Achilles (or any other courageous
person). For this reason, not to mention the fact that courageousness is part of the
lion schema itself, it is awkward to speak of a correlation between courage and
"lionhood," or between courage and any other particular features of the lion
schema, which could be the motivation for the metaphor BRAVE PEOPLE ARE
LIONS.
To understand even more clearly that correlation is not a direct motivation
for this metaphor, let's review the kinds of correlation which form the basis for
metaphors like MORE IS UP and PURPOSES ARE DESTINATIONS. In each of these
cases, two quite distinct concepts are cognitively linked because they are tightly
88 JOSEPH GRADY
correlated in certain recurring types of experience. Vertical elevation varies

directly with quantity in many situations, though our means of judging these two
parameters are very different. (We can judge quantity in the absence of vertical
elevation, and vice versa.) We often experience a sense of gratification as a
consequence of arriving at a particular spatial location, but our means of
determining location and our emotional capacity for feeling gratification are
distinct, too, of course. Notice, by the way, that there are many times when we
move through space to a new location but do not feel this same sense. For
instance, on some occasions I might accidentally move to the wrong location, or
I might be pushed to a place I had no intention of going. In these cases, the
distinction between arriving at a spatial location and achieving a purpose is plain.
There is no way to analyze the conceptual correspondence underlying
"Achilles is a lion" as an association of this sort, between fundamentally distinct
concepts which are correlated in some type of experience. Consider "Brave people
are lions," "Acting courageously is acting like a lion," "Courage is the instinctive
fearlessness of a lion," etc. No matter how we phrase the metaphor, it seems, the
cues that prompt us to attribute bravery (and which relate to some aspect of
perceived behavior) are the same for the people and the lions. The simplest
explanation for the metaphor is that, in some sense, brave people and lions are
perceived to resemble one another.
4.2 The 'similarity theory' vs. the 'resemblance hypothesis'

In their discussion of "Achilles is a lion" Lakoff & Turner note that the
courage of a lion is itself a metaphorical projection from a human character trait
onto an aspect of the lion's instinctive behavior. If so, does this observation defeat
the suggestion that lions and brave people bear a perceived resemblance? Not at
all. Lakoff & Turner's discussion of the "Great Chain Metaphor" offers important
insights into why the statement is taken as referring to Achilles' character, rather
than his hair color or gait, for instance, but does not address the question of why
the association between people and lions would arise in the first place, which is
the issue we are considering here. Why do we project human bravery onto aspects
of lions' instinctive behavior, and vice versa, rather than associating brave people
with chickens or goldfish, for instance? The most plausible explanation is that we
perceive something in common between stereotypical lions, whatever the basis for
this schema, and brave people. Lions and courageous people both (appear to)
confront dangerous opponents without fear.
For centuries, various scholars who have treated the phenomenon of
metaphor - including Aristotle, in the Poetics - have suggested that metaphors are
basically expressions of the similarity between two concepts. Recent researchers
in the cognitive linguistic tradition have argued compellingly against this 'simila-
rity theory' of metaphor (e.g. Lakoff & Turner 1989:198). Simply put, there is
often no literal similarity to point to between concepts which are associated by
metaphor. For instance, it is difficult to see how a metaphor like HAPPY IS UP
(Lakoff & Johnson 1980), as in "She is in high spirits," could be based on an
objective similarity between mood and vertical elevation. Nor is coldness "simi
lar" to lack of emotion, as Searle acknowledged. Yet these concepts are metapho
rically equated. In short, the similarity theory fails for a number of important cases
(and in particular, for metaphors based on correlation).
Because of this controversy, I emphasize that I am not advocating the
discredited similarity theory, which may at this point be a straw man in any case.
My proposal does not depend on any literal similarity between brave people and
lions. It seems inevitable, though, to conclude that the metaphorical association
between them - involving projection in whichever direction - is most plausibly
based on the perception of common aspects in their behavior. I will call this
proposition the 'resemblance hypothesis,' in order to distinguish it from the
similarity theory, and to highlight the role of our perceptions and representational
schemas, as opposed to facts about the world.
There is some precedent within conceptual metaphor theory for allowing
that there can be a sort of metaphorical association based on (the perception of)
shared features. Lakoff & Turner (1989:90) described the phenomenon of 'image
metaphors', offering as an example the mapping of a woman's waist onto an
hourglass, made possible "by virtue of their common shape". In Lakoff & Turner's
view, this kind of metaphor has a special status, since conceptual structure and
inferences are not mapped from one domain to another. Instead the source and
target of the metaphor share some feature in a single perceptual domain, such as
color or shape. Since features of lions other than their alleged courage are not
projected onto brave people - e.g., there is nothing about a brave person which
corresponds to the lion's tawny coat, or to its habit of sleeping most of the day -
we might point out that here too there is a very limited correspondence, which we
might even hesitate to call a mapping. "Achilles is a lion" is obviously not an
image metaphor, since it makes no claims about Achilles' physical form, but it
may reflect a type of very limited conceptual projection, in the same way that
image metaphors do.
As we have seen, the correlation metaphors considered in previous sections
are best accounted for in terms of co-occurrence, rather than resemblance. For
instance, achieving an objective and arriving at a location do not share a feature
which makes them suitable as a source-target pair; neither do quantity and
elevation. (In both cases we might say that there actually is a shared feature:
punctual aspect in the first case and scalar structure in the second, but while these
aspects of "superschematic" shared structure are probably necessary for the
90 JOSEPH GRADY
formation of metaphoric connections, they are not sufficient motivations for the
respective pairings. If they were, then any punctual experience, such as breaking
a dish or blowing out a candle, should stand metaphorically for achieving an
objective, and any scalar phenomenon, such as the blueness of the sky or pitch of
an acoustic signal, should serve as a source concept for quantity.) Resemblance
is not the basis for the sorts of entrenched mappings which prompted the
development of conceptual metaphor theory. If it is the basis for the conceptual
ization underlying "Achilles is a lion," then this is a reason to consider this
metaphor different in kind from those which are derived from recurring
correlations in experience.
4.3 A network model

There is a simple network model which helps illustrate the difference
between resemblance and correlation metaphors. If we think of metaphors as
patterns of association within activation networks then primary metaphors could
be characterized as links between distinct concepts, perhaps based on numerous
experiences where the concepts are tightly correlated and therefore simultaneously
activated. This pattern is schematized in figure 1.
Figure 1: Schematic network representation of a correlation-based metaphor
The concepts PILE, QUANTITY and ELEVATION are used as examples. Lower nodes
represent features of objects at higher nodes. The node at the top of the figure
represents the concept of a pile - a conceptualization in which quantity and vertical
dimension are correlated. The dashed line represents the association which is the
basis of the metaphor MORE IS UP.4
A metaphor like "Achilles is a lion," on the other hand, would have a
different kind of representation. In figure 2 the circled section represents
overlapping activation - in this case, activation of the notion of courage. The
dashed line represents the association between lions and brave people, based on
the feature shared by their respective schemas.
LION BRAVE PERSON
Figure 2: Schematic representation of a simple resemblance metaphor
We can draw no such diagram for MORE IS UP or ACHIEVING AN OBJECTIVE IS

ARRIVING. These concepts do not share a feature which motivates the mapping
between them; they are instead linked by co-occurrence (as in figure 1). If a
certain configuration can represent one metaphor but not another, this suggests
that there is a substantive difference between the two metaphor types.
5. 'GENERIC-IS-SPECIFIC' metaphors
There is one more type of relationship between concepts, besides correlation
and resemblance, which might motivate a metaphoric association between them.
In More than cool reason, Lakoff & Turner (1989:162) state that, "There exists
a single generic-level metaphor, GENERIC IS SPECIFIC, which maps a single
specific-level schema onto an indefinitely large number of parallel specific-level
schemas that all have the same generic-level structure as the source domain
schema." They illustrate the pattern with discussions of several Asian proverbs,
including "Blind blames the ditch." The situation depicted in the proverb, they
propose, instantiates a more general schema, in which a person blames his own
mistakes on circumstances he should have anticipated. We understand the
meaning of the proverb by recognizing the relationship between the particular
scenario and the more generic schema. (The metaphorical correspondence between
vision and understanding is also evident here, of course, and Lakoff & Turner
point out that these proverbs often rely on conventional mappings as well as the
GENERIC-IS-SPECIFIC structure.)
We probably do not want to treat GENERIC IS SPECIFIC as a metaphor per se,
if we would like to reserve the term for particular figurative pairings of concepts.
Nonetheless, it is worth considering whether this type of metaphorical pattern
might add to the typology we have developed so far: there may be metaphors
based on the 'isa' relationship, instantiation. Other cases where the source concept
appears to be a specific instance of the more generic target concept include RISK-
92 JOSEPH GRADY
TAKING IS GAMBLING ("A career change is a high-stakes gamble") and CO

OPERATIVE ACTIVITY IS MUSICAL HARMONIZING ("There's been harmony in the
family lately, thank goodness").
Any discussion of categories and instantiations in the context of a metaphor
study must call to mind Glucksberg & Keysar's position (e.g. 1993:408) that
"metaphors are class inclusion assertions." Glucksberg & Keysar's analysis of
statements like "my job is a jail" holds, in part, that the source term (or vehicle)
refers to a broad category of objects - in this case, any "involuntary, unpleasant,
confining, punishing, unrewarding situation" (1993:414). Glucksberg, Keysar,
Lakoff and Turner would probably all agree that the specific concept jail may
stand (metaphorically) for the generic category of involuntary, unpleasant,
confining, punishing, unrewarding situations, that ditch may stand for threatening
circumstances, and, more generally, that prototypical cases often stand metaphori
cally for generic categories.5
Returning to the question of whether we must add a new category to our
typology, note first that it is difficult to make a clear distinction between GENERIC-
IS-SPECIFIC metaphors and metaphors of the resemblance type. For example, if
"Achilles is a lion" is motivated by perceived resemblance between the behavior
of a brave man called Achilles and the stereotypical behavior of a lion, then we
might argue that the lion stands for the more generic category of brave things -
i.e. that Achilles and the lion share a generic-level representation, along the lines
of "COURAGEOUS BEINGS." (Equivalently, one could argue that the behavior of
courageous people and lions are instances of a more general schema for
courageous behavior.)
On either analysis a particular entity is mapped onto another entity with
which it shares salient perceived features, and therefore, an identical representa
tion at a higher level of generality. Figure 3 sketches this situation. (Here, lower
nodes represent instances of categories at higher nodes.)
The lines represent association between concepts and show that this association
can be traced along either of two routes in both cases - either a direct association
or an association by way of a shared underspecified, or generic, representation. In
essence, then, the existence of Lakoff & Turner's GENERIC-IS-SPECIFTC pattern is
evidence against a strong "anti-abstractionist" position: the generic scenario is an
abstraction over a range of more particular cases, which are easy to map onto one
another precisely because of this shared structure. (This structure also foreshadows
the 'generic space' in Fauconnier & Turner's theory of 'conceptual blending'.)
While the correlation cases do not appear to involve abstraction - and have
provided compelling evidence against a strong abstractionist position - the
resemblance cases and GENERIC-IS-SPECIFIC cases arguably do.
The cases considered so far suggest a fairly neat distinction between
correlation metaphors on one hand and resemblance or GENERIC-IS-SPECIFIC
metaphors on the other. The distinction seems to be challenged, though, when we
consider that supporting a heavy burden is an instance of enduring a difficult
situation, arriving at a destination is an instance of achieving a purpose, being
hungry is an instance of experiencing desire, and so forth. In other words, the
primary, correlation-based metaphors discussed in earlier sections might somehow
be analyzable as GENERIC-IS-SPECIFIC metaphors. And since we have seen that
GENERIC-IS-SPECIFIC metaphors can be construed as resemblance metaphors,
perhaps the primary metaphors should after all be accounted for based on shared
aspects of schemas. For instance, enduring a family crisis resembles supporting
a heavy weight in that both provoke feelings of stress and displeasure. An
important day resembles a large object in that both tend to command our attention.
Does the typology actually collapse to a single category?
No, there is still a basis for preserving the distinctions. First, we will briefly
review the relationship between GENERIC-IS-SPECIFIC and resemblance metaphors.
Once again, simple diagrams help clarify the argument. In each case the arrow
points from source to target:
Figure 4a: Risk-taking is gambling (Generic-is-specific)

94 JOSEPH GRADY
Figure 4b: Achilles is a lion (Resemblance)
In figure 4a, the GENERIC-IS-SPECIFIC case (RISK-TAKING IS GAMBLING), the source

is a specific instance of the target, a generic schema.6 The resemblance metaphor
structure represented in figure 4b ("Achilles is a lion") is very similar except that
here the target is taken to be a specific instance of the generic schema; a different
instance serves as the source. Cases a and b are fundamentally similar, and differ
only with respect to which levels of specificity are highlighted. We can show that
they are variants by considering hypothetical metaphors like "COURAGEOUS
BEINGS ARE LIONS" and "MAKING A RISKY CAREER MOVE IS GAMBLING," which
would look like GENERIC IS SPECIFIC and resemblance metaphors, respectively.
Figure 4c: Size is importance (Correlation)
A correlation metaphor (IMPORTANCE IS SIZE), as represented in 4c, looks

somewhat different. Here there is a particular feature of the source concept, not
itself the basis for a resemblance link, which is relevant to the metaphorical
mapping. In cases a and b there is no particular feature of the source image which
is conventionally associated with the target concept. We have seen, for instance,
that lions' appearance and sleep habits are not conventionally mapped onto
courageous people. While we could try to treat case "c" the same way as "Achilles
is a lion" - i.e. by identifying the metaphor as "An important day is a large
object" and arguing that it is based on the shared feature important (i.e.
"commands our attention"), as in figure 4d,
Figure 4d: "An important day is a large object"

we would be missing an important observation: there is more than one aspect of

our understanding of large objects that is relevant to this conceptualization, and
in fact the relationship between two of these aspects - physical size and
subjective importance - constitutes a pairing with special significance, one that
enters our conceptual repertoire as a projection pattern that may serve as the basis
for more elaborate mappings.
In sum, "GENERIC-IS-SPECIFIC" and "resemblance" do appear to be
alternative ways of construing what is essentially the same conceptual relation
ship, differentiated from each other only with respect to which link is profiled, to
borrow a term from Cognitive Grammar. Correlation metaphors, on the other
hand, involve salient relationships between aspects of single concepts, of a kind
not evident in the other sorts of metaphor. These relationships derive from
correlations within the recurring experience types that give content to those
concepts.
6. Comparison between the types of metaphor

Considering the evidence, we now have two distinct classes of metaphors -
the resemblance class, including GENERIC-IS-SPECMC metaphors, on the one hand
and the correlation-based metaphors, including primary metaphors, on the other.
This is not an elaborate typology, but it is one which involves some critical
distinctions. In addition to what has been said above, there are a number of other
significant ways in which these two classes of metaphors appear to differ.
6.1 Directionality
Some resemblance metaphors appear to violate the principle of
unidirectionality that is usually attributed to conceptual metaphors. For instance,
consider the hypothetical statements, "Einstein is the modern Pythagoras" and
"Pythagoras was the Einstein of his age," intended as comments about comparable
intellectual achievement.7 Another metaphor which appears to be based on
resemblance is DEATH IS SLEEP, as in Hamlet's "to sleep perchance to dream." (We
could argue, by the way, that this correspondence is based on a shared generic-
level schema involving inactivity.) This metaphor, too, works in reverse, as in the
expression "dead to the world," said of someone who is asleep or unconscious.
Metaphors of the GENERIC-IS-SPECIFIC type also seem to be symmetrical,
allowing projection in either direction, as we would expect from Lakoff &
Turner's description. We might respond "Blind blames the ditch" when a hasty
person blames an injury on a hammer instead of his own carelessness, and we can
also imagine using a hypothetical proverb like "Hasty blames the hammer" when
a person falls into a ditch. Either instance may serve as source to the other's target.
In this important respect, then, both resemblance and GENERIC-IS-SPECIFIC
96 JOSEPH GRADY
metaphors are like image metaphors, which work equally well in either direction:
a woman's waist can be an "hourglass;" an hourglass can have a slender "waist."
Of course all these cases involve the projection of subtly different conceptual
material depending on direction - e.g., an hourglass may be "feminized" when
its narrow portion is called a waist - but for our purposes it is sufficient to note
that a salient shared feature leads to the possibility of projection in either direction,
which is not the case where metaphors based on correlation are concerned.
Difficulty may not stand metaphorically for simple physical weight, and so forth.
6.2 Ontology
Correlation metaphors and resemblance metaphors make different demands
on the objects which serve as source and target. Resemblance metaphors may
involve correspondences between concepts of the same type, whereas correlation
metaphors link concepts of different types. For instance, weight and difficulty are
two concepts linked in the primary metaphor DIFFICULTIES ARE BURDENS (e.g.,
"Caring for an elderly relative places a heavy burden on a family"). The
phenomenon of physical weight is recognized and judged by cognitive faculties
very distinct from those which underlie the notion of difficulty - i.e. exertion,
discomfort, stress, etc. The same principle applies to the correspon-dences
between quantity and vertical elevation, between similarity and proximity,
between logical organization and physical part-whole structure, etc. In each of
these cases, the linked concepts are fundamentally distinct in the way they are
perceived and understood. In fact, it is typical of the source and target concepts of
primary metaphors that they are characterized by very distinct properties. For
instance, source concepts tend to involve sensory content whereas target concepts
involve our cognitive responses to sensory input (see Grady 1997.) Resemblance
metaphors, on the other hand, may involve objects of identical or nearly identical
types, as we have seen. One state of inactivity is mapped onto another; one type
of physical mishap is mapped onto another, one intelligent person is mapped onto
another, etc.
6.3 Conventionality
Because the human imagination is boundless in its capacity to impose
resemblance on disparate objects, resemblance metaphors would appear to be
nearly unconstrained. The moon and a monkey wrench surely do have something
in common, at least in the way we perceive them. (As Turner points out,
1991:154, both "can expand and contract".) Our ability to perceive resemblance,
of course, is constrained by the cognitive mechanisms of perception, possibly in
cluding the structuring role of the "image schema." (See, e.g., Johnson 1987,
Lakoff 1987, Turner 1991). Nonetheless, this constraint still leaves open a nearly
infinite range of potential pairings of concepts and images which somehow remind
us of each other. The same is true of GENERIC-IS-SPECIFIC metaphors, which may
involve links between, as Lakoff & Turner put it, "an indefinitely large number of
parallel specific-level schemas."
Correlation-based metaphors, on the other hand reflect specific, recurring
experience types, and are therefore much more constrained. This is especially clear
when we look at primary metaphors, which have direct experiential motivation.
Similarity, for instance, corresponds metaphorically to Proximity, and not to other
sorts of spatial relations. The same associations arise in language after language -
apparently because the experience types which motivate them are so basic that
they characterize human life in all times and places - and these associations fall
into well-defined sets of patterns.
7. Conclusion
Debate about the nature of metaphor has been sharp and long-running in
several different scholarly traditions. Part of the reason may be that researchers
have pointed to objects of different kinds in support of their own preferred
definitions. If we make a distinction between these types many of the controversial
issues about metaphor might be resolved. Supporters of versions of the similarity
theory, i.e. researchers who refer to shared features and structural analogy as the
basis for metaphor, would have to acknowledge that there are metaphors which
are not based on resemblance or perceived resemblance - the correlation
metaphors. Conceptual metaphor theorists, who are used to arguing that similarity
is not the basis for metaphors, might allow that there is a class of linguistic and
conceptual phenomena which is motivated by the perception of a resemblance
between distinct objects, a resemblance which would, of course, have to be
described in terms of cognitive mechanisms of perception and categorization.
Other claims about metaphors - besides the extent to which similarity plays
a role in motivating them - also fall out from the two positions, and might be
resolved by recognizing a taxonomy. For instance, the traditional understanding
of metaphor as an exceptional, creative product of imagination may have resulted
from a focus on metaphors of a particular type. Many of the metaphors which have
appeared in traditional philosophical discussions of metaphor have fallen into the
class of resemblance metaphors. If the resemblance hypothesis is correct, then
many expressions like "Achilles is a lion" or "Man is a wolf," which appear over
and over in these discussions, are based on perceived parallelism between their
source and target concepts - or to put it another way, the perception that there is
a superordinate category which includes both concepts. I propose that the
relatively unconstrained nature of resemblance and GENERIC-IS-SPECIFIC
metaphors underlies various scholarly claims that metaphor is ungoverned by rules
98 JOSEPH GRADY
or principles, and that it is a tool for adding originality and color to texts (to be
used with abandon or with caution depending on the authority one consults). The
finite list of conventional, highly-motivated associations proposed by scholars like
Lakoff & Johnson, on the other hand, might be associated with metaphors based
on correlation.
Scholars such as Lakoff & Turner have already opened the door to the
classification of metaphors based on distinct properties. Their proposals regarding
the "GENERIC IS SPECIFIC metaphor" and image metaphors, for instance, suggest
that metaphors can involve quite different sorts of cognitive mechanisms and
structures. The taxonomy I have suggested here follows up on proposals like
those. I argue that we can refine our sense of the different types of metaphor
further by carefully considering the motivations for these metaphors. When we do,
we arrive at a classification which seems to explain some of the long-standing
disagreements about the nature of metaphor. While it might prove to be the case
that not all metaphors fall neatly into one or the other of the categories I suggest,
the prototypical cases, and the differences between them, are clear, and the
distinctions should help guide our continuing research into the nature of
metaphorical thought and language.
Notes
1. I would like to thank the editors of this volume for their questions, comments and
suggestions, which have been particularly helpful in the preparation of this paper.
2. Primary metaphors are the same as 'primitive metaphors' (Grady et al. 1996).
3. For a discussion of how schematic, conventional metaphors are elaborated in given
instances, see Grady et al. (this volume).
4. In the very crude representations in this section I ignore a number of important issues -
perhaps chief among them, directionality, which is certainly a feature of primary
metaphors.
5. Glucksberg & Keysar are primarily interested in the nature of metaphorical statements of
the form "A is B," - including such issues as sentence ordering - which is not my focus
here. Note that many basic patterns of metaphorical conceptualization show up in other
sorts of linguistic contexts - e.g. "They have extracted some new information from the
photograph," where the source term is a verb referring to a metaphorical action.
6. Looked at another way, the diagram represents the fact that the target, risk-taking, is an
aspect or feature of the source scenario, gambling.
7. These statements might not strike some readers as metaphorical, but certainly strike others
as being so. Glucksberg & Keysar (1993:421) refer to the statement "Xiao-Dong [a
Chinese actor] is a Bela Lugosi" as a metaphor. The question here is one of degree of
metaphoricity, and individuals differ regarding where the line ought to be drawn between
metaphors per se and other sorts of reference.
References
Gibbs, Raymond W. Jr. 1994. The Poetics of Mind: Figurative Thought,
Language, and Understanding. Cambridge: Cambridge University Press.
Glucksberg, Sam, & Boaz Keysar. 1993. "How Metaphors Work.". Metaphor and
Thought, 2d edition, ed. by A. Ortony, 401-424. Cambridge: Cambridge
University Press.
Goossens, L. 1995. "Metaphtonymy: The Interaction of Metaphor and Metonymy
in Figurative Expressions for Linguistic Action". By Word of Mouth:
Metaphor, Metonymy and Linguistic Action in Cognitive Perspective, ed. by
Louis Goossens, Paul Pauwels, Brygida Rudzka-Ostyn, Anne-Marie Simon-
Vandenbergen, & Johan Vanparys, 159-174. Amsterdam/Philadelphia: John
Benjamins.
Grady, Joseph. 1997a. Foundations of Meaning: Primary Metaphors and Primary
Scenes. University of California, Berkeley, Ph.D. dissertation.
1997b. "THEORIES ARE BUILDINGS Revisited". Cognitive Linguistics 8.
267-290.
1998. "The Conduit Metaphor Revisited: Reassessing Metaphors for
Communication". Conceptual Structure, Discourse, and Language II, ed.
by J.-P. Koenig, 205-218. Stanford: CSLL
, & Sarah Taub, Pamela Morgan. 1996. "Primitive and Compound
Metaphors". Conceptual Structure, Discourse, and Language, ed. by A.
Goldberg, 177-187. Stanford: CSLI
, & Christopher Johnson. To appear. "Converging Evidence for the Notions
of Primary Scene and Subscene". Proceedings of the twenty-third annual
meeting of the Berkeley Linguistics Society. Berkeley: BLS.
Johnson, Mark. 1987. The Body in the Mind. Chicago: University of Chicago
Press.
Lakoff, George. 1993. "The Contemporary Theory of Metaphor". Metaphor and
Thought, 2d edition, ed. by A. Ortony, 202-251. Cambridge: Cambridge
University Press.
, & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of
Chicago Press.
, & Mark Turner. 1989. More Than Cool Reason: A Field Guide to Poetic
Searle, John R. 1979. "Metaphor". Expression and Meaning. 76-116. Cambridge:
Cambridge University Press.
Sweetser, Eve. 1990. From Etymology to Pragmatics: Metaphorical and Cultural
Aspects of Semantic Structure. Cambridge: Cambridge University Press.
Turner, Mark. 1991. Reading Minds: The Study of English in the Age of Cognitive
Science. Princeton: Princeton University Press.
100 JOSEPH GRADY
1987. Death is the Mother of Beauty. Chicago: University of Chicago

Press.
BLENDING AND METAPHOR
JOSEPH E. GRADY, TODD OAKLEY & SEANA COULSON

George Mason University, Case Western Reserve University &
University of Arizona
1. Introduction
The framework sometimes referred to as 'conceptual metaphor theory', with
its origins in Lakoff & Johnson (1980), is one of the central areas of research in
the more general field of cognitive linguistics. Within this field, the notions of
'source domains' and 'target domains', 'invariance', 'mappings', and so forth
have become a common, though not universal, vocabulary for discussing the
linguistic and conceptual phenomena of metaphor. The findings and principles of
this framework have been applied in numerous studies, both within and outside
of the field of linguistics.
A more recent framework, proposed by Fauconnier and Turner (1996,1998)
seeks to explain much of the same linguistic data, and also to unify the analysis
of metaphor with the analysis of a variety of other linguistic and conceptual
phenomena. This framework - referred to variously as the theory of 'blending',
'conceptual blending', and 'conceptual integration' - shares many aspects of
conceptual metaphor theory (CMT). For instance, both approaches treat metaphor
as a conceptual rather than a purely linguistic phenomenon; both involve
systematic projection of language, imagery and inferential structure between
conceptual domains; both propose constraints on this projection; and so forth.
However, there are also important differences between the approaches: CMT
posits relationships between pairs of mental representations, while blending theory
(BT) allows for more than two; CMT has defined metaphor as a strictly directional
phenomenon, while BT has not; and, whereas CMT analyses are typically
concerned with entrenched conceptual relationships (and the ways in which they
may be elaborated), BT research often focuses on novel conceptualizations which
may be short-lived.
In this article we explore the relationship between BT, CMT and the
phenomena they address, arguing that the two approaches are complementary. In
particular, the cross-domain relationships which have been identified by CMT
researchers shape and constrain the more complex process of conceptual blending.
The nature of this relationship has relevance for anyone interested in the
102 JOSEPH GRADY, TODD OAKLEY & SEANA COULSON
conceptual analysis of language and, more broadly, for anyone interested in

conceptual structure.
We begin with an overview of the BT framework, focusing on similarities
and differences with the CMT framework.
2. Blending theory and conceptual metaphor theory

2.1 Domains vs. mental spaces
In the CMT framework, metaphors are analyzed as stable and systematic
relationships between two conceptual ' domains'. In a metaphorical expression like
(1) The committee has kept me in the dark about this matter.
language and conceptual structure from the 'source' domain of vision is used to
depict a situation in the 'target' domain of knowledge and understanding.
Particular elements of the source and target domains are picked out through a
combination of the source language used ("in the dark") and the relevant
conceptual metaphor, a 'mapping' - presumably stored as a knowledge structure
in long-term memory - which tells us how elements in the two domains line up
with each other. In this metaphor, knowledge structures which concern seeing
have been put into correspondence with structures concerning knowledge and
awareness. Because the mapping is principled, ignorance is associated with
darkness as well as other conditions which preclude sight. In fact, thanks to the
general mapping between visual perception and intellectual activity, nearly any
concept related to the experience of vision is likely to have a clear counterpart in
the realm of knowledge and ideas. We easily understand a novel sentence like
"You'd need an electron microscope to find the point of this article" - and the
conceptual metaphor is the mechanism by which we interpret such references.1
In BT, by contrast, the basic unit of cognitive organization is not the domain
but the 'mental space' (Fauconnier 1994 [1985]), a partial and temporary
representational structure which speakers construct when thinking or talking about
a perceived, imagined, past, present, or future situation. Mental spaces (or,
'spaces', for short) are not equivalent to domains, but, rather, they depend on
them: spaces represent particular scenarios which are structured by given domains.
For instance, a BT account of example 1 would involve a space in which the agent
is standing in the dark. While this representation appeals to our knowledge of
visual experience, the recruited structure is only a small subset of knowledge of
that domain. In short, a mental space is a short-term construct informed by the
more general and more stable knowledge structures associated with a particular
domain.
BLENDING AND METAPHOR 103
2.2 Two domains vs. four spaces

While CMT analyses involve mappings between precisely two conceptual
structures, BT typically makes use of a four-space model. These spaces include
two 'input' spaces (which, in a metaphorical case, are associated with the source
and target of CMT), plus a 'generic' space, representing conceptual structure that
is shared by both inputs, and the 'blend' space, where material from the inputs
combines and interacts. A BT account of example 1 would include the following
spaces: an input space drawing on the domain of vision, in which a person (A) is
surrounded by darkness; another input space, drawing on the domain of
intellectual activity, in which a committee has withheld information from an
individual (A'); a mapping between these spaces, specifying that A and A' are to
be taken as one and the same person, that the person's inability to see corresponds
to unawareness, and so forth; a generic space containing the shared material the
two inputs have in common (roughly, 'a person who has no access to a particular
stimulus'); and the blended space, in which a committee is causing an individual
to remain in the dark.
Note that in the 4-space model material is projected from both the source
and target spaces to the blend. This arrangement contrasts with the simple,
unidirectional projection posited by CMT, in which mappings are from source to
target.
2.3 Emergent structure

One of the chief motivations for BT, according to proponents, is that the
four-space model can account for phenomena that are not explicitly addressed by
mechanisms of the two-domain model. Consider, for example, the well-worn
metaphor
(2) This surgeon is a butcher.
intended as a damning statement about an incompetent practitioner (Veale 1996).

Initially, the metaphor may seem to be explainable in terms of direct projection
from the source domain of butchery to the target domain of surgery, guided by a
series of fixed counterpart mappings: "butcher" onto "surgeon"; "animal" (cow)
maps onto "human being"; "commodity" onto "patient"; "cleaver" onto "scalpel";
and so forth. This analysis of the cross-domain relationships, however, cannot by
itself explain a crucial element of the statement's meaning: the surgeon is
incompetent. A butcher, though less prestigious than a surgeon, is typically
competent at what he does and may be highly respected. The notion of incompe
tence is not being projected from source to target.
Discussions in the CMT tradition have touched on some related points.

Lakoff & Turner (1989:79), for instance, ask, in the course of discussing
personifications of death, "why is the reaper grim?" After all, real reapers are not
necessarily grim, any more than butchers are necessarily incompetent. Their
answer, in part, is that "[t]he way we feel about the appearance and character of
the personification must correspond to the way we feel about the event." This is
an intuitively satisfying explanation for the reaper's grimness, but, as Lakoff &
Turner point out, there are independent reasons why death is personified as a
reaper in the first place, including a metaphorical conceptualization of the human
lifecycle as the lifecycle of a plant. We cannot apply the same logic to the case of
the incompetent butcher: why would we select a butcher as an appropriate source
image for a surgeon, and how would that selection (in itself, without requiring us
to specify "a bad butcher" or the like) communicate the notion of incompetence?
The intuitive answer is that the selection of the source image, and the interpreta
tion of the sentence, depend partly on contrasts between surgeons and butchers;
this is a factor which the mechanisms of CMT cannot cope with directly.
The BT model accounts for the inference of incompetence as follows. First,
the blend inherits some structure from each of the inputs (in accordance with
constraining principles, discussed below). From the target input space, structured
by the domain of SURGERY, it inherits such elements as the identity of a particular
person being operated on (i.e. the speaker), the identity of another individual who
is performing the operation, and perhaps details of the operating room setting.
From the source input space, which draws on the domain of BUTCHERY, it inherits
the role "butcher" and associated activities. The two input spaces share some
structure, represented in the generic space, in which a person uses a sharp
instrument to perform a procedure on some other being.
In figure 1, solid lines represent the cross-space correspondences that
constitute the mapping between the input spaces, dotted lines represent projections
between spaces, and the dashed line between the Surgeon role in Input 1 and the
Butcher role in the blend represents the fact that the butcher in the blend is
associated with the surgeon in the target space (see the discussion of 'fusion with
accommodation' in section 5.1).
Besides inheriting partial structure from each input space, the blend
develops 'emergent' content of its own, which results from the juxtaposition of
elements from the inputs. In particular, the BUTCHERY space projects a means-end
relationship incompatible with the means-end relationship in the SURGERY space.
Figure 1 : Conceptual integration network: surgeon as butcher

In butchery, the goal of the procedure is to kill the animal and then sever its flesh
from its bones. By contrast, the default goal in surgery is to heal the patient. In the
blended space, the means of BUTCHERY have been combined with the ends, the
individuals and the surgical context of the SURGERY space. The incongruity of the
butcher's means with the surgeon's ends leads to the central inference that the
butcher is incompetent (see the box within the blended space in figure 1). This
emergent property of the blend cannot be captured so explicitly within a CMT-
style analysis focusing on correspondences and projections from source to target.
2.4 On-line processing and entrenchment

Imagine we were observing a young, apprentice butcher at work, taking too
much time and being too tentative as he cut up a piece of meat. Someone might
comment,
(3) He's not a butcher, he's a surgeon.
In context, this sentence could be intended and understood as a negative

evaluation of the butcher's competence. Casting him as a surgeon highlights the
incongruity between his methods and those appropriate to a butcher.
Since the blend is probably novel at the time it is uttered, this example
illustrates the conception of blending as an on-line, real-time process that creates
new meaning through the juxtaposition of familiar material. A sentence like 2
probably draws on conventional associations with the word butcher, and the
blending analysis may really be an account of the historical derivation of such
usages, rather than of the on-line processing a hearer might use today. But
sentence 3, which depends on a very similar conceptual integration network, calls
more strongly for explanation in terms of real-time processing by means of a
cognitive structure like the one represented in the blending diagrams.
Whereas CMT has been primarily concerned with identifying regular,
conventional patterns of metaphorical conceptualization (and explaining
motivated extensions of these conventional structures), BT has often explicitly
addressed itself to novel and unique examples which do not arise from entrenched
cross-domain relationships. Since we encounter so many novel blends - e.g. in
cartoons, jokes, newly coined terms, terms we apply in unusual ways, etc. - and
since we create and understand them so effortlessly, such examples suggest that
the processes used to generate and interpret blends are well-developed, basic
elements of our cognitive machinery.
2.5 Basic processes of blending

As conceived within BT, blending involves three basic processes -
'composition', 'completion', and 'elaboration'. Composition, the most straightfor
ward process, refers to the projection of content from each of the inputs into the
blended space. Sometimes this process involves the 'fusion' of elements from the
inputs, as when the blend contains only a single individual who is associated with
the butcher from one space and the surgeon from the other. The representations
resulting from the composition process may or may not be realistic. For instance,
it is not plausible that a butcher would be allowed to operate on a surgery patient,
but nonetheless we easily construct and manipulate such a blended image.
Completion is the filling out of a pattern in the blend, evoked when structure
projected from the input spaces matches information in long-term memory. For
example, when we mentally project a butcher into an operating room, we end up
introducing the notion of incompetence and/or malice into the scene as well, in
order to make sense of the scene. We complete our understanding of the scenario
by introducing a new feature of the person, prompted by the juxtaposition of
elements from the inputs. The idea of destructive, inappropriate action calls to
mind the notion of an incompetent and/or malicious person. In this way, the
completion process is often a source of emergent content in the blend.
Finally, elaboration is the simulated mental performance of the event in the
blend, which we may continue indefinitely. For instance, we might proceed from
the image of a butcher carving a patient to the even more grotesque image of a
butcher packaging the patient's tissue as cold cuts. Once the connections to long
term knowledge about operations and butchery have been made, we are able to
imagine scenarios which unfold along various possible trajectories.
At each of these stages there is the potential for emergence of new content,
not available from either of the input spaces. New juxtapositions, new frames, new
features all arise when we combine elements from distinct mental spaces. These
bits of emergent structure (cf. Hofstadter's notion of "slippage") are chief
diagnostics for the occurrence of blending.2
2.6 Optimality principles of BT

Fauconnier and Turner (1998) lay out five "optimality principles" of
conceptual blending, constraints under which blends work most effectively. These
are:
Integration: The scenario in the blended space should be a well-integrated
scene.
Web: Tight connections between the blend and the inputs should be
maintained, so that an event in one of the input spaces, for
instance, is construed as implying a corresponding event in the
blend.
Unpacking: It should be easy to reconstruct the inputs and the network of
connections, given the blend.
Topology: Elements in the blend should participate in the same sorts of
relations as their counterparts in the inputs.
Good Reason: If an element appears in the blend, it should have meaning.
An additional principle, leading to some of the fanciful imagery encountered in

blends, is referred to as Metonymic Tightening: relationships between elements
from the same input should become as close as possible within the blend. For
instance, Western images of personified Death often depict the figure as a
skeleton, thus closely associating the event of death with an object that, in our
more literal understandings, is indirectly but saliently associated with it.
There is tension among some of these principles, and so each blend satisfies
them to varying degrees.3
Next, we move to a more detailed discussion of a particular metaphoric
blend.
3. The ship of state

This sentence taken from a piece of political commentary illustrates the
common conceptualization of a nation or society as a ship:
(4) With Trent Lott as Senate Majority Leader, and Gingrich at the helm in the House,
the list to the Right could destabilize the entire Ship of State.4
Before examining the details of this particular blend, let us look at the conven
tional mapping it builds upon. As it is used in popular discourse, the Nation-as-
Ship metaphor includes at least the following cross-domain correspondences:
Nation Ship
National policies/actions Ship's course
Determining national policies/actions Steering the ship
National success/improvement Forward motion of the ship
National failures/problems Sailing mishaps (e.g., foundering)
Circumstances affecting the nation (e.g. on Sea conditions
the political or economic levels)
Consider the following attested instance of the metaphor:
(5) Without the consent of our fellow citizens, we lose our moral authority to steer the
ship of state.5
The metaphorical correspondences underlying example 5 reflect the conventional

mapping described above, with the ship's course standing for the nation's policies,
and determining the ship's course (steering it) corresponding to determining the
nation's policies. The next example evokes a richer scenario.
(6) The [Sri Lankan] ship of state needs to radically alter course; weather the stormy
seas ahead and enter safe harbor.6
Here we have the image of a harbor in addition to the more standard notion of sea
conditions. The harbor stands presumably for stable political and economic
circumstances.
While the Nation-as-Ship is a conventional conceptualization, it is also
related to more fundamental metaphorical mappings, such as ACTION IS SELF-
PROPELLED MOTION, COURSES OF ACTION ARE PATHS, TIME IS MOTION, A SOCIAL
RELATIONSHIP IS PHYSICAL PROXIMITY (e.g., within a single sailing vessel),
CIRCUMSTANCES ARE WEATHER, STATES ARE LOCATIONS and SO forth. All these
conventional metaphors help motivate the framing of a nation and its history as
a ship plying the seas. The idea that simple metaphors interact to yield more
elaborate conceptualizations has been discussed by researchers working in the
CMT framework. (See, for instance, Lakoff & Turner's 1989 discussion of
'composite' metaphors, and Grady's 1997 more explicit analysis of the 'unifica
tion' or 'binding' of metaphors.) The blending framework offers a neat way of
representing this complex interaction of concepts and links, since it explicitly
allows for multiple spaces and multiple iterations of the integration process. One
blend may be the input for another.
More significantly, the blending framework here offers a way of accounting
for those elements of the Nation-as-ship image that have no specific counterparts
in the target space of nations and politics. For instance, ships have very particular
shapes and are made of particular materials. These important aspects of ships have
no conventional counterparts in the target domain of nations, but they figure
nonetheless in any metaphorical projection of the ship frame. We simply cannot
conceive of ships without evoking some aspects of their physical character.
Within the blending framework, we can account for this fact in terms of
pattern completion: once we have evoked, by means of more basic metaphors, the
image of a large container holding many people, or of a society moving forward
through space, and/or the idea that political events are partially determined by the
(metaphorical) weather, these images may match, and call up, stored representa
tions of a ship, and then all other elements of the ship domain are immediately
available for recruitment (i.e. they are 'primed'). The ship image in the blend
integrates a number of metaphorical understandings of society. Once it is evoked,
it may become as elaborate as our imaginations will allow, and like any other
conceptualization it has the potential to become conventional.
The Lott and Gingrich example in 4 provides a clear example of how
metaphoric expressions may recruit more mappings than those between a single
source and target domain. For instance, this example introduces the notion of
right-hand directionality (i.e. starboard, in the context of a ship), which is
independent of the Nations-as-Ships metaphor. The standard association between
right-left polarity and conservative-liberal alignments is clearly not based on the
ship model, as it is frequently encountered in contexts where there is no ship
imagery.
Furthermore, 4 suggests that the presence of two individuals will predictably
cause a ship to list dangerously to one side. While we can imagine a complicated
scenario in which their actions could lead to such an outcome - e.g. their handling
of very heavy cargo, or their steering and handling of the sails in particular wind
conditions - the sentence implies a simpler and more direct causal connection than
this. This causal structure appears not to be projected from the source domain of
ships, but from target domain logic, in which the Senate Majority Leader and the
Speaker of the House inevitably have a considerable, direct influence on national
policies and the overall political orientation of government. Blending theory
suggests that selective projection from the two input spaces yields an image which
is inconsistent with our understanding of the source space - two people whose
presence is likely to cause a ship to list to one side - but that the web of underlying
connections allow us to draw inferences from the blend nonetheless. When we
encounter sentence 4, we easily infer that the strong shift towards conservatism
may lead to political instability.7
4. Metaphors as inputs to blending

If conceptual metaphor theory is primarily concerned with well-established
metaphoric associations between concepts, and blending theory focuses on the
ability to combine elements from familiar conceptualizations into new and
meaningful ones, then conceptual metaphors are among the stable structures
available for exploitation by the blending process. As we have just seen in the
ship-of-state examples, conventional metaphors feed the blending process by
establishing links between elements in distinct domains and spaces. In this section
we explore this relationship in a bit more detail.
4.1 Types of counterpart connection

The network of connections which ultimately constitutes a blend depends
first on the establishment of links between the input spaces (Fauconnier &Turner
1998). These counterpart relations guide the construction of the blend.
Cross-space counterparts may be related to each other in a variety of ways.
For instance, in the case of an individual "kept in the dark" by a committee (see
Section 2), the counterpart relationship between the person (in one input) who is
in darkness and the person (in the other input) who is kept uninformed is based on
Identity. The same individual is represented in each input space, and these two
representations are, quite naturally, linked, in a way that helps guide the
construction and interpretation of the blend.
Other types of counterpart relationship across mental spaces include the
connection between a role and a value - e.g., the connection between "Jocasta"
and "Oedipus' mother," discussed by Fauconnier (1994 [1985]) - and the
connection between an entity and a representation of the entity, such as a man and
his portrait. Similarity and Analogy are relations which play obvious roles in many
conceptual integration networks, including ones we call metaphorical blends. For
instance, surgeons and butchers share the generic structure of a person wielding
a sharp object to cut flesh.
Conventional metaphors can also provide the counterpart mappings to
launch blends. For instance, the metaphorical association between nations and
ships is thoroughly conventional, and forms part of many people's conceptual
repertoires. What started out (undoubtedly) as some individual's creative, on-line,
conceptual achievement has become a shared, entrenched conceptualization,
presumably because the blend proved successful for some purpose, therefore arose
again, and through repeated experience became conventional. As a result, the
metaphorical mapping between the nation and the ship, the nation's history and
the ship's course over the sea, and so forth, is now stored in memory and provides
a trigger that allows conceptual blending to proceed, including the kinds of
creative conceptual manipulation we examined in the last section.
Of course for a conventional metaphoric blend to have arisen in the first
place, it must, itself, be based on some kind of counterpart mapping. This is an
area where CMT, and the associated body of work accumulated over the past
eighteen years, informs the blending framework. Numerous principles regarding
the kinds of concepts which become associated by conventional metaphor have
been uncovered and described, including patterns in the relationship between the
image-schematic structure of source and target ('Invariance', Brugman 1990;
Lakoff 1990; Turner 1991), the relationships holding among different mappings,
the kinds of content that may be associated with source and target8, and the ways
in which source and target may or may not be similar.
Importantly, there is a class of entrenched metaphors which are not based

on similarity or analogy, and which are therefore unlike the metaphoric counter
part relations which arise on-line.
(7) a. These two colors are not particularly close [i.e. similar].
b. His sunny smile lit up the room.
c. Tomorrow is a big day for this organization.
These sentences are illustrations, respectively, of the following conventional

metaphors: SIMILARITY IS PROXIMITY, HAPPINESS IS BRIGHTNESS, IMPORTANCE IS
SIZE (all known by various names in the conceptual metaphor literature). There is
no obvious sense in which the concepts paired in these metaphors are similar or
analogous to one another. Each is scalar in some sense, but this is not sufficient
motivation for the particular pairings evidenced here. (Consider the fact that
Brightness may not stand for Similarity, and so forth.) Instead these metaphors are
most plausibly explained as entrenched conceptual associations arising from
recurring correlations in experience. Just as the recurring correlation between
quantity and height (e.g. of a pile) motivates the metaphor MORE IS UP (as in,
"Crime figures have soared"), these metaphors are motivated by recurrent types
of episodes which bring together particular dimensions of experience. For
instance, brightness is correlated with warmth and increased visibility, both of
which trigger contentment (cf. HAPPINESS IS BRIGHTNESS). Lakoff & Johnson
(1980) argued convincingly that various metaphors relating "UP" to other concepts
could not be based on objective similarity or shared features, and the same holds
in the cases mentioned here. They are not based on similarity or analogy, but must
instead be based on experiential correlation.
Metaphors like SIMILARITY IS PROXIMITY, HAPPINESS IS BRIGHTNESS, and
IMPORTANCE IS SIZE are 'primary metaphors' (see Grady et al. 1996; Grady 1997;
Lakoff & Johnson in press), a special class of entrenched associations, based on
neither similarity nor analogy. They seem to constitute a distinct sort of counter
part connection on which blends may be based.
4.2 Complex metaphorical blends

Since blending is an opportunistic process of on-line space-building, any
conceptualization that starts out as a primary metaphor, or other simple conceptual
association, is susceptible to being elaborated. The source concept of any basic
metaphor can trigger the construction of a richer image. If difficulty is understood
as heaviness - due to a correlation between, on the one hand, our sensory judgment
of mass and, on the other, affective states associated with exertion - then we can
talk about tons of work. If a cheery disposition is metaphorically associated with
bright light, then we understand what a thousand-kilowatt smile must be like

(given some additional input from our knowledge of electricity). If the experience
of moving forward is correlated with an affective state telling us we are about to
achieve some purpose, then the Ship of State makes headway as the nation works
to accomplish its objectives.
The Nation-as-Ship example also illustrates the way in which multiple
simple metaphors can be relevant within a single complex blend. The ship's
forward motion is understood in terms of a conventional metaphorical association
with goals more generally. The notion of "safe harbor," as in example 7, derives
from a metaphorical understanding of circumstances as locations and surround
ings. While the image of a "lookout" is not a conventional part of the Nation-as-
Ship blend, it can easily be incorporated, and linked to a target domain notion of
anticipating future events (i.e. foresight), based on a metaphorical association
between vision and thought (cf. KNOWING IS SEEING). The metaphorical right-left
orientation of political parties is another conventional counterpart connection
which can be recruited to enrich the blend, as we have seen.
The role of basic metaphors in complex blends illustrates an important
principle about the relationship between metaphor and blending: it is particular
connections within an entire conceptual integration network which we regard as
metaphoric. For instance, it is, in a way, misleading to refer to the Lott-Gingrich
example, or the cognitive representation that motivates the words, as "a meta
phor." Within the conceptual complex that underlies the sentence there are several
distinct metaphoric connections - e.g. nation/ship, conservative/right - and the
blend as a whole does not represent the systematic mapping of one domain onto
another.
Furthermore, metaphoric blends may contain figurative links that are not,
themselves, metaphoric. For instance, when we personify death as a skeleton
carrying a sickle, we are dealing with a metaphorical image, but one which has
been elaborated via the addition of details which do not derive from a metaphoric
mapping. The relationship between skeletons and death is not metaphorical but
metonymic; skeletons figure literally in scenarios involving death. In accordance
with the principle of Metonymic Tightening (see Section 2), the skeleton becomes
even more closely associated with Death in the blend than it is in the source input.
In short, conventional metaphoric relationships may be the starting points for the
process of creating complex conceptual blends. And identifying a metaphoric
relationship holding between source and target elements is sometimes only the
starting point for analyzing a blend.
5. What makes a blend metaphoric

Given that many of the blending examples discussed in the BT literature are
not metaphoric, it is helpful to understand what characterizes metaphoric blends
and distinguishes them from others. As we have seen, some blends depend on
counterpart relations dictated by conventional metaphoric associations, such as the
one between nations and ships. There are other aspects of blends, though - relating
to their structure, their content, and the linguistic and conceptual setting in which
they appear - that make them seem metaphoric to us.
5.1 Fusion with accommodation

In a metaphoric blend, prominent counterparts from the input spaces project
to a single element in the blended space - they are 'fused'. A single element in the
blend corresponds to an element in each of the input spaces. A ship in the blend
is linked to a ship in the source space and a nation in the target, a surgeon is linked
to both a surgeon and a butcher, and so forth. Intuitively speaking, the point of
metaphors is precisely that one thing is depicted as or equated with another. In the
blending framework this means a single element in the blended space has links to
each of the input spaces.
By contrast, in other sorts of blends these counterparts may project to
distinct elements in the blended space. For example, Fauconnier & Turner (1998,
and elsewhere) have discussed the following passage, in which a modern
philosopher describes his "debates" with Kant - i.e. his musings over particular
topics, in relation to Kant's views of the same topics:
(8) I claim that reason is a self-developing capacity. Kant disagrees with me on this
point. He says it's innate, but I answer that that's begging the question, to which
he counters, in Critique of Pure Reason, that only innate ideas have power. But I
say to that, what about neuronal group selection? He gives no answer.
The sentences arise from a blended conceptualization in which the two philoso
phers are imaginatively juxtaposed with each other and engage in conversation
about particular issues. In this blend, which strikes us as fictive but not metaphori
cal, the philosophers who correspond to each other in the two input spaces (and
are therefore connected by an Analogy link) are not, in fact, fused in the blended
space. Instead, they retain their individual identities, and the nature of their
interaction is the focus of the blend.
While the philosophers are projected as distinct participants, other aspects
of these input spaces are fused in the blend. For example, the languages of the
philosophers are fused into a single language (not necessarily specified), the
historical gap between them is collapsed, the geographical settings are also
merged, and so forth. Thus fusion alone does not identify metaphors.
Another sort of non-metaphorical fusion occurs in 'framing', a variety of
conceptual integration which operates by the same basic principles outlined above
(Fauconnier & Turner 1998). In framing we identify a particular entity with a slot
in a more general conceptual frame. For instance, the statement, "Carl is a
bachelor" depends on the following conceptual operation: a particular unmarried
man we know ("Carl") is associated with our cultural model of bachelors, which
in turn is informed by our models of marriage and so forth (see Fillmore 1982).
Our knowledge of Carl and of the BACHELOR frame represent the input spaces for
a conceptual integration. In the blend, Carl is fused with the frame role "bachelor."
This example, like framing examples in general, does not strike us as
metaphorical, since it represents a particular variety of fusion: the elements which
are counterparts in the cross-space mapping are combined by composition in the
blend. While all blends are selective in that they only draw on some of our
knowledge of the input domains, framing involves counterparts which are
essentially compatible, such that information about each serves to specify the
fused element in the blend.
Metaphorical blends, on the other hand, involve a different kind of fusion,
in which certain very salient aspects of input domain structure are prohibited from
entering the blend, and in which some salient structure in the blended space is
prevented from floating back to the inputs. That is, there is information from one
of the inputs (the target) that must be ignored in the blend: nations do not move
across the sea, ignorance is not literally associated with darkness, etc. An
important feature of metaphorical fusion of counterparts, then, is that it involves
overriding, and therefore not projecting, salient aspects of our knowledge of the
target. This sort of asymmetrical projection occurs in any case where the
organizing frame in the blend is projected from one input at the expense of the
other, e.g. the ship frame in the Nation-as-Ship cases. The fact that source and
target must be incompatible in some sense relates to an old claim about metaphor,
which can be considered here in a new light.
Philosophers (e.g. Davidson, Grice, Searle) have argued that listeners are
cued to interpret a particular reference as metaphorical by anomalies of meaning.
On this view, when we hear a statement such as "Inflation soared," the impossibil
ity of the event is our cue that the statement is intended metaphorically. Arguing
against this claim, Keysar (1989) has shown that subjects are able to interpret a
statement like "Paul is a magician" as a metaphorical reference to Paul's abilities
as an accountant, even when Paul is actually a magician by trade. In other words,
the recognition of metaphor does not depend on surface anomalies of meaning. In
the blending framework the notion of anomaly can be defined with greater subtlety
and specificity: the network of conceptual connections which comprises the

meaning of the utterance includes a counterpart relation between entities which
we know to be incompatible in some important sense. (In Paul's case the
counterpart relation is an analogical connection between skillful bookkeeping and
the supernatural manipulation of matter). Whether or not this fact is relevant to
on-line processing of metaphorical language - still a controversial question among
philosophers and psycholinguists - part of what defines metaphors is that they
involve (temporary) suppression of critical knowledge of a given conceptual
domain, and therefore are not compatible with our understanding of reality. We
refer to this particular phenomenon, in which structure from one fused element is
blocked, as 'accommodation': the target material yields to the source material,
which is explicitly represented in the blend.
Knowing whether the fusion of elements from two inputs involves
accommodation depends crucially on how specifically those elements are
construed. Consider the issue of language in the Debate-with-Kant example in 8.
We do not feel it is metaphoric to report the debate as though it happened in
English, even though Kant was a German speaker. This is almost certainly
because the details of the languages are not at issue in any part of this conceptual
network, and so we might say that each input simply contains the generic notion
Language. For the purposes of this blend we are not 'construing' (in the sense of
Langacker 1987) the languages of the philosophers as, specifically, English and
German.
To highlight the importance of construal in this case, consider that it is easy
to create a context in which a mapping between German and English does feel
metaphorical, or at least more metaphorical than in the Kant blend. Imagine
reading a philosophy essay written in dense, convoluted English, perhaps overly
influenced by translations of Heidegger, and exclaiming, "This isn't English, it's
German!". Here we have a blend based on the same pairing of counterparts, and
yet this case is metaphorical where the previous one is not. This is because in the
Bad Essay case we are interested in the particulars of the languages and their
differences, while in the Kant case we are only interested at the level of unspeci
fied "Language," as a means of communication and medium of debate. That is,
one construal profiles features of English and German while the other profiles
entities at a more schematic level. In the Kant case, the active representations in
the input spaces do not include particulars about language, and so there is no
conflicting information to resolve or accommodate. The Bad Essay example,
though, does have fusion with accommodation, because it represents a construal
at a different level of schematicity; consequently, it is felt to be (more) metaphori
cal.
Our discussion of fusion with accommodation echoes, in new terms, various

discussions of metaphor as a phenomenon of "category extension," or "class
inclusion," and is compatible with psycholinguistic results showing that subjects
are more apt to see metaphor in cases where there is greater semantic distance
between elements. The CMT principle that source and target come from different
"domains" is also in the same spirit as our more general statement that metaphors
involve the fusion of saliently distinct elements from two inputs. Note, though,
that metaphoric counterparts do not obligatorily come from different conceptual
domains or frames. For instance, a modern philosopher might come out of a
colleague's office and mutter, "I've just spent the afternoon debating with
Immanuel Kant!"10 Here "Kant" and his modern counterpart are understood in
terms of the same frames and domains, yet the statement would strike some as
metaphorical.
Finally, note that if metaphor depends on salient differences between the
relevant concepts, this implies that there are degrees of metaphoricity. Many
researchers have suggested this (e.g., Fauconnier & Turner 1998), and the
examples here offer further support. While some utterances are prototypically
metaphoric and others prototypically nonmetaphoric, there seems to be no hard
and fast distinction between these categories.
5.2 Directionality and asymmetric topicality

Another important feature of metaphoric blends is that their input spaces do
not have equal status as topics. In the non-metaphorical Debate-with-Kant blend,
both philosophers, along with their positions, are the focus of attention. It is the
interaction between the two, and a consideration of their relative merits, that
motivate the blend. In other words, each of the inputs has high topicality. A given
inference may relate to one more than the other, but both are held up to scrutiny
and comparison by means of the blend, and the blend's function is to give us a
means of examining the relationship between the two.
Metaphors, by contrast, are distinguished by asymmetric topicality. One of
the inputs is topical and the other provides a means of re-framing the first for
some conceptual or communicative purpose; these are, respectively, the target and
source inputs of the metaphor. For instance, in the Nation-as-ship blend, the nation
is the actual topic of interest, the target space; when we use the blend we are
interested in conceptualizing, picturing, or describing aspects of the nation, not in
understanding more about ships. Similarly, "My surgeon is a butcher" is a
(damning) statement about a surgeon, not a butcher.
Coulson (1997) has pointed out that some metaphorical blends allow us to
project inferences in more than one direction. In a joke about the "Menendez
Brothers [Computer] Virus," the blend establishes a (darkly) humorous connection
between actions of the virus and alleged actions of human agents. Erik and Lyle
Menendez, two brothers in their twenties, killed their parents and subsequently
inherited their substantial wealth. At their widely-publicized trial, the brothers
argued that they had been the victims of long-term abuse, and that the killings had
therefore been a form of self-defense, although their parents were unarmed at the
time. According to the joke, the virus "eliminates your files, takes the disk space
they previously occupied, and then claims it was a victim of physical and sexual
abuse on the part of the files it erased" (Coulson 1997:252). While the joke uses
details of the criminal case to explain the virus, which is in this sense the target
input of the blend, it also invites inferences about the brothers. Because the
criminal case was controversial, one of the effects of the joke is to support a
particular view, namely that the Menendez brothers were guilty of murder, and
that the defense they offered was absurd. Given that the same network of
connections is used to make inferences about the brothers and about the virus, this
example is an apparent exception to the principle that metaphors involve
asymmetric topicality: a single conceptual integration network - which feels
metaphorical and involves fusion (with accommodation) between profiled
participants - allows inferences in either direction, and invites us to focus on
aspects of each input.
However, the Menendez Brothers Virus blend operates on distinct levels
(and possibly in distinct stages) and different directionality is associated with
each. An initial understanding of the virus depends on successfully mapping the
description of human actions onto the domain of computer operations and files.
Understanding the implications about the criminal case is a separate process which
involves unpacking one of the input spaces on which the joke is based. (To put it
another way, this process involves retrieval of presuppositions, guided by the
connections in the network.) Topicality is asymmetrical during each of these
processes.
Moreover, topicality is not the only factor determining the directionality of
metaphor. A metaphoric blend which recruits conventional mappings inherits the
directionality of those mappings, as the Nation-as-ship blend inherits the
directionality of metaphors for change, time, society, political orientation, etc. and
maps source concepts onto all these target concepts. Furthermore, there is a long
tradition of describing the greater concreteness of metaphoric sources as opposed
to targets. Topicality probably correlates with these other factors in that certain
kinds of topics are more likely to evoke metaphoric counterparts, which in turn are
likely to be relatively rich in sensory content.
5.3 Metaphors vs. counterfactuals: The role of linguistic context

Like metaphors, counterfactuals involve counterpart relations between
entities that are construed as essentially different. Consider this hypothetical
example, spoken by a senior professor to a junior colleague:
(9) If I were you I'd be working on finishing my book.
Like metaphors, the conceptual blend underlying this sentence involves

counterparts, construed as crucially different, which are fused in the blended
space; a single entity there corresponds to a different person in each of the inputs.
The hypothetical professor does not (and could not) have all the properties of both
input professors; it is their differences which motivate the blend. Specifically, the
professor in the blend is in the situation of the junior professor, who must publish
a book in order to be tenured, but has the attitudes and priorities of the senior
professor. If this blend includes fusion of distinct entities, why does it not strike
us as metaphorical?
It is likely that one of the factors is the perceived degree of difference
between the counterparts. As we mentioned earlier, and as many previous works
on metaphor have noted using various terminology, the perceived difference
between two entities is an important determinant of how metaphorical an
association between them may seem. A sentence starting with "If I were a cloud"
strikes us as more figurative than one starting with "If I were you." A sentence
starting with "If I were Napoleon" probably falls in between. This relative scale
is plausibly based on the degree of perceived category difference, at the relevant
level of abstraction, between the "I" element and the counterpart in each case.
Another factor, though, is the construction of the sentences themselves. The
rhetorical force - i.e. the profiling effects - of counterfactual statements may run
contrary to those we associate with metaphor. A sentence starting with "If I were
a cloud" may strike us as less metaphoric than one starting with "I am a cloud,"
since the counterfactual specifically negates the proposition that the two entities
can be equated in some sense. That is, while both sentences may be interpreted
based on the same network of conceptual links and projections, the profiling may
be different in the two cases, such that one is more consistent with our prototype
of metaphors. Metaphors typically assert counterpart connections without drawing
explicit attention to incongruities between the connected entities.
The above concerns suggest that metaphor is not a sharply delineated phenome
non, and underscore the need for a framework like BT which can account for the
mapping operations that underlie central and peripheral cases alike.
6. Conclusion
Differences between conceptual metaphor theory and blending theory, such
as the distinct nature of directionality in the two frameworks, have led some
researchers to treat them as competing theories (e.g. Coulson 1996). Alternatively,
one might consider the two approaches to be incommensurable. After all, CMT
addresses recurring patterns in figurative language, while BT seems to focus on
the particulars of individual cases. And the phenomena accounted for by CMT
consist of stable knowledge structures represented in long-term memory, while BT
seeks to model the dynamic evolution of speakers' on-line representations.
In this paper we have taken neither of these positions. Rather, we propose
that because they tackle different aspects of metaphoric conceptualization, the two
frameworks are largely complementary. The conventional conceptual pairings and
one-way mappings studied within CMT are inputs to and constraints on the kinds
of dynamic conceptual networks posited within BT.
If we establish that the findings of CMT and BT are consistent, the potential
rewards are significant, since this allows us to unify two streams of research into
a more general and comprehensive treatment of linguistic and conceptual
phenomena. BT researchers have argued that the same principles which speakers
use to understand metaphor operate similarly across a wide range of nonmetapho-
rical phenomena.
The generality of conceptual blending theory derives in part from its roots
in mental space theory which treats metaphor as a special case of indirect
reference. As our examples illustrate, metaphoric and non-metaphoric conceptual
izations alike rely on selective projection from two or more input spaces into a
blended space, the establishment of cross-space mappings, structuring the blended
space via processes of composition, completion, and elaboration, and subsequent
projection of structure from the blended space to the inputs. By treating all sorts
of mappings as formally identical at a certain level we can understand the transfer
of structure in metaphor as fundamentally similar to the transfer of structure in
non-metaphoric instances.
Among the non-metaphoric types of linguistic structure which can be treated
in a blending framework are counterfactuals and conditionals. A number of
researchers working within the framework of conceptual blending have addressed
its implications for how people reason about events which could have happened,
but did not (e.g. Fauconnier 1997; Oakley 1995, 1998; Turner 1996). The tools of
blending theory, including the cline between identity, similarity, and analogy
links, have also proven useful in explaining the variety of complex concept
combinations coded for by modified noun phrases. For example, blending theory
has been used to explore issues of concept combination in seemingly simple cases
like "red pencil" (Turner & Fauconnier 1995; Sweetser in prep.), more exotic
cases like "land yacht" and "dolphin-safe tuna" (Turner & Fauconnier 1995), and
privative constructions such as "alleged affair" and "fake gun" (Coulson &
Fauconnier in press).
Conceptual metaphor theory has often emphasized the role played by
metaphors in structuring abstract concepts with cognitive models projected from
more concrete source domains. With its additional machinery for recruiting
knowledge structures, blending theory has also proven to be powerful in
explaining how abstract concepts can be understood with the help of blended
models. Although blended models are not always plausible - cf. the debate
between Kant and a modern philosopher - blends can promote integrated
construals that help us reason about abstract phenomena. Accordingly, a number
of researchers have demonstrated the importance of particular blends in the
invention of mathematical concepts (Fauconnier & Turner 1998; Lakoff & Nunez
in press) and proofs (Robert 1998). Moreover, Maglio & Matlock (1998)
demonstrate the roles of distinct conceptual blends as experts and novices interact
with Web browsers.
Blending theory has also been taken up by literary theorists interested in the
cognitive underpinnings of verbal creativity. For example, Brandt (in press) shows
how integration networks can be used to represent the complex flow of inferences
and imagery in the poetry of Baudelaire. Turner (1996) shows how the machinery
of conceptual blending operates in a wide range of literary genres from simple
parables, to the imagery in Dante's Inferno, to Shakespearean drama.
More surprising, perhaps, is the suggestion that the very same integrative
mechanisms underlie the most banal aspects of language processing (Turner &
Fauconnier 1995; Mandelblit 1997). Sweetser (in prep.) demonstrates the ubiquity
of blending phenomena and shows how its processes are used to combine the
semantic properties of grammatical constructions with the lexical semantics of the
words used in their instantiations. Similarly, Fauconnier & Turner (1996) have
suggested that integrative mechanisms of blending are needed to understand
particular instances of the caused-motion construction such as "He sneezed the
napkin off the table" (cf. Goldberg 1996).
In arguing that conceptual metaphor theory and blending theory provide
largely complementary formalisms, we have suggested that many of the
differences between them reflect their motivation in different aspects of the same
data. While the metaphor theorist strives to capture generalizations across a broad
range of metaphoric expressions, the blending theorist typically focuses on the
particulars of individual examples. Because it is useful to separate entrenched
associations in long-term memory from the on-line processes that recruit them, we
have argued that the former issue is the province of metaphor theory, and the
latter, the province of blending theory. Consequently, metaphor theory will
continue to address such questions as which concepts are conventionally

associated with each other, how and why such conventional associations arise, and
how cross-domain mappings are structured. As argued above, such issues are
central to the question of how metaphoric blends arise, and may have important
implications for the quasi-metaphoric blending in other sorts of examples. To be
sure, a full understanding of the conceptual feats that underlie the examples
considered above will require both a rich theory of metaphor and a fully specified
model of conceptual blending.
Notes
1. Grady (1997) has argued that conceptual domains are often too general as units of analysis
for conceptual metaphors, and that many mappings are better described as associations
between particular source and target concepts, belonging to distinct domains. Both
approaches treat metaphors as relationships between established, long-term knowledge
structures.
2. This is not to say that emergent structure is a necessary feature of conceptual blends : some
blends are truth-functionally compositional. However, it is the frequent need to account for
emergent structure that motivates BT.
3. See Gentner (1983) for another approach to constraining and optimizing cross-domain
mappings. Gentner's framework applies to relations between (what BT treats as) input
spaces to a blend.
4. From Carol R. Campbell, "Cave Man Bill And The Doleful State of American Politics,"
published by The Written Word, an on-line journal of economic, political and social
commentary.
5. Bruce E. Johnson, "Making a difference," Federal Executive Institute Alumni Association
Newsletter President's Report, April 1997, No. 225.
6. From "Two years of PA [the People's Alliance] : the state of the Nation," Editorial in "The
Sunday Times [of Sri Lanka] on the Web" Aug. 18, 1996.
7. While a reference to the ship's course (rather than to listing) might have been more
conventional in this context, the fact that we easily interpret the sentence demands that we
account for it as it stands. In the BT framework it does not matter whether such an
improbable image results from deliberate innovation or the accidental "mixing" of
metaphors.
8. See, e.g., Grady's (1997) discussion of primary metaphors, in which source concepts have
"image content" while target concepts have "response content."
9. For more on the contrast between resemblance metaphors and correlation metaphors, see
Grady (this volume).
10. This statement will strike some readers, but not others, as metaphorical. The dividing line
between metaphor and other sorts of figurative reference is not sharply drawn or
universally agreed upon.
References
Brandt, Per Aage. 1998. "Cats in space". Acta Linguistica.
Brugman, Claudia. 1990. "What Is the Invariance Hypothesis?". Cognitive

Coulson, Seana. 1996. "The Menendez Brothers Virus: Analogical Mapping in
Blended Spaces." Conceptual Structure, Discourse, and Language, ed. by
Adele Goldberg. 67-83. Stanford, Calif.: CSLI.
, & Gilles Fauconnier. 1998 in press. "Fake Guns and Stone Lions:
Conceptual Blending and Privative Adjectives". Cognition and Function in
Language, ed. by Daniel Jurafsky, Laura Michaelis, & Barbara Fox.
Stanford, CA: CSLI.
1997. "Semantic Leaps: The Role of Frame-Shifting and Conceptual
Blending in Meaning Construction".Ph.D. Dissertation. University of
California, San Diego.
Fauconnier, Gilles. 1994 [1985]. Mental Spaces: Aspects of Meaning Construc-
tion in Natural Language. Cambridge: Cambridge University Press.
1997. Mappings in Thought and Language. Cambridge, U.K.: Cambridge
University Press.
, & Mark Turner. 1996. "Blending as a Central Process of Grammar".
Conceptual Structure, Discourse, and Language, ed. by Adele Goldberg.
183-203. Stanford, Calif.: CSLI.
, & Mark Turner. 1998. "Conceptual Integration Networks". Cognitive
Science 22. 133-187.
Fillmore, Charles. J. 1982. "Frame Semantics". Linguistics in the Morning Calm.
111-137, The Linguistic Society of Korea. Seoul, South Korea: Hanshin
Publishing Co.
Gentner, Dedre. 1983. "Structure-Mapping: A Theoretical Framework for Ana
logy". Cognitive Science 7. 155-170.
Goldberg, Adele E. ed. 1996. Conceptual Structure, Discourse, and Language.
Stanford, Calif.: CSLI.
Grady, Joseph, Sarah Taub,, & Pamela Morgan. 1996. "Primitive and Compound
Metaphors". Conceptual Structure, Discourse, and Language, ed. by Adele
Goldberg, 177-187. Stanford, Calif.: CSLI.
1997. "Foundations of Meaning: Primary Metaphors and Primary Scenes".
Ph.D. Dissertation. University of California, Berkeley.
This volume. "A Typology of Motivation for Conceptual Metaphor:
Correlation vs. Resemblance".
Keysar, B. 1989. "On the Functional Equivalence of Literal and Metaphorical
Interpretations in Discourse". Journal of Memory and Language 28. 375-
385.
Lakoff, George,, & Mark Johnson. 1980. Metaphors We Live By. Chicago:
Univeristy of Chicago Press.
1990. "The Invariance Hypothesis: Is abstract Reason Based on Image

Schemas?" Cognitive Linguistics 1. 39-74.
, & Rafael Nunez. In press. "The Metaphorical Structure of Mathematics:
Sketching Out Cognitive Foundations for a Mind-Based Mathematics".
Mathematical Reasoning: Analogies, Metaphors, and Images, ed. by Lyn
English. Hillsdale, N.J.: Erlbaum.
Maglio, P.P., & T. Matlock, 1998. "Metaphors We Surf the Web By". Workshop
on Personalized and Social Information Space, Stockholm, Sweden.
Mandelblit, N. 1997. "Grammatical Blending: Creative and Schematic Aspects in
Sentence Processing and Translation". Ph.D. Dissertation. University of
California, San Diego.
Oakley, Todd. 1995. "Presence: The Conceptual Basis of Rhetorical Effect".
Ph.D. Dissertation.University of Maryland.
1998. "Conceptual Blending, Narrative Discourse, and Rhetoric".
Cognitive Linguistics 9. 321-360.
Robert, Adrian. 1998. "Blending in the Interpretation of Mathematical Proofs".
Conceptual Structure, Discourse, and Language II, ed. by Jean-Pierre
Koenig. 337-350. Stanford, Calif.: CSLI.
Sweeter, Eve. In preparation. "Mental Spaces and Cognitive Linguistics: A
Cognitively Realistic Approach to Compositionality".
Turner, Mark. 1991. Reading Minds: The Study of English in the Age of Cognitive
Science. Princeton, N.Y.: Princeton University Press.
1996a. "Conceptual Blending and Counterfactual Argument in the Social
and Behavioral Sciences". Counterfactual Thought Experiments in World
Politics, ed. by Philip Tetclock, & Andrew Belkin. 291-295. Princeton,
N.Y.: Princeton University Press.
1996b. The Literary Mind. Oxford: Oxford University Press.
, & Gilles Fauconnier. 1995. "Conceptual Integration and Formal Expres
sion". Metaphor and Symbolic Activity 10:183-203.
Veale, Tony. 1996. "Pastiche". Unpublished manuscript.
SELF AND AGENCY IN RELIGIOUS DISCOURSE
PERCEPTUAL METAPHORS FOR KNOWLEDGE AT
A MARIAN APPARITION SITE
VICTOR BALABAN
Emory University
"My dear children of America, are you listening? Are your ears closed or
open? Are your hearts closed or open?"
the Virgin Mary to Nancy Fowler, October 13, 1991
1. Introduction
This is a study of how perceptual metaphors for knowledge can be used by
English speakers as part of a pragmatic strategy to reduce the speaker's own
agency in the events described. In this case the narratives are accounts of
miraculous signs recounted by pilgrims at a Marian apparition site. The hypothesis
tested was that many features of the discourse used in pilgrims' narratives are
manifestations of an underlying conflict of establishing the authenticity of a
religious experience. It was proposed that pilgrims are under pragmatic pressures
to reduce their agency in the events described, in order to attribute divine authority
to those experiences, and that one linguistic manifestation of this pressure would
be pilgrims' increased use of non-visual metaphors for knowledge, as a way to
emphasize a passive relationship to the supernatural.
This was tested by comparing pilgrims' use of visual and non-visual
metaphors for knowledge in their discourse. It was found that pilgrims consis
tently used more non-visual than visual metaphors for knowledge, but that they
did so in different contexts of communications with different demand characteris
tics: when discussing religious events, in secular narratives, and in un-elicited
narratives communicated among fellow believers. These results could support the
hypothesis that pilgrims routinely reduce their own agency by use of non-visual
perceptual metaphors, but alternate explanations cannot be ruled out until this
study is replicated on a control group of "mainstream" Americans.
126 VICTOR BALABAN
2. Pilgrims' narratives
In 1989 Nancy Fowler began having visions of the Virgin Mary. Since that
time, her home in Conyers, Georgia has become one of the largest and longest
lived Marian Apparition sites in the United States (Balaban 1996a, 1996b, in
press). As many as 80, 000 people gather on the 13th of every month to hear Ms.
Fowler relate a message for the United States from the Virgin. Pilgrims visit and
observe devotions such as taking Holy Water from a Holy Well, leaving petitions
and prayers in baskets in the Apparition Room, and taking pictures of the sun to
look for signs of Jesus and Mary in the images. Pilgrims come to Conyers for
healing, to give thanks, or to reaffirm their faith; but whatever the explanation
given, the implicit reason is almost always to directly experience for themselves
some of the graces of Mary.
After Sunday prayer meetings pilgrims in Conyers will mill around on the
Holy Hill and tell each other their stories. Although the events are told informally,
they are by no means unimportant. Actually seeing or hearing the Virgin Mary is
considered to be a rare privilege that only a very few are ever granted. Most
pilgrims experience a host of traditional signs and confirmations of Mary's
presence such as seeing the sun moving in the sky, seeing the appearance of Mary
in miraculous photographs, seeing rosary beads changing to gold, or smelling the
scent of roses. Other events might include healings, conversions or other life
changes. The common thread among these narratives is that they are all stories of
encounters with the divine in one form or another. For both the narrators and the
audience, these signs testify to the continual presence of Mary and confirmation
of the fact that She does intervene in the lives of the faithful, particularly at these
holy places. In the process, these divine encounters are transformed from private
experiences into meaningful social rituals (Zimdars-Swartz 1991). Telling the
story allows believers to both celebrate and reaffirm this transformation in their
lives (Stromberg 1992).
The narratives that the Conyers pilgrims tell have many features in common
with a narrative genre that has been well documented: conversion narratives.
Conversion narratives describe the experience of how the presence of the divine
entered an individual's life and transformed them, usually in context of how they
became a believer. Stromberg (1992) analyzed a series of Evangelical Christian
conversion narratives and concluded that these narratives can be understood as a
form of ritual performance where the narrator can express and come to terms with
contradictory aims. In a conversion narrative, the believer can either fulfill a
purpose in their relationship to God that might be forbidden in other contexts; or
else attribute unacknowledged aims to God, so that God does what the believer
cannot or should not do. In this way, believers can invoke conflicts in their lives
and accommodate discordant or otherwise disturbing thoughts and actions by
SELF AGENCY IN RELIGIOUS DISCOURSE 127
reframing them in the canonical language of Evangelical Christianity. Once these

undesirable elements are attributed to an outside entity, there are no inconsisten
cies any more and the dilemma is solved. In a simple sense, the supernatural agent
does what the person cannot or feels they should not do or feel.
The difficulty for the pilgrims in Conyers lies in that fact that the secular
culture of the United States does not accept the reality of apparitions (Stromberg
1992). The popular assumption is that people who have experiences similar to
those described in conversion narratives are either mistaken or possibly even
unstable. Although many clinical studies have found that hallucinations and
illusions often occur in non-psychotic individuals (see e.g. Bliss & Clark 1962),
the folk model is such that any reported vision or supernatural encounter can be
considered a sign of instability. As members of the wider culture of the United
States, the pilgrims believe this too. (This is, of course, in no way meant to
suggest that any of the people who have religious experiences in Conyers are
hallucinating, only that in American society describing such an experience has a
high probability of being interpreted as such.)
Therefore, when telling their religious narratives, pilgrims are negotiating
between two very different belief systems. One is within the local culture of the
Apparition Site, where it is acceptable to attribute thoughts and actions to an
outside agent; and the second is in the wider, secular culture where this way of
thinking can be considered a sign of loss of control. This can be thought of as an
attempt to negotiate between two contradictory cultural models.
3. Cultural models
Recent work in cognitive anthropology has led to a growing recognition of
the role of cultural models, cognitive schemata that are shared by a cultural group
(D'Andrade 1990). It has been argued by cognitive anthropologists that much of
our everyday social life is mediated by these cultural models which organize
experience, create expectations, motivate behavior, and provide a framework for
people to remember, describe and reconstruct events (see e.g. D'Andrade 1990;
D'Andrade and Strauss 1992; Dougherty 1985; Holland and Quinn 1987).
Multiple models of social experiences are clearly possible, and many competing
and contradictory values can coexist in the same cultural systems (see e.g. Geertz
1983; R. Rosaldo 1989; Shore 1996; Sperber 1985).
For example, Quinn (1985, 1987) has illustrated the presence of multiple
cultural models in her analysis of American cultural models of marriage. In these
studies, narratives of subject's accounts of their marriages are analyzed, and
different underlying metaphors of marriage are identified. These metaphors reflect
an underlying cultural model of marriage and in turn seem to be based on four
abstract schemas: RELATION i.e. marriage is something between two people,
128 VICTOR BALABAN
CONTAINER i.e. we sailed into the marriage, TRAJECTORY i.e. that was a turning
point in our marriage, and ENTITY i.e. marriage is an institution. Quinn shows
how each model highlights different aspects of the experience of marriage, and
different speakers may use different schemas.
Discussing marriage in terms of one metaphor as opposed to another
presumably has important consequences for whether a couple thinks of their
relationship as something that can be "fixed" or as a trip that has "come to an
end". When pilgrims to Conyers attempt to explain apparently supernatural
experiences, which model of self and agency a speaker chooses has crucial
consequences for how the narrator is perceived. The difficulty has to do with
fundamentally different models of agency and intentionality that are implied by
the two different models. In order to describe an experience of the supernatural,
a speaker must invoke a specific model of self, a model that allows people to have
thoughts and experiences that are attributable to an outside agent. The pilgrims to
Conyers are, nonetheless, perfectly aware that secular culture operates under a
different model of self, and that invoking a more passive model may bring with
it the implication that the speaker may be mistaken or unstable.
Describing a supernatural experience is thus not a simple process, but a
negotiation between different, mutually inconsistent, cultural models of self and
agency. Only if pilgrims can be positive that their experiences are divinely
inspired can they be justified in identifying their experiences as the works of an
intentional divinity. The way this conflict is manifested is through pragmatic
pressures on the pilgrims to justify the authenticity of their experiences.
The premise of this study is that the language used in pilgrims ' narratives
will reflect this ongoing process of negotiation, as personal experiences are made
to fit pre-existing schemas. Many scholars have described this process, which
Zimdars-Swartz (1991:19) calls "the transformation of private experience to
public meaning"; Obeyesekere (1981:50) calls "the conventionalization of
personal symbols"; Slater (1990:109) calls "the exteriorization of private
experience"; and Shore (1996) discusses in terms of the relationship between
shared social models and more individual cognitive models. For the pilgrims to
Conyers, recounting a religious experience creates a particular social reality
(Rappaport 1977), and a particular identity (Stromberg 1992). The way this is
accomplished is through language.
4. Language and agency

One of the key insights of recent work in cognitive linguistics is that
language is grounded systematically in cognition. According to this view,
grammatical regularities do not arise in language as a result of the application of
formal rules. Instead, it is proposed that the use of various linguistic devices are
organized around clusters of related notions such as agency, control, volition and
animacy (e.g. Comrie 1981; DeLancey 1984; Givon 1979, 1984; Hopper and
Thompson 1980, 1994; Lakoff 1987; Langacker 1991, 1994; Slobin 1981, 1985).
In this view, grammar is shaped by all the factors - cognitive, social, interactive
and cultural - that are involved in how language is used. Regular patterns of
grammatical usage arise as a result of strategies that speakers use in negotiating
what they want to communicate to listeners. Contrary to many earlier views of
language, this means that the relationship between grammar and semantics is quite
flexible, and implies that one needs to understand the demands of particular
communicative interactions in order to understand how the event is being
grammaticalized.
In the case of a conversion narrative, narrators are attempting to express
unacknowledged, inconsistent or unacceptable thoughts and/or actions into an
acceptable form (Stromberg 1992). The key to understanding this sort of discourse
is the central dynamic of attributing discordant thoughts and feelings to an
external, divine agent. Many theorists have noted the importance of the speakers'
lessening agency as a central element of ritual discourse. Geertz (1973) discussed
this in terms of anonymization in ritual; and Levi-Strauss (1969:18) wrote that
myth is "a message that, properly speaking, is coming from nowhere". Hymes
(1972:49) and Labov (1972:353) discuss the depersonalizing aspect of communi
cation and ritual through the use of formulaic speech; and Dubois (1986:330)
writes "the first step in establishing authority for ritual speech is to make it appear
to be of a personal origin, or at least to approach this".
Studies of the language used in evangelical Christian groups confirm these
theorists' assertions. Coleman (1980) found that "born-again" Evangelical
Christians systematically avoided using transitive sentences with first-person
subjects as a way to avoid portraying themselves as volitional agents. She
characterized evangelical language as an effort to conceal the nature of the
narrator's actions and the speaker's role in them. Szuchewycz (1994) found that
participants in a Charismatic Catholic prayer meeting used linguistic constructions
that reduced their agency by attributing divine authority and authorship to their
own speech acts. Harding (1987) also found that the rhetorical devices used in
Baptist conversion narratives functioned to redescribe personal experiences in
divine terms.
The prototypical model that underlies the notion of agenthood in American
society is that of a responsible, in-control human who intentionally performs an
action (Croft 1994; Langacker 1991). Bates and MacWhinney (1982:217)
suggested that prototypical agents are physical entities which have features such
as "animacy", "intention", "cause", and "human". Van Oosten (1985:213) studied
a large corpus of English conversations and concludes that the central characteris-
130 VICTOR BALABAN
tics of a prototypical agent are that it has the primary responsibility for the action
that occurs in the sentence, and that it is intentional. DeLancey (1984), using
examples from English, Newari and Hare, further agues that different languages
emphasize different aspects of these attributes of agentivity. DeLancey found that
the general tendency in English is to equate the source of an utterance with the
speaker. This is at odds with the interpretation sought by the pilgrims, which is to
locate the source as being outside the speaker. This makes the narrator a
non-prototypical agent and has the effect of making most intentional agents human
since only humans are generally presumed to have intentionality (Van Oosten
1985; Langacker 1991).
In purely linguistic terms, there should be no difficulty for an English
speaker to attribute thoughts and feelings to an outside agent. There are a variety
of linguistic devices that can be used to mark speakers as atypical (non-volitional)
agents. However, in the case of the pilgrims to Conyers, there is a complication:
the kind of non-prototypical agent they are trying to convey, one that attributes
thoughts and feelings to an outside agent, is the prototype for another category of
agent, the category of mentally or emotionally unstable agents.
Thoughts or actions that are inconsistent or contradictory are considered
troubling in American society. These notions of agency map on to conceptions of
mental health and illness in American culture, where there are folk-beliefs that
people are agents with coherent intentions. Many symptoms of emotional or
mental illness do have the quality of having been produced without the individual
having had any intent of doing so. Symptoms such as unintentional movements,
memory lapses, compulsive thoughts and behaviors, overly intense or labile
emotional states, all have the quality of being "ego-alien" (Stromberg 1992).
Therefore, the pilgrims find themselves in a bind. They are motivated to tell
their religious narratives to as many people as possible in order to give witness to
the glory of God, but in so doing they risk reinforcing the stereotype of religious
devotees as being unstable. On the other hand, if pilgrims portray themselves as
more typical agents, they run the risk of having their experiences interpreted as
ordinary events. This is an insoluble dilemma. One way for pilgrims to negotiate
this impasse is to try to construct their accounts in such a way that not only is the
authenticity of a reported apparition established, but it is also distinguished from
other possible explanations of the same phenomena i.e. hallucination, coincidence,
etc. This can be accomplished by presenting the thoughts and feelings in such a
way that they had to have come from an external agent, but at the same time to use
a variety of linguistic devices to mark the source and reliability of knowledge.
To summarize, it is hypothesized that the pilgrims in Conyers need to assert
that their thoughts or feelings come from an outside agent, but at the same time
make clear the speaker's control over his/her own mind. The way this can be done
is to employ linguistic devices that shift the speaker to a non-volitional role; and
at the same time use other devices that either mark the source and reliability of the
speaker's knowledge, or else highlight the reliability of the speaker themselves.
In this way, the conflict can be negotiated. It is hypothesized that one linguistic
manifestation of this pressure will be pilgrims' use of non-visual metaphors for
knowledge, as a way to emphasize the passive relationship to the divine that
pilgrims are attempting to convey in their narratives.
5. Perceptual metaphors for knowledge

Cognitive linguists propose that language use reflects inherently metaphori
cal understanding of many areas of experience. Metaphor, in this view, is not just
a linguistic device, but a central organizing factor in language and cognition
(Johnson 1987; Lakoff 1987; Lakoff and Johnson 1980).
Metaphor, in cognitive linguistic theories, is characterized as a process
where one experiential domain is conceptualized in terms of another (Taylor 1989;
Sweetser 1990). Briefly, cognitive linguists propose that language use reflects
inherently metaphorical understanding of many areas of experience (Lakoff &
Johnson 1980; Lakoff 1987; Johnson 1987; Sweetser 1990). Metaphors are
systematic interrelations of multiple expressions which map one relatively stable
'domain' on to another. The basic process is that a concrete domain (the source
domain) is mapped on to a more abstract domain (the target domain). An example
which will be described in more detail later is the fact that the target domain of
"knowledge" is often understood in terms of the source domain of "vision" i.e.
Your logic is crystal clear.
Many cognitive linguists have noted that the same terms that pertain to
sensory perception are also used metaphorically to indicate specific aspects of
thought or knowledge (e.g. Lakoff & Johnson 1980; Lakoff 1987; Johnson 1987;
Sweetser 1990, 1995). For instance, verbs that refer to vision (i.e. I see) are used
to refer to knowing or understanding (i.e. I see what you mean). Sweetser (1990,
1995), argues that a coherent and structured system of metaphors underlies and
motivates the tendency to use vocabulary from peoples' external (social and
physical) domains when discussing internal (psychological and emotional)
domains.
Perceptual metaphors are motivated by folk models of different perceptual
abilities. For example, vision is considered to be the most objective and verifiable
sense and is generally considered to be peoples' primary source of objective
information about the world. Target domains of vision verbs commonly develop
abstract senses of mental activity and intellect, either knowledge or mental
'vision'. The source domain for this is generally considered to be the greater
channeling and focusing ability that vision affords. One can visually "pick out"
132 VICTOR BALABAN
and attend to specific items amid many other stimuli. Examples of this would
include not only obvious uses such as I see what you mean, but more abstract uses
such as In my view or His argument threw up a smoke screen. In all these cases
visual terms refer not to actual visual perception, but to understanding or
knowledge. Furthermore, the limited domain of vision and the ability to monitor
give rise to metaphors of control.
Matlock (1989) argues that an underlying KNOWING IS SEEING model
motivates the use of visual perceptual metaphors for knowledge. The KNOWING
IS SEEING metaphor is highly conventionalized and is found in most Indo-Euro
pean languages (Matlock 1989; Sweetser 1990). KNOWING IS SEEING is an
example of the broader MIND AS BODY metaphor (Lakoff & Johnson 1980;
Sweetser 1990) where the internal self is pervasively understood in terms of the
bodily external self, and hence is described by means of vocabulary drawn from
the physical domain. Matlock (1989) asks why vision is used as a source domain
to structure knowledge, rather than hearing or any other sensory system. She
explains this by noting that vision is associated with more certain and direct
knowledge, while other senses are more associated with indirect or inferred
knowledge. In simpler terms, seeing is believing.
Different sensory verbs, with different source domains, can also be used as
metaphors for knowledge (see table 1 for examples). Taste and touch require
physical contact, and distance is connected with objectivity and intellect, closeness
with subjectivity, intimacy and emotion. Therefore metaphors of taste and touch
usually have deeper emotional connotations. Taste generally has deep links to
internal self, e.g. The whole thing left a bad taste in my mouth, and touch maps on
to emotional sensations, e.g. I was so touched by your letter. Smell has few
metaphorical connections to mental domains, but can still be used, e.g. I smell an
opportunity here. In the auditory domain, it is more difficult to localize a sound
than to visually focus on an object; therefore hearing is often connected with the
communicative aspects of understanding rather than with cognition in general.
Since hearing is the primary mode for understanding language, and hence for
influencing people, either intellectually or emotionally, verbs of hearing often
come to mean "to be receptive", "to heed" or "to obey". An example of this would
be the general acceptance communicated by I hear you or You 're coming in loud
and clear as opposed to the more active I see what you mean (Sweetser 1990:42).
Since pilgrims are under pragmatic pressures to reduce their agency in the
events described, it is possible to hypothesize about what perceptual metaphors
for knowledge could be used to convey reduced agency. Vision, as noted above,
tends to be connected with the intellect and uses the active monitoring and
focusing abilities that visual perception affords as a source domain. Metaphors
based on vision, therefore, presume an active agent in the world. Such an active
agent would be inconsistent with the more passive, receptive relationship to the
divine that pilgrims are seeking to convey. A more passive agent might be
conveyed by metaphors based on other sensory modalities. It is therefore
hypothesized that pilgrims to Conyers will use non-visual metaphors for
knowledge more than visual metaphors as a way to portray themselves as passive
recipients of divine experiences by reducing their agency in the events described.
6. Method
6.1 Participants and materials
A total of 191 narratives were collected from pilgrims in Conyers, Georgia;
and from the Apparitions-List (APAR-L). The face-to-face narratives consisted
of 17 tape recorded testimonies collected from pilgrims in Conyers between
October 1996 and March 1997. When conducting face-to-face interviews, I
explained to all participants that I am respectful of their beliefs, I am not Catholic
(I am Jewish) and I am a graduate student in psychology. All 17 participants then
answered an open-ended question "Is there any testimony you would like to
share?" Of those participants, 14 also agreed to tell of a secular experience. The
prompt question used was "How did you meet you husband/wife?" Pilot studies
showed that pilgrims did in fact describe how they met their spouses in non-reli
gious terms, in that none mentioned God or any other supernatural agent being
involved. Taped interviews were transcribed verbatim.
The on-line narratives were 55 un-elicited e-mail messages from the
APAR-L electronic mailing list (a forum for believers to discuss their experiences
with Mary), and 5 on-line testimonies sent to the author in response to requests for
testimonies posted to the same list.
6.2 Coding and analyses

The collected narratives were divided into four groups: 1) Face-to-Face
Religious Narratives, 2) Face-to-Face Secular Narratives, 3) Un-Elicited On-line
Religious Narratives, and 4) Elicited On-line Religious Narratives.
Each group of narratives was analyzed for each instance of the use of
perceptual metaphors for knowledge. Perceptual metaphors were operationalized
as any use of perceptual vocabulary to indicate knowledge. Eight different
modalities of perceptual metaphor were coded for: 1) visual, 2) auditory, 3)
olfactory, 4) taste, 5) tactile, 6) unspecified, 7) non-perceptual, biological
metaphors (metaphors for knowledge that have some aspect of human physiology
other than perceptual systems as a source domain); and 8) non-biological
metaphors that use mechanical or other non-biological source domains (see table
1 for examples). The mean frequency of use of Non-Visual metaphors was also
134 VICTOR BALABAN
coded for, with Non-Visual metaphors being defined as all the different types of
perceptual metaphors, except Visual, added together.
Two metaphors were excluded from the analysis: the visual metaphor
contained in the expression let' s see, and the tactile metaphors contained in
expressions of emotion such as I felt sad. Both usages were considered to be so
highly conventionalized that they probably do not provide any indication of
pilgrims underlying conceptual models.
The mean frequency of use of each type of metaphor was then expressed as
a proportion of the total number of metaphors used per subject
Four group comparisons were conducted: a) religious vs. secular experi
ences, b) elicited vs. non-elicited accounts, c) face-to-face vs. on-line narratives,
and d) and perceptual metaphors vs. perceptual vocabulary. An analysis of
variance (ANOVA) showed that the four groups of narratives differed signifi
cantly in length, F(1, 3)=36.97, p<.001. The longest narratives were the Elicited,
On-Line narratives (M = 287.4 propositions), followed by the Face-to-Face
Religious narratives (M = 166.53), the On-Line Un-Elicited narratives (M =
42.07), and the shortest were the Face-to-Face Secular narratives (M = 25.5). A
comparison of just the Face-to-Face narratives showed that the religious narratives
were significantly longer than secular narratives, t(13)= 4.25, p<.001.
7. Results and discussion

The main hypothesis tested in this study was that pilgrims would use more
non-visual metaphors than visual metaphors for knowledge in their religious
testimonies as a way to reduce their own agency in the events they described.
Overall pilgrims used metaphors in 5.9% of their propositions. The most
common metaphors were Non-Biological metaphors such as: it's starting to
compute, or the doorway is starting to creep open in my brain, or l am so grateful
to Godfor the gifts He's given me or Ifeel as if I haven't got the direct line to god
yet. Pilgrims also used a great many tactile metaphors when describing their
religious experiences. Examples include: I feel as though I was truly physically
touched by Our Blessed Mother's broken Heart, or Ifelt gripped by the Spirit, or
My life was touched on that day, or the Lord started really knocking me over the
head. In addition there were many Biological metaphors used that were
Non-Perceptual, including: it was the Blessed Mother's way of giving the boys a
kiss on the cheek, or I knew it in my heart, or my whole brain was going back and
forth, doubt, worry, or my spiritual growth took root 14 years ago. There were
very few auditory metaphors used, although there were a few: and [the Lord] said
"Wake up!", or the ears perked up, or For the longest time I felt that God was
calling me to something. Although the majority of pilgrims' metaphors for
knowledge were Non-Visual, there were many Visual metaphors used as well: it
dawned on me, or Jesus showed me the way, or He was showing me little truths
of the faith. There were no Olfactory or Taste metaphors used by any of the
pilgrims.
A 4(narrative type) x2(visual vs. non-visual metaphors) analysis of variance
(ANOVA) was performed to test whether the frequency of use of visual and
non-visual perceptual metaphors for knowledge differed among groups. No
significant effects of group were found on the frequency of use of either visual or
non-visual perceptual metaphors. Pilgrims used more non-visual {M =.91) than
visual metaphors (M =.09) for knowledge, F(1, 90) = 2.86, p<.01. There was no
interaction effect, indicating that in all four groups of narratives this pattern holds,
as can be seen in figure 1.
There are several factors in addition to the desire to reduce speaker agency that
might affect how speakers use perceptual metaphors for knowledge. Since the
main purpose of this study is to test how the pragmatic pressures of recounting a
religious experience are manifested in discourse it is necessary to have control
groups, in order to assess how pilgrims' use of metaphors is changed in different
contexts of communication when they are not under those pressures. Four group
comparisons were conducted: a) religious vs. secular experiences, b) elicited vs.
non-elicited accounts, c) and face-to-face and on-line narratives, and d) metaphori-
136 VICTOR BALABAN
cal vs. non-metaphorical perceptual vocabulary. Each of these comparisons is

explained in more detail below.
7.1 Religious vs. secular experiences

When describing secular events, pilgrims are presumably under less pressure
to reduce their own agency or assert the source of their knowledge. It could
therefore be hypothesized that when discussing secular experiences, pilgrims
would show a different pattern than when recounting religious events and use
more visual metaphors than non-visual metaphors for knowledge. This was tested
by comparing the use of metaphors in the Face-to-Face Religious and Secular
narratives.
Pilgrims used significantly more non-visual (M =.84) than visual (M =.1)
metaphors for knowledge, t(16) = 10.83, p<.001. When recounting a secular
experience, pilgrims did not use significantly more non-visual (M =.35) than
visual (M = 0) metaphors for knowledge. This might indicate that pilgrims' use
of more non-visual than visual perceptual metaphors for knowledge in their
discourse is limited to the context of recounting a supernatural experience.
However, the fact that pilgrims used exclusively non-visual metaphors for
knowledge in the Secular narratives suggests that this pattern is not limited to
religious discourse and that a larger sample might have shown significant
differences.
7.2 Elicited vs. non-elicited accounts

One unavoidable confound of collecting interviews at Conyers is that the
pilgrims are recounting their experiences to a non-believer (the author) and so may
be speaking in ways that reflect the demand characteristics of the interview
situation. This was tested by collecting pilgrims' accounts of their religious
experiences gathered from the archives of the Apparitions List (APAR-L), an
electronic mailing list program devoted to distributing the Virgin's messages
on-line.
These On-Line accounts, unlike the Face-to-Face Religious narratives, were
not elicited and presumably were written to be shared with a group of fellow
believers. Elicited and Un-Elicited interviews were then compared to assess
whether or not there are differences in the use of perceptual metaphors for
knowledge in response to the interview situation.
If the task of explaining supernatural events to a non-believer forces subjects
to use discourse strategies that convince the listener of the genuineness of their
experiences, then it would be hypothesized that the Un-Elicited religious
narratives would not show the same pattern of metaphor use as the Elicited
narratives.
This hypothesis was not supported. As in the Elicited narratives, there were
significantly more non-visual (M =.63) than visual (M =.07) metaphors for
knowledge in the Un-Elicted narratives, t(54)=-7.64, p<.001. Therefore pilgrims'
use of more non-visual than visual perceptual metaphors for knowledge in their
religious narratives was not simply the result of demand characteristics of
speaking to a non-believer.
7.3 Face-to-Face vs. on-line narratives

The Un-Elicited narratives in the previous analysis are communicated
On-Line and therefore might not have been directly comparable to the oral
Face-to-Face narratives (see e.g. Jones 1995; Spears and Lea 1992). As a way of
compensating for this, on-line testimonies were elicited from members of APAR-L,
providing a final control, a group of Elicited On-Line narratives.
There were significantly more non-visual (M =.94) than visual (M =.06)
metaphors, in pilgrims' Elicited On-line Religious Narratives, t(4)= 11.89,p<.001.
Therefore pilgrims' use of more non-visual than visual perceptual metaphors in
their un-elicited narratives is not just a function of their being communicated
on-line.
7.4 Metaphorical vs. non-metaphorical perceptual vocabulary

One final alternative explanation for any observed pattern of use of
perceptual metaphors that must be ruled out is the possibility that pilgrims might
simply be reporting more non-visual than visual experiences. Pilgrims' use of
perceptual metaphors might not be related to the context of communication, but
simply reflect the patterns of use of non-metaphorical perceptual vocabulary
overall. Therefore the use of non-metaphorical perceptual vocabulary in the
narratives was also analyzed.
Non-metaphorical perceptual vocabulary was operationalized as any use of
perceptual vocabulary to indicate direct sensory experience. Six different
modalities of perceptual vocabulary were coded for: 1) visual, 2) auditory, 3)
olfactory, 4) taste, 5) tactile, and 6) unspecified. The mean frequency of use of
each type of perceptual vocabulary was then expressed as a proportion of the total
amount of perceptual vocabulary used per subject.
A 4 (narrative type) x2 (visual vs. non-visual perceptual vocabulary)
analysis of variance (ANOVA) was performed to test whether the frequency of use
of non-metaphorical visual and non-visual perceptual vocabulary differed between
groups. In contrast to the pilgrims' use of metaphors, main effects of group were
found on the frequency of use of both visual (F(1, 3)= 2.80, p<.05) and non-visual
(F(1,3)= 5.55, p<.01) perceptual vocabulary, as well as an interaction effect (F(12)=
19.05, p<.001). Also in contrast to their pattern of use of metaphors, pilgrims used
138 VICTOR BALABAN
more visual (M =.45) than non-visual (M =.21) non-metaphorical perceptual

vocabulary, F(1, 90) = 1.78, P<05.
Pilgrims used more visual than non-visual non-metaphorical vocabulary in
their Un-Elicted On-Line narratives (visual M =.50, non-visual M =.15, t(54) =
-3.93. p<.001). There was no significant difference in their use of visual and
non-visual perceptual vocabulary in Face-to-Face Religious Narratives (visual M
=.46, non-visual M =.48), when recounting a Secular experience (visual M=.14,
non-visual M=.07), or in their Elicited On-line Religious Narratives, (visual M
=.64, non-visual M =.36) (see figure 2).
Figure 2: Non-metaphorical perceptual vocabulary
The pattern of pilgrims' use of non-metaphorical perceptual vocabulary therefore

is very different than their use of perceptual metaphors for knowledge, so it seems
that pilgrims' use of metaphors does not simply reflect patterns of use of
perceptual vocabulary overall.
8. Conclusion
Pilgrims to the Marian Apparition site in Conyers, Georgia used more
non-perceptual than perceptual metaphors for knowledge in their religious
narratives. These results are consistent with the initial hypothesis that pilgrims are
attempting to reduce their own agency by use of non-visual perceptual metaphors

when describing their religious experiences. However, they consistently used more
non-visual than visual metaphors across a range of conditions with different
demand characteristics: when telling of a secular experience, in un-elicited
accounts, and on the Internet.
There are at least three different hypotheses that could explain these results.
First, it might be the case that all Americans habitually use more non-visual than
visual metaphors for knowledge in all areas of discourse. If this were the case,
then these results would indicate that pilgrims simply do not use perceptual
metaphors for knowledge to reduce their agency. A related explanation could be
that meeting one's spouse is perceived by Americans as embodying a similar loss
of control as a religious experience. This would mean that pilgrims are using
perceptual metaphors to reduce their agency, and that they do so in the same way
that Americans do in other situations where there is a loss of control involved.
A third explanation might be that pilgrims consistently attempt to present
themselves as non-volitional agents in all areas of discourse, something similar to
what Coleman (1980) calls a "worldviewlect". This explanation would be
supported if it were to be found that mainstream Americans do not use the same
kinds of metaphors when discussing how they met their spouses, although such
a study has not yet been done. Without a control group of "mainstream"
Americans discussing similar events, there is no way to eliminate any of these
hypotheses at this time.
There are several factors in particular that need to be explored in more
depth. The first is the fact that the Face-to-Face Secular narratives are so much
shorter than the Face-to-Face Religious narratives. The fact that these secular
narratives are so brief suggests that the demand characteristics of the interview
situation, with the secular questions being presented as an informal follow-up to
the requested testimony, may have affected how the pilgrim constructed their
replies. If so, these secular accounts may not be a reflection of how pilgrims might
talk about their marriages in other contexts.
A far larger question that remains to be addressed is how representative the
pilgrims' discourse styles are of members of American society. These results
highlight the need for a control group of "mainstream" Americans. It is entirely
possible that most members of American society use predominantly non-visual
metaphors when describing how they met their spouses. If this were the case, then
the discourse style used by the pilgrims in Conyers might not necessarily be a
reflection of any religious influence.
Another important element that needs to be better understood is the role of
creative vs. conventionalized metaphors (see Gibbs 1994 for a review). Although
highly conventionalized metaphors such as lef s see and I feel were not included
140 VICTOR BALABAN
in this analysis, pilgrims' use of metaphors ranged from the conventionalized, e.g.
I heard the call, or My whole life was straightened out, to some extremely creative
metaphors, e.g. I was helped along by a drive-by shooting of grace from the Holy
Spirit! or, It was like all the poison was coming out of my body. It is not clear at
this point if more conventionalized or more creative metaphors are used in
different contexts or for different functions.
Finally, studies such as this one can help shed light on a fundamental
theoretical disagreement about the nature of metaphor. Theorists such as Johnson
(1987), Lakoff (1987) and Sweetser (1990) have argued that basic level metaphors
structure our understanding of the world and that underlying metaphors motivate
subsequent language use. Other theorists, notably Quinn (1987, 1991), Strauss and
Quinn (1998) argue the opposite, that the point a speaker wants to make motivates
the metaphor they choose. In this view, the results of this study could be
interpreted as reflecting widely shared cultural models instead of underlying
cognitive models.
At this point there would be no way to distinguish whether the patterns of
metaphor usage by the Conyers pilgrims reflects something about the nature of
vision itself or whether this is an example of a very widely used cultural model.
It would be interesting to see what perceptual metaphors for knowledge are used
in non-Indo-European languages, and perhaps even more importantly, what
perceptual metaphors for knowledge one finds in deaf ASL signers where
language reception is primarily visual and not auditory. However, I feel that the
methodology being developed in this study could provide a way to begin to test
which metaphorical usages are more contextually motivated, and which others are
so consistently and widely used that they might be considered to be universal.
These preliminary results, while suggestive, are not conclusive. They do,
however, show how quantitative and ethnographic methodologies can be fruitfully
combined to study the role of metaphor in cognitive linguistics. Analyzing how
linguistic devices are used in different contexts of naturally occurring discourse
within a specific local culture can help to delineate the cognitive, social, pragmatic
and cultural factors that influence the forms which narratives can take, which in
turn can lead to a better understanding of exactly how language is grounded in
cognition.
MODALITY NON- PERCEPTUAL

METAPHORICAL METAPHORS
PERCEPTUAL F O R KNOWLEDGE
VOCABULARY
VISUAL I saw the sun move. I saw that I was sinner.

I watched it happen. The meaning was clear.
AUDITORY I heard the thunder. I hear and obey.

I listened to them speak. You're coming in loud and
clear.
OLFACTORY I smelled roses. I smell an opportunity.

The deal just stinks.
TASTE The wine tasted sweet. The experience left a bad

taste in my mouth.
TACTILE I touched my rosary. I was touched.

I felt a hug land on me. It was like Jesus hitting me
over the head.
UNSPECIFIED I sensed someone was near. I could tell things were

I experienced something. changing.
I have noticed that my faith
has increased.
BIOLOGICAL NA I knew it in my heart.

NON-PERCEPTUAL
NON-BIOLOGICAL NA The doors in my mind open-

ed.
The wheels in my head start-
ed to turn
Table 1: Examples of Metaphorical and Non-Metaphorical Perceptual Vocabulary
References
Balaban, V. 1996a. 'The Marian Apparition Site at Conyers, Georgia". Religions
of Atlanta, ed. by G. Laderman. 215-241. Atlanta, Ga.: Scholars Press.
1996b. "Photoessay: The Marian Apparition Site at Conyers, Georgia".
Religious Studies News.
In press. "Religious Metaphor and Cognitive Linguistics". Proceedings
of the 1997 LAUD symposium, ed. by R. Dirven, & R. Bischoffs. London:
Peter Lang.
142 VICTOR BALABAN
Bates, E., & B. MacWhinney. 1982. "Functionalist Approaches to Grammar".

Language acquisition: The State of the Art, ed. by L. Gleitman, & E.
Wanner, 173-218. Cambridge: Cambridge University Press.
Bliss, E.L., & L.D. Clark. 1962. "Visual Hallucinations". Hallucinations, ed. by
L.J. West, 92-107. New York: Grune and Stratton.
Coleman, L. 1980. "The Language of 'Born Again' Christianity". Proceedings of
the Berkeley Linguistics Society 6. 133-142.
Comrie, B. 1981. Language Universals and Linguistic Typology. Chicago:
University of Chicago Press.
Croft, W. 1994. "Voice: Beyond Control and Affectedness". Voice: Form and
Function, ed. by B. Fox, & P.J. Hopper, 89-127. Amsterdam/Philadelphia:
John Benjamins.
D'Andrade, R. 1990. "Some Propositions about the Relations between Culture
and Human Cognition". Cultural Models in Language and Thought, ed. by
N. Quinn, & D. Holland, 112-48. Cambridge: Cambridge University Press.
DeLancey, S. 1984. "Notes on Agentivity and Causation". Studies in Language
8. 181-213.
Dougherty, J.W.D. 1985. Directions in Cognitive Anthropology. Urbana:
University of Illinois Press.
Dubois, J. 1986. "Self-Evidence and Ritual Speech. Evidentiality: The Linguistic
Coding of Epistemology, ed. by W. Chafe, & J. Nichols, 313-333. Norwood,
N.J.: Ablex.
Durand, A. 1994. From Grammar to Politics: Linguistic Anthropology in a
Western Samoan Village. Berkeley: University of California Press.
Geertz, C. 1973. "Religion as a Cultural System". The Interpretation of Cultures,
87-125. New York: Basic.
Gibbs, R.W. 1994. The Poetics of Mind: Figurative Thought, Language and
Givon, T. 1979. On Understanding Grammar. New York: Academic Press.
1984. Syntax: A Functional-Typological Introduction. Amster
dam/Philadelphia: John Benjamins.
Harding, S. 1987. "Convicted By the Holy Spirit: The Rhetoric of Fundamental
Baptist Conversion". American Ethnologist, 167-181.
Holland, D., & N. Quinn. 1987. "Culture and Cognition". Cultural Models in
Language and Thought, ed. by D. Holland, & N. Quinn, 3-40. Cambridge:
P.J. Hopper, & S.A.Thompson. 1980. Transitivity in Grammar and Discourse.
Language 56. 251-299. New York.
Hopper, P.J., & S.A. Thompson. 1994. "Language Universals, Discourse
Pragmatics and Semantics". Language Sciences 15. 357-376.
Hymes, D. 1972. "Models of the Interaction of Language and Social Life".

Directions in Sociolinguistics, ed. by J. Gumperz, & D. Hymes, 35-71. New
York: Holt.
Jones, S. 1995. Cybersociety: Computer Mediated Communication and Commu-
nity. Thousand Oaks, Calif.: Sage.
Labov, W. 1972. "Rules for Ritual Insults". Language in the Inner City: Studies
in the Black English Vernacular. Philadelphia: University of Pennsylvania
Press.
Lakoff, G. 1987. Women, Fire and Dangerous Things: What Categories Reveal
about the Mind. Chicago: University of Chicago Press.
Chicago Press.
Langacker, R.W. 1990. Concept, Image and Symbol: The Cognitive Basis of
Grammar. Berlin: Mouton de Gruyter.
1994. "Grammatical Traces of Some 'Invisible' Semantic Constructs.
Language Sciences 15. 323-355.
Levi-Strauss. 1969. The Raw and the Cooked. New York: Harper and Row.
Matlock, T. 1989. "Metaphor and the Grammaticalization of Evidentials".
Proceedings of the Berkeley Linguistics Society 15. 215-225.
Obeyesekere, G. 1981. Medusa's Hair: An Essay on Personal Symbols and
Religious Experience. Chicago: University of Chicago Press.
Quinn, N. 1985. "'Commitment' in American Marriage: A Cultural Analysis".
Directions in Cognitive Anthropology, ed. by J.W.D. Dougherty, 291-320.
Champaign: University of Illinois Press.
1987. "Convergent Evidence for a Cultural Model of American Marriage".
Cultural Models in Language and Thought, ed. by D. Holland, & N. Quinn.
195-221.Cambridge: Cambridge University Press.
, &C. Strauss. 1997. A Cognitive Theory of Cultural Meaning. Cambridge:
Rappaport, R. 1977. "The Obvious Aspects of Ritual". Ecology, Meaning and
Religion, ed. by R. Rappaport, 173-221. Richmond, Calif.: North Atlantic
Books.
Reingold, H. 1993. The Virtual Community: Homesteading on the Electronic
Frontier. Reading. Mass.: Addison-Wesley.
Rosaldo, R. 1989. Culture and Truth. Boston: Beacon Press.
Shore, B. 1996. Culture in Mind. Oxford: Oxford University Press.
Slater, Candace. 1986. Trail of Miracles: Stories from a Pilgrimage in Northeast
Brazil. Berkeley: University of California Press.
144 VICTOR BALABAN
Slobin, D. 1981. "The Origins of Grammatical Encoding of Events". The Child's

Construction of Grammar. London: Academic Press.
1985. "Crosslinguistic Evidence for the Language Making Capacity. The
Crosslinguistic Study of Language Acquisition, ed. by D. Slobin. Hillsdale,
N.J.: Erlbaum.
Spears, R., & M. Lea. 1992. "Social Influence and the Influence of the 'Social' in
Computer-Mediated Communication.". Contexts of Computer Mediated
Communication, ed. by M. Lea, 30-65. New York: Harvester Wheatsheaf.
Sperber, D. 1985. On Anthropological Knowledge. Cambridge: Cambridge
University Press.
Stromberg, P. 1992. Language and Self-Transformation: A Study of the Christian
Conversion Narrative. Cambridge: Cambridge University Press.
Sweetser, E. 1990. From Etymology to Pragmatics: Metaphorical and Cultural
Aspects of Semantic Structure. Cambridge:Cambridge University Press,
1992.
1995. "English Metaphors for Language: Motivations, Conventions, and
Creativity". Poetics Today 13. 705-724.
Szuchewycz, B. 1994. Evidentiality in Ritual Discourse: The Social Construction
of Religious Meaning. Language in Society 23. 389-410.
Taylor, J.R. 1989. Linguistic Categorization: Prototypes in Linguistic Theory.
Oxford: Oxford University Press.
To Bear Witness that I Am the Living Son of God, Reported Teachings and
Messages to the World from Our Lord and Our Loving Mother in Conyers,
Georgia Volume I.
Traugott, E.C. 1989. "On the Rise of Epistemic Meanings in English: An
Example of Subjectification in Semantic Change". Language 65. 31-55.
1982. "From Propositional to Textual and Expressive Meanings: Some
Semantic-Pragmatic Aspects of Grammaticalization". Perspectives on
Historical Linguistics, ed. by W.P. Lehmann, & Y. Malkiel, 245-271.
Amsterdam/Philadelphia: John Benjamins.
Turner, V., & E. Turner. 1978. Image and Pilgrimage in Christian Culture:
Anthropological Perspectives. New York: Columbia University Press.
Van Oosten, J. 1985. The Nature of Subjects, Topics and Agents: A Cognitive
Explanation. Bloomington: Indiana University Linguistics Club.
Zimdars-Swartz, S. 1991. Encountering Mary: Visions of Mary from LaSalette to
Medjugorje. Princeton, N.Y.: Princeton University Press.
TAKING METAPHOR OUT OF OUR HEADS AND
PUTTING IT INTO THE CULTURAL WORLD
RAYMOND W. GIBBS, JR.

University of California, Santa Cruz
Is metaphor linguistic, conceptual, or both? Despite centuries of widespread belief

that metaphor is a special linguistic, rhetorical device, much research in cognitive
linguistics over the past twenty years has demonstrated that metaphor is not
merely a figure of speech, but is a specific mental mapping that influences a good
deal of how people think, reason, and imagine in everyday life (Johnson 1987,
1993; Lakoff & Johnson 1980; Lakoff 1987; Lakoff & Turner 1989; Sweetser
1990; Turner 1991). A significant claim of this work is that many concepts,
especially abstract ones, are structured and mentally represented in terms of
metaphor. This claim is empirically supported by different research looking at
systematic patterns of conventional expressions, novel usages, and historical shifts
in word meanings, as well as work looking at the importance of metaphor in
grammatical forms.
A good deal of empirical evidence from cognitive psychology and
psycholinguistics has demonstrated the psychological validity of many of the ideas
about conceptual metaphor from cognitive linguistics (Gibbs 1994). For instance,
various psycholinguistic evidence supports the idea that metaphors such as ANGER
IS HEATED FLUID IN A CONTAINER are really conceptual and not, more simply,
generalizations of linguistic meaning. These include studies that have looked at
people's mental imagery for idioms (Gibbs & O'Brien 1990), people's con
text-sensitive use of idioms (Nayak & Gibbs 1990; Gibbs & Nayak 1991) and
euphemistic phrases (Pfaff, Gibbs, & Johnson 1997), people's folk understanding
of how the source domains in conceptual metaphors constrain what idioms (Gibbs
1992) and proverbs mean (Gibbs, Strom, & Spivey-Knowlton 1997), people's use
of conceptual metaphors in organizing information in text processing (Allbritton,
McKoon, & Gerrig 1995), and people's use of conceptual metaphors in drawing
inferences when reading poetic metaphors (Gibbs & Nascimento 1996). These
psycholinguistic findings support the hypothesis that different kinds of metaphoric
thought partly explain why many metaphors and idioms have the meanings they
do for contemporary speakers.
146 RAYMOND W. GIBBS, JR.
The claim that pre-existing conceptual metaphors influence all aspects of

how people understand idiomatic and conventional language has been criticized
by scholars interested in immediate metaphor and idiom comprehension
(Glucksberg & Keysar 1990; Glucksberg, Brown, & McGlone 1993; Kreuz &
Graesser 1991 ; McGlone 1996). But even more recent research on on-line idiom
processing shows that people appear to quickly activate underlying conceptual
metaphors when reading idiomatic phrases arising from these same metaphors
(Gibbs, Bogdonovich, Sykes, & Barr 1997).
In general, there is quite a bit of empirical evidence from both cognitive
linguistics and cognitive psychology consistent with the idea that many of our
cognitive models (e.g., ICMs) for abstract concepts are inherently structured via
metaphor (as well as metonymy and related cognitive material). My aim in this
chapter is to explore exactly what it means to say that metaphor is conceptual.
Most cognitive scientists supportive of the conceptual view of metaphor tacitly,
and sometimes explicitly, assume that conventional metaphorical mappings must
be internally represented in the individual minds of language users. I want to
examine this assumption and suggest that cognitive linguists and cognitive
psychologists, like myself, should think about metaphor and its relation to thought
as cognitive webs that extend beyond individual minds and are spread out into the
cultural world. There are two parts to this message. First, our understanding of
what is conceptual about metaphor involves significant aspects of cultural
experience, some of which is even intimately related to our embodied behavior.
Under this view, there need not be a rigid distinction between cultural and
conceptual metaphor. Second, public, cultural representations of conceptual
metaphors have an indispensable cognitive function that allows people to carry
less of a mental burden during everyday thought and language use. This possibility
suggests that important parts of metaphoric thought and language are as much part
of the cultural world as they are internalized mental entities in our heads.
1. Metaphoric representations and embodied experience

What is the best way to think about metaphorical concepts? For the most
part, cognitive linguists and myself have tended to assume that conventional
metaphorical mappings pre-exist as independent entities in long-term memory. For
instance, the systematic pattern of conventional expressions for the abstract
concept of love and love relationship (e.g., We got off to a great start, My
marriage is on the rocks, We are at a turning point, We are making great strides
in our relationship) reflect an underlying conceptual metaphor LOVE IS A JOURNEY
that exists as one of several different ways that the concept of love is represented
in long-term memory.
METAPHOR AND CULTURE 147
There are at least two possible ways that metaphor might structure
conceptual representations (Murphy 1996). The strong view proposes that many
concepts are not understood via their own representations but by metaphorical
connections to knowledge in different domains. For instance, people do not have
an independent, non-metaphorical concept for love, but have a concept for love
that is closely connected via metaphorical links to other, truly independent,
concepts such as that of journeys. The weak view proposes that people have
well-developed, independent concepts, but these are often metaphorically linked
to other concepts with similar structure. Thus, people have a distinct, non-meta
phorical concept for love, but this concept has well-established connections to
distinct concepts from different domains of experience, like journeys, which are
structured similarly in that both source and target domain concepts share similar
underlying attributes or relations.
The empirical evidence from cognitive psychology is presently unable to
distinguish between the strong and weak views of metaphoric representation (see
Murphy 1996 for criticism of the strong view). I think it fair to say that cognitive
linguistic research also does not provide a firm basis for distinguishing between
these two possibilities, although some cognitive linguists tend to argue for a strong
view (e.g., Johnson 1987; Kvecses 1991). Both the strong and weak views
presuppose, nonetheless, that metaphors are conceptual in the sense that metaphor
plays a major role in people's mental representations of many, particularly
abstract, concepts.
One question that linguists and psychologists have debated concerns why
certain conceptual metaphors, but not others, are used by people in speaking about
abstract concepts. There is a large body of work in cognitive linguistics suggesting
that much metaphorical thinking arises from recurring patterns of embodied
experience (Johnson 1987; Lakoff 1987, 1990). For example, central to our
understanding of the conceptual metaphor ANGER IS HEATED FLUID IN A CON
TAINER, which gives rise to expressions such as blew your stack, hit the ceiling,
make my blood boil, is the embodied experience of containment. People have
strong kinesthetic experiences of bodily containment ranging from situations in
which bodies are in and out of containers (e.g., bathtubs, beds, rooms, houses) to
experiences of bodies as containers in which substances enter and exit. An
important part of bodily containment is the experience of our bodies being filled
with liquids including stomach fluids, blood, and sweat. Under stress, people
experience the feeling of their bodily fluids becoming heated. These various,
recurring bodily experiences give rise to the development of an experiential
gestalt, called an image schema, for CONTAINMENT (Johnson 1987).
Image schemas cover a wide range of experiential structures that are

pervasive in experience, have internal structure, and can be metaphorically
elaborated to provide for our understanding of more abstract domains (Gibbs &
Colston 1995; Johnson 1987; Lakoff 1987). Our CONTAINMENT schema, for
instance, is metaphorically elaborated to explain some of the complex ways that
we structure single abstract concepts. For instance, the conceptual metaphor
ANGER IS HEATED FLUID IN A CONTAINER takes the image schema for containment
as part of its source domain and maps this image-schematic structure onto anger,
which gives rise to a number of interesting entailments. Thus, people know that
when the intensity of anger increases, the fluid in the container rises (e.g., His
pent-up anger welled up inside of him), people know that intense heat produces
steam and creates pressure on the container (e.g., Bill is getting hot under the
collar, Jim 's just blowing off steam, and He was bursting with anger), and people
know that when the pressure of the container becomes too high, the container
explodes (e.g., She blew up at me).
There are several studies from cognitive psychology that support the idea
that people's embodied experiences gives rise to their metaphorical structuring of
abstract concepts, which, in turn, constrains speakers' use and understanding of
language. Let me briefly describe two such research projects. I will then suggest
that some of my conclusions from this research paint too narrow a view of
metaphor in ordinary cognition. Following this, I outline an alternative view about
metaphor in thought, one that does not rely exclusively on the assumption that for
metaphors to be "conceptual," they must only be part of people's internal mental
representations.
One research project in psycholinguistics on the embodied nature of
language focuses on how people use and understand idiomatic phrases (Gibbs
1992). The traditional view of idiomaticity assumes that expression such as blow
your stack, flip your lid and hit the ceiling are "giant lexical items" whose
meanings result from "dead" metaphors. But cognitive linguistic research (Lakoff
1987) and psycholinguistics studies (Gibbs 1994) demonstrate that idioms are not
simple, "dead" metaphors, but actually retain a good deal of their metaphoricity
because they arise from metaphorical mappings between dissimilar source and
target domains. For example, the figurative meanings of blow your stack, flip your
lid, and hit the ceiling are specifically linked to two independently existing
elements in our conceptual system - MIND IS A CONTAINER and ANGER IS HEAT IN
A PRESSURIZED CONTAINER.
One set of psycholinguistic studies examined how people's intuitions of the
bodily experience of containment, and several other image schemas, which serve
as the source domains for several important conceptual metaphors, underlie
speakers' use and understanding of idioms. These studies were designed to show
that the specific entailments of idioms reflect the source to target domain
mappings of their underlying conceptual metaphors (Gibbs 1992). Most
importantly, these metaphorical mappings preserve the cognitive topology of these
embodied, image-schematic source domains.
Participants in a first study were questioned about their understanding of
events corresponding to particular bodily experiences that were viewed as
motivating specific source domains in conceptual metaphors (e.g., the experience
of one's body as a container filled with fluid). For instance, participants were
asked to imagine the embodied experience of a sealed container filled with fluid,
and then they were asked something about causation (e.g., What would cause the
container to explode?), intentionality (e.g., Does the container explode on purpose
or does it explode through no volition of its own?), and manner (e.g., Does the
explosion of the container occur in a gentle or a violent manner?).
Overall, the participants were remarkably consistent in their responses to
the various questions. To give one example, people responded that the cause of a
sealed container exploding its contents out is the internal pressure caused by the
increase in the heat of the fluid inside the container. They also reported that this
explosion is unintentional because containers and fluid have no intentional
agency, and that the explosion occurs in a violent manner. These brief responses
provide a rough, nonlinguistic profile of people's understanding of a particular
source domain concept (i.e., heated fluid in the bodily container). These profiles
are rough approximations of what cognitive linguistics and others refer to as the
image-schematic structures of the source domains (Gibbs & Colston 1995; Lakoff
1990; Turner 1991, 1996).
These different image schematic profiles about certain abstract concepts
allowed me to predict something about people's understanding of idioms. My idea
was that people's intuitions about various source domains map onto their
conceptualizations of different target domains in very predictable ways. For
instance, people's understanding of anger should partly be structured by their folk
concept for heated fluid in the bodily container as described above. Several studies
showed this to be true (Gibbs 1992). Not surprisingly, when people understand
anger idioms, such as blow your stack, flip your lid, or hit the ceiling, they inferred
that the cause of anger is internal pressure, that the expression of anger is
unintentional, and is done in an abrupt violent manner. People do not draw these
same inferences about causation, intentionality, and manner when comprehending
literal paraphrases of idioms, such as get very angry.
More interesting, though, is that people's intuitions about various source
domains map onto their conceptualizations of different target domains in very
predictable ways. For instance, several later experiments showed that people find
idioms to be more appropriate and easier to understand when they are seen in
discourse contexts that are consistent with the various entailments of these
phrases. Thus, people find it easy to process the idiomatic phrase blow your stack
when this was read in a context that accurately described the cause of the person's
anger as being due to internal pressure, where the expression of anger was
unintentional and violent (all entailments that are consistent with the entailments
of the source to target domain mappings of heated fluid in a container onto anger).
But readers took significantly longer to read blow your stack when any of these
entailments were contradicted in the preceding story context.
These psycholinguistic findings provide additional evidence that people's
metaphorical concepts underlie their understanding of what idioms mean in
written texts. Moreover, they provide significant experimental evidence that
people's intuitions about their embodied experiences can predict something about
their use and understanding of idioms, expressions that are partly motivated by
bodily based conceptual metaphors.
A different series of experiments demonstrates that people appear to
compute or access metaphorical representations during their immediate under
standing of idioms like blew his stack (Gibbs, Bogdonovich, Sykes, & Barr 1997).
In these studies, participants read stories ending with idioms and then quickly gave
lexical decision responses to visually presented letter-strings that reflected either
something about the conceptual metaphors underlying these idioms (e.g., "heat"
for ANGER IS HEATED FLUID IN A CONTAINER having just read John blew his stack)
or letter-strings that were unrelated to these conceptual metaphors (e.g., "lead").
There were two important findings from this study. First, people were faster
to make these lexical decision responses to the related metaphor targets (i.e.,
"heat") having just read idioms than they were to either literal paraphrases of
idioms (e.g., "John got very angry") or control phrases (e.g., phrases still
appropriate to the context such as "John saw many dents"). Second, people were
faster in recognizing related metaphorical targets than unrelated ones having read
idioms, but not literal paraphrases or control phrases. This pattern of results
suggests that people are immediately computing or accessing at least something
related to the conceptual metaphor ANGER IS HEATED FLUID IN A CONTAINER when
they read idioms. In another experiment, participants were faster to make lexical
decision responses to metaphor targets (e.g., "heat") having read an idiom
motivated by a similar conceptual metaphor (e.g., John blew his stack) than an
idiom with roughly the same figurative meaning but motivated by a different
conceptual metaphor (e.g., John bit her head off, which is motivated by the
conceptual metaphor ANGER IS ANIMAL BEHAVIOR). Again, it appears that people
compute or access the relevant conceptual metaphor for an idiom during some
aspect of their processing of these phrases.
It is important to be careful in interpreting the results of psycholinguistic

studies like those just described. Thus, the Gibbs et al. (1997) data should only be
understood as showing that people quickly see a tight association between their
understanding of certain idioms and particular conceptual metaphors. These
results do not necessarily imply that people actually compute or access conceptual
metaphors when they are actively processing the meanings of idioms in real-time.
Nor do these data tell us whether people must compute or access an idiom's
underlying conceptual metaphor in order to interpret what that idiom figuratively
means. For instance, the fact that reading John blew his stack primes people's
understanding of the concept heat from ANGER IS HEATED FLUIDINA CONTAINER
does not imply that reader must access the conceptual metaphor to comprehend
the figurative meaning of the idiom.
Furthermore, the Gibbs et al. (1997) findings do not tell us whether people
actively construct metaphoric representations (i.e., the conceptual metaphor) when
understanding idioms or whether people merely access in an associative manner
pre-existing conceptual metaphors when processing certain idioms. When people
read an idiomatic expression like John blew his stack they may very well quickly
access the conceptual metaphor ANGER IS HEATED FLUID IN A CONTAINER given
that this metaphor is so closely tied to the idiom, even if the metaphor is not
needed to actually understand what the idiom means in discourse. People may not
actually compute from scratch a source (HEATED FLUID FN A CONTAINER) to target
(ANGER) domain mapping, and draw the complex set of inferences associated with
the conceptual metaphor, during ordinary understanding of conventional language.
Finally, the fact that conceptual metaphors may be active during some part
of idiom understanding does not mean that people are activating embodied image
schemas. Thus, in understanding John blew his stack, readers do not necessarily
activate or re-experience specific image schemas such as CONTAINMENT as part
of their interpreting the figurative meaning of the idiom.
I offer these comments about the Gibbs et al. (1997) findings to urge that
cognitive linguists and others do not overinterpret the results of experimental data
from psycholinguistics. There is clearly a need for more detailed studies to better
understand some of the intricacies in the relationship between conceptual
metaphors and linguistic expressions. Part of this work, as I now argue, demands
consideration of the cultural contexts in which conceptual metaphors arise and
support particular uses of language.
2. What's missing from the psycholinguistic evidence

My thinking about the role of conceptual metaphor in people's use and
understanding of language has mostly embraced an individualistic view of
cognition. Much of my empirical research (e.g., Gibbs 1992; Gibbs et al. 1997) on
the embodied nature of metaphorical thought and language use generally assumes
that people create embodied, metaphorical representations from their phenome-
nological experiences of the body and their sensori-motor interactions with the
physical world. People's metaphorical understanding of certain abstract concepts
are intimately tied to image schemas that partly arise from recurring bodily
experiences. People clearly also learn about conceptual metaphors from their
experience with language. Yet I interpreted my data as supporting the idea that
people have independent mental representations for abstract concepts, such as
anger, that are structured by metaphorical mappings constrained by the image
schematic structure of the source domain (i.e., heated fluid in a container). These
complex metaphoric representations are encoded as part of people's internal,
long-term memory systems.
I still believe that this characterization of metaphorical concepts is partly
accurate as a psychological theory of human conceptual systems. But metaphor
can still be cognitive yet not simply encoded as an internal mental representation.
Many scholars argue that theories of cognition should stop maintaining the idea
that cognitive structures are necessarily "in the head," and instead should
acknowledge that they are dynamic systems of "structural couplings" which model
how people interact with the world, including different linguistic environments.
The traditional view of cognition assumes that representations are
exclusively in the mind (e.g., propositions, schemas, productions, mental images,
connectionist networks). External representations (e.g., real-world objects,
situations, codified aspects of language) are seen to, at best, have only a peripheral
role in cognitive behavior. Yet a variety of cognitive scientists now question the
need for elaborate mental machinery in explaining the genesis of human linguistic
and nonlinguistic behavior, especially the well-known belief that there are internal
mental representations with computational processes that operate upon them (see
Clark 1996; Hutchins 1995; Shannon 1993; Thelen & Smith 1994).
Several branches of psychology have adopted this perspective on cognition.
The socio/cultural approach maintains that it is the continuous internalization of
the information and structure from the environment and the externalization of
internal representations into the environment (e.g., "off-loading") that produce
high-level psychological functions (Leontiev 1981; Luria 1976; Vygotsky 1978,
1986). The situational cognition approach argues that the activities of individuals
are situated in the social and physical contexts around them and knowledge can
be considered as a relation between the individual and the situation (Greeno 1989;
Suchman 1987). Many of these scholars argue that theories of cognition should
not continue to maintain that cognitive structures are necessarily "in the head," but
should recognize how "wide" or "distributed" cognition is, in the sense of being
spread out into the world (Hutchins 1995; Wilson 1994).
Under this "wide" or "distributed" view, cognition is what happens when the
body meets the world. One cannot talk about, or study, cognition apart from our
specific embodied interactions with the cultural world (and this includes the
physical world which is not separate from the cultural one in the important sense
that what we see as meaningful in the physical world is highly constrained by our
cultural beliefs and values). Scholars cannot, and should not assume, that mind,
body, and culture can somehow be independently portioned out of human
behavior as it is only appropriate to study particular "interactions" between
thought, language, and culture, respectively. Theories of human conceptual
systems should be inherently cultural in that the cognition which occurs when the
body meets the world is inextricably culturally-based. What is missing from the
psycholinguistic work, and from aspects of the work on metaphor in cognitive
linguistics, is an explicit acknowledgment of culture and its important, perhaps
defining, role in shaping embodiment and, consequently, metaphorical thought.
In particular, certain cultural representations of metaphor enable people to
"off-load" some aspects of conceptual metaphor out into the cultural world such
that people need not rely exclusively on internal mental representations when
solving problems, making decisions, using language, and so forth.
3. The cultural basis for embodied metaphor

Over the past ten years, several linguists and anthropologists have
persuasively argued that the systematic presence of metaphor in linguistic
expressions reflects not just the operation of individual mental structures, but the
workings of different cultural models (cf. Csordas 1994; Emanation 1995, this
volume; Fernandez 1993; Holland & Quinn 1987; Kvecses this volume).
Cultural models are intersubjectively shared cultural schemas that function to
interpret experience and guide action in a wide variety of domains including
events, institutions, and physical and mental objects. To take an early, and still
influential model, Lakoff and Kvecses' (1987) cultural model for "anger" shows
how metaphor structures not only how individuals within a language community
might conceptualize of their various anger experiences, but also how metaphor
shapes the cultural knowledge that individuals share, to some degree or another,
in a particular cultural community. Kvecses (in press) has gone on to show how
metaphor motivates many different cultural models. For instance, according to
Kvecses, an emotion concept like "love" is best viewed as being constituted by
a large number of cognitive models centered around a small number (or just one)
prototypical model(s). The conceptual contents of various cognitive models,
especially the prototypical ones, are structured by metaphors, metonymies, and
"related concepts."
Both cognitive and cultural models are to some extent similar in that each
is assumed to provide the substrate for various linguistic and nonlinguistic
behaviors. That is, cultural models are not epiphenomenal, but are presumed to do
real work for individuals and collective communities in shaping what people
believe, how they act, and how they speak about the world and their own
experiences. What is the relationship between metaphor's role in structuring
cognitive models and in shaping cultural models? Should culture be located
externally to the individual, as the products of prior human activity, or should it
be located internally as part of knowledge and beliefs? Both views have support
in anthropology (D'Andrade 1995). A cognitive model for some complex
concept, like for anger, might be more limited than the cultural model, which
perhaps reflects a widespread model of cognition distributed across members of
a speech community. For example, the emphasis within cognitive linguistics on
conceptual systems underlying the speech of idealized native speakers may better
be viewed as capturing something about the supra-individual, or social/cultural,
basis for metaphor rather than anything about the psychology of individual
speakers (Steen 1994). Psychologists and linguists should be very careful not to
assume that (a) cognitive models must be explicitly represented in people's heads,
and (b) that conceptual metaphors, as a significant part of these cognitive models,
are only represented as internal mental structures.
One implication of the "wide" or "distributed" view of cognition is that even
image schemas, which arise from recurring embodied experiences, and which
often serve as the source domains for conceptual metaphors, might very well have
a strong cultural component to them, especially in terms of which aspects of
embodied experience are viewed as particularly salient and meaningful in people's
lives. Anthropologists have in recent years spent considerable effort looking at the
role of embodiment in culture, and have in several cases shown how embodied
experience itself is culturally constituted (Csordas 1994; Strathern 1996). As
Quinn (1991) pointed out, many of our embodied experiences are rooted in
social-cultural contexts. For instance, the notion of CONTAINMENT (identified as
an image schema in the cognitive linguistics literature) is based on one's own
bodily experience of things going in and out of the body, and of our bodygoing in
and out of containers. But containment is not just a sensori-motor act, but an event
full of anticipation, sometimes surprise, sometimes fear, sometimes joy, each of
which are shaped by the presence of other objects and people that we interact with.
Image schemas are not therefore simply given by the body but constructed out of
culturally governed interactions. One dramatic demonstration of culture's role in
shaping elementary body sensations and experience is seen in the work on the
permeable boundaries of the Hindu body-self (Bharati 1985).
The bodily experiences that form the source domains for conceptual
metaphors are themselves complex social and cultural constructions (Kirmayer
1992). Anthropologists have demonstrated in a variety of cultural settings how
many elementary embodied experiences are shaped by local cultural knowledge
and practice (cf. Csordas 1994). People instill cultural meaning to bodily
processes such as breathing, blushing, menstruation, birth, sex, crying and
laughing, and value the products of the body (e.g., blood, semen, sweat, tears,
feces, urine, and saliva) differently in changing cultural contexts. More complex
bodily experiences ranging from nerves (Low 1994), to rape (Winkler 1994), to
pain and torture (Jackson 1994; Scheper-Hughs & Lock 1987), to name just a few,
have been studied by anthropologists to explore the linkages between embodi
ment, metaphor, and cultural meaning. These studies demand that cognitive
linguists and others acknowledge that embodied metaphor arises not from within
the body alone, and is then represented in the minds of individuals, but emerges
from bodily interactions that are to a large extent defined by the cultural world.
Taking this idea seriously raises the possibility that the results of Gibbs
(1994) and Gibbs et al. (1997) may not simply reflect the guiding presence of
conceptual metaphors in experimental participants' heads as much as the fact that
people quickly see connections between idioms and certain culturally-shared
metaphors of thought. Although I interpreted my findings as supporting the idea
that embodied metaphor plays a major role in both thought and language, this
work did not explicitly acknowledge that social and cultural constructions of
experience fundamentally shape embodied metaphor. For instance, metaphorical
thought about anger might indeed be structured in terms of embodied experiences
of autonomic arousal and cardiovascular response (i.e., anger is heated fluid under
pressure within the bodily container). But our embodied experience of anger is
also understood in terms of social situations in which there is an offending event
and offender and some attempt, usually, to seek retribution (Lazarus 1991). Our
social/cultural interpretation of an event (i.e., its appraisal) shapes how our bodies
experience emotions. Thus, wide-spread conceptual metaphors, such as ANGER IS
HEATED FLUID IN A CONTAINER, are shaped by specific social/cultural knowledge
as well as, if not more so than, bodily experiences per se.
To make this idea more concrete, imagine a very simple scenario of a
passerby kicking you in the leg as you sit on a park bench. You feel great physical
pain as a result of this injury. But in what cases will you experience anger and how
might you conceptualize this emotion? My claim is that you do not feel angry
simply because of the physical event of someone else kicking you in the leg.
Moreover, you do not metaphorically conceptualize of the anger experienced here
simply because your body feels heated and under pressure. Instead, you feel angry
precisely because you evaluate or appraise the action of someone kicking you in
the leg as a particular kind of social behavior, such as that the person intentionally
wishes to do you harm and attempts to achieve this goal by purposely kicking you
in the leg. When you appraise a person doing something intentionally to harm you,
you may, depending on the situation and your past history, conceptualize this
angry feeling in metaphorical terms. For instance, you may believe that the other
person's intentional actions put you under some pressure that you feel quite
viscerally with your body in such as way as to think of the event as ANGER IS
HEATED FLUID IN AN (EMBODIED) CONTAINER. In other situations, however, you
may feel physical pain but not experience any anger because you appraise the
person's act as completely unintentional, perhaps due to your carelessly sticking
your leg out on the walkway. What defines each of these situations, and your
embodied reactions to them, is your cultural interpretation of the event and the
motivations, or reasons, for the other person's behavior towards you (never mind
the important cultural influences that subtly shape your conceptualization of the
very idea of your body as a container). Under this view, it makes little sense to
think of embodiment, and the metaphors that arise from it,
apart from the very specific cultural evaluations we unconsciously make for our
actions, the actions of others, and the world around us.
Recognizing that what is cognitive (and embodied) is inherently cultural
should be a fundamental part of how we do our work as cognitive psychologists,
linguists, and anthropologists. There might be far fewer differences between
cognitive and cultural models than often suggested by cognitive linguistics and
anthropologists. Some of the debates over whether metaphor is constitutive of
cultural models (e.g., Kvecses and others- see Gibbs 1994) or whether metaphor
only reflects non-metaphorical cultural models (e.g., Quinn 1992) tend to
disappear if we embrace the possibility that all cognition is embodied in cultural
situations.
4. Metaphor as external representations

The inseparability of mind, body, and world, and of cognitive and cultural
models, points to the important idea that metaphor is an emergent property of
body-world interactions, rather than arising purely from the heads of individual
people. We need not talk of metaphor as only part of our mental representations
for concepts (e.g., anger), or as expressed by language (e.g., She bursted with
anger). Metaphor is a kind of tool that arises from body-world interactions which
we can "re-experience" in an embodied way, and is not simply accessed from
long-term memory, in different ways in different real-world situations. As
Kirmayer (1992:335) phrased it: "metaphors are tools for working with experi
ence."
When I talk of moving metaphor "out into the world" I am thinking of

metaphor as a kind of tool, available as a "public representation" for all to use
when needed, without having to explicitly encode all conceptual metaphors as part
of our internal mental representations. To give an idea of what it means to
"off-load" metaphor into the world as a tool, first consider the following. Most of
us have the ability to do long-division arithmetic, but we can best do this when we
employ paper and pencil so as to reduce memory limitations and so forth in doing
the calculations. Paper and pencils exists as tools "out in the world" for us to use
to be smarter and have less of a burden to maintain complex machinery in our
heads. But doing arithmetic with pencil and paper is not necessarily the same as
doing it in one's head. The medium one uses affects performance. For example,
Asian children are taught arithmetic using an abacus takes considerable skill and
practice. Yet studies of middle to late-grade school children shown that Asian
children perform mental arithmetic as if they were calculating on an abacus,
making the same kinds of mistakes when actually using an abacus (e.g., losing
track of 5s). Similarly, American children do mental arithmetic as if they were
calculating using pencil and paper, making the same kinds of mistakes one makes
in such calculations (e.g., mistakes in borrowing) (Stigler 1984). These examples
demonstrate the significant interaction between internal and external representa
tions in cognitive behavior. They also illustrate the difficulty in assuming that
internal representations are not influenced by real-world, cultural representations.
Thus, one might incorrectly conclude that a child's mistakes in doing mental
arithmetic are solely due to errors in internal cognitive processes and neglect the
important influence that interactions with public representations (e.g., pencil and
paper, abacuses, etc) have on the very formation of internal mental representa
tions. This problem is similar to assuming, for instance, that the development of
conceptual metaphors arise only from internal representations and not from
people's embodied, cultural interactions with the real-world.
Many metaphors are "off-loaded" into the cultural world to enable people
to better solve problems, make decisions, and perform skilled action in the exact
same way that having paper and pencil and abacuses allows us to do complex
arithmetic. Anthropologists often describe metaphorical symbols or events that
guide people's culturally appropriate behavior. For instance, Ohnuki-Tierney
(1992) explores the role that the monkey has as a "playful trope" in Japanese
culture, while Colby (1992) demonstrates the importance of the Japanese tea
ceremony in providing metaphorical coherence to some aspects of Japanese life.
In both cases, the external symbols, which illustrate the interplay of metaphor and
metonymy, serve as a kind of tool or scaffolding for culturally appropriate
behavior, and greatly reduce the need for individuals to structure experience
purely from their own internal metaphorical abilities.
An important way that people are able to "off-load" their conceptual

metaphors into the cultural world is through their continued interactions with
source domains that help motivate culturally appropriate metaphors. For instance,
the metaphor ANGER IS HEATED FLUID IN A CONTAINER is culturally instantiated as
part of our interactions with all sorts of containers in which fluid is heated or put
under pressure. Various cultural representations of this in art (e.g., paintings,
cartoon characters with steam blowing out of their ears and the top of their heads),
in the behavior of real-life pots and kettles all correlate with aspects of embodied
experience to make the metaphorical concept ANGER IS HEATED FLUID IN A
CONTAINER an appropriate way of thinking and behaving when angry. In other
situations, our interactions with real and symbolic animals helps instantiate the
wider cognitive web that makes up the conceptual metaphor ANGER IS ANIMAL
BEHAVIOR (e.g., jump down his throat, bite his head off). Thus, our continual
interaction with real-world source domains reduces the need to represent different
conceptual metaphors in a purely internalized manner.
It is interesting to note that examination of therapeutic discourse reveals just
how often people talk about their emotional experiences in metaphorical terms
based on their interactions with real-world objects that take on symbolic character.
The psychologist Milton Erickson describes many such cases (Haley 1973). For
example, Erickson reported one case involving a couple in therapy because of
conflicts over their sexual relations. Erickson chose one aspect of their life that
was analogous to sexual relations and he asked the couple to talk about this
experience to see how it might relate to their problem. Erickson specifically talked
to the couple about having dinner together to draw them out about their individual
preferences. Thus, the wife liked appetizers before dinner, while the husband
preferred to get right into the main course of meat and potatoes. The wife liked to
have a leisurely dinner, while the husband preferred eating quickly. By having the
couple work to have a dinner that was enjoyable to both of them, Erickson
encouraged the couple to approach their sexual relations in a more satisfactory
manner. The point here is that the metaphor HAVING SEXUAL RELATIONS IS
SHARING A MEAL is not just something that exists as a private, internalized
metaphor in the minds of the husband and wife. Instead, the metaphor has a
public, cultural face that enables the couple to experience, and re-experience, the
metaphor as embodied action out in the world and not just in their heads.
Erickson's analysis, and therapy, illustrates why, for instance, the conceptual
metaphor ANGER IS HEATED FLUID IN A CONTAINER partly emerges from people's
interaction with well-known cultural repertoires. Various experimental work, in
fact, shows that an emotion such as anger is prototypically understood to be
composed of a sequence of events beginning with a set of "antecedent conditions",
a set of "behavioral responses", and a set of "self-control procedures" (Shaver,
Swartz, Kirson, & O'Connor 1987). The temporal organization of many emotions
like anger is reflected in the language used to express those emotions (Lakoff
1987). Thus, English has different idioms fitting each of the three stages for our
concept of anger. The antecedent conditions (Stage 1) for anger concern a sudden
loss of power, status, or respect, ideas exemplified by idioms such as eat humble
pie, kick in the teeth, and swallow one's pride. The behavioral conditions (Stage
2) for anger concern people's behavioral responses to the emotion, an idea best
reflected in idioms such as getting red in the face, getting hot under the collar, or
blowing your stack. The self-control procedures (Stage 3) concern an individual's
efforts to maintain composure, an idea that is best reflected in idiomatic phrases
such as keep your cool, or hold your temper.
People's different metaphorical understandings of anger, including ANGER
IS HEATED FLUID IN A CONTAINER, is as much a part of their participating in
culturally appropriate repertoires as it is caused by full-blown metaphoric
representations in their heads. This suggests the critical point that for a representa
tion to be "conceptual" it need not only be part of our private mental apparatus.
Thus, by adopting a specific cultural identity in being angry at a particular time,
an individual takes advantage of the cultural narrative of being angry with its
conventional ways of acting in a series of steps (i.e., the anger script). This series
of culturally-defined activities (e.g., antecedent conditions, behavioral conditions,
and self-control procedures) provide a structure that is metaphorically analogous
to what happens when people think about or experience fluid in containers that are
heated or put under pressure. It is difficult, then, to separate the conceptual
metaphor from the cultural contexts from which it arises. People can rely on
interpersonal scripts for behaving when angry, or even when they wish to act as
if they are angry, rather than simply generating anger behaviors based on an
internal, metaphorical representation in their heads. Many emotion theorists now
embrace these social/cultural, "out of one's head," ways of describing emotion
concepts and behavior as an alternative to viewing emotion concepts as purely
private, mental phenomena (Parkinson 1994).
Consider another example of the public side of metaphor. Imagine that you
encounter a situation where you must estimate the time it will take to complete a
particular task. Cognitive linguistic analyses have shown that there are several
metaphorical ways of structuring the vague, abstract concept of time, most notably
by thinking of time in terms of space (Alverson 1994; Lakoff & Johnson 1980).
There are many cultural tools for thinking about time in terms of space. We have
all sorts of concrete conventional ways, for instance, of mapping past, present,
and future, before and after, sooner and later onto left and right, up and down,
clockwise and counterclockwise. Monday is to the left of Tuesday for most of us,
and four o-clock is tucked under three o'clock on the right hand side of every day
and night. Our spatialization of time extends to science where familiar systems of
diagrams have been developed to display different kinds of information. We think
of profits, or the temperature, or the loudness of a stereo, rising up from the left
to the right with the passage of time. We use our sense of space to see the passage
of time (usually from left to right in standard convention, except in evolutionary
diagrams, in which earlier eras are often shown at the bottom, with today at the
top). Our ability to imagine these diagrams is parasitic on our ability to draw and
see them, off-loading them at least temporarily into the external, cultural world.
Within the modern world, we constantly bump up against other public
representations of metaphor, such as in pictorial advertisements, artworks, media
images, and a wide range of tools and products (see Forceville 1996; Kennedy
1982; Kennedy et al. 1993). For example, metaphor is used by designers of
computer software, interfaces, and display formats to help attune people to
aspects of the environment that support skilled action. Research shows, for
instance, that visual displays in aircraft often employ metaphoric signs. Thus,
seeing a flight path as a highway not only direct flights accurately in forward and
lateral directions, it also indicates optimal altitude by suggesting that the pilot not
fly into the solid surface depicted on the highway (Dent-Reed, Klein, and
Eggleston 1994). The ranges of electromagnetic radar are viewed as surfaces. In
another display, safe areas for flight close to the terrain are depicted as ponds of
varying depths. In both the "radar are solid surfaces" metaphor and the "flights
paths are highway" metaphor, the affordances (Gibson 1979) of the object used
metaphorically in the depiction (the solid surface or highway) are the same as the
affordances in the natural situation. Solid surfaces do not afford flying through and
radar is to be avoided; highways afford smooth and accurate forward motion in a
vehicle and the flight path is the best path forward in the aircraft. Such invariance
of affordances across types of vehicles seem critical for a successful pictorial
metaphor in displays that guide action.
A wonderful place to find metaphors as public representations is in the
world of computers and the World Wide Web. Consider some of your own
experiences in using computers. Computer uses constantly engage in metaphorical
actions such as surfing, navigating, traveling through webs, manipulating objects
on a desktop, selecting options off of menus, creating different windows,
dragging, dropping, point-and-clicking, and inhabiting on-line communities.
Interface designers have moved beyond the two-dimensional desktop metaphor
and created more embodied digital environments such as town squares, shopping
malls, personal assistants, living rooms that users may interact with other people,
use and develop different program and products, and connect with other computer
services. New programs appear as objects in rooms. When you install a spread-
sheet on your hard-drive, an adorable obsolete mechanical calculator appears on

your shelf space. We intuitively understand these fictional devices as different
metaphors, most of which have explicit iconic representations that we can see and
manipulate. In the fast-changing world of computers, these public metaphoric
representations have an increasingly indispensable cognitive function.
5. Implications for empirical research

My claim that conceptual metaphors are as much social and cultural as
"inside" the heads of individuals has several implications for empirical research
on metaphor in thought and language. Consider the Gibbs (1994) studies showing
that patterns of embodied experience serve as the source domains for conceptual
metaphors underlying people's understanding of idioms. Adopting a "distributed"
view of cognition would force scholars, like myself, to examine not just people's
individual physical conceptions of, for example, the behavior of fluid in the bodily
container. Instead, scholars should look extensively at people's social/cultural
experiences of embodied containment, such as what causes people to move in and
out of containers, how the escape of fluids from different containers is socially
understood, how people's experiences of containment shapes interpersonal
relationships and their own sense of identity/autonomy. Moreover, looking at how
people interact in particular cultural situations and with specific concrete objects
is essential toward seeing how ordinary metaphors are spread out between body
and culture. By systematically assessing people's intuitions of their social/cultural
experiences of containment, one may discern more about what motivates
metaphorical concepts for emotions, and other knowledge domains, which in turn
may provide deeper insights into the context-sensitive meanings of linguistic
expressions. This strategy of seeking links between embodied action and cultural
experience is not typical of research in cognitive psychology, especially given the
bias of cognitive science toward viewing what is cognitive as mental constructs
in the heads of individuals. Nonetheless, certain aspects of what is cognitive about
metaphor, and other things, may be inextricably linked to social/cultural
experience.
6. Conclusion
The vast majority of the discussion about conceptual metaphor in cognitive
linguistics and cognitive psychology has implicitly assumed that for something to
be conceptual, it must be explicitly mentally represented in the minds of
individual people. Even those scholars who have emphasized the importance of
human embodiment in language and thought, especially in motivating metaphor's
important role in thinking, still view conceptual metaphors in terms of what is
inside of people's heads.
My aim in this paper has been to urge cognitive linguists and others to adopt
a distributed perspective on what it means for something to be "conceptual" and
to recognize that cognition arises, and is continually re-experienced, when the
body interacts with the cultural world. In saying this, I do not want to suggest that
culture is somehow just objectively given out in the world, because the body
creates the cultural world as much as culture defines embodied experience. At the
same time, my claim that some aspects of conceptual metaphor are "off-loaded"
does not mean that people never have abstract, metaphorical concepts in their
individual minds. The main point is that our use of metaphors to structure
concepts, such as anger or time, is strongly shaped by (a) how we culturally
conceptualize of situations, like getting angry and sensing time, and (b) by our
interactions with social/cultural artifacts around us. Under this view, metaphor is
as much a species of perceptually guided adaptive action in a particular cultural
situation as it is a specific language device or some internally represented structure
in the mind of individuals. My plea that cognitive scientists move metaphor out
of our heads and into the embodied and public world in this way does not make
metaphor any less cognitive than if we had long lists of metaphors nicely encoded
in our heads. All this move attempts to do is acknowledge the culturally
embodied nature of what is cognitive and to suggest that there is much less of a
difference between what is cognitive and what is cultural than perhaps many of us
have been traditionally led to believe.
One possibility for further exploration is that we can continue to talk about
a human agent as being locked within the envelope of skin and skull, but that
beliefs, knowledge, and many other mental states, including all that is metaphori
cally structured, now depend on physical vehicles that can at times be spread out
to include certain aspects of the local cultural environment. This picture preserves
the idea of a human being as the combination of body and biological brain;
allowing us to sometimes speak of the individual as occasionally manipulating and
structuring external resources (including public representations of metaphor) to
extend and off-load his or her own problem-solving activities. At the same time,
this view acknowledges that in "reaching out" to the world we often create wider
cognitive and computational webs. Thus, the cognitive models we create surely
extend beyond the individual. Understanding these "cognitive webs" will require
the application of tools and concepts of cognitive science to include larger, hybrid
entities comprising brains, bodies, and a variety of cultural systems (see Clark
1996).
References
Allbritton, D., G. McKoon, & R. Gerrig. 1995. "Metaphor-Based Schemas and
Text Comprehension: Making Connections Through Conceptual Meta-
phors". Journal of Experimental Psychology: Learning, Memory, and

Alverson, H. 1994. The Semantics of Experience. Baltimore: Johns Hopkins
University Press.
Bharati, A. 1985. "The Self in Hindu Thought and Action". Culture and Self:
Asian and Western Perspectives, ed. by A. Marsella, G. De Vos, & F. Hsu,
185-230. New York: Tavistock Publications.
Clark, A. 1996. Being There. Cambridge: MIT Press.
Colby, B. 1992. 'The Japanese Tea Ceremony: Coherence Theory and Metaphor
in Social Adaptation". Beyond Metaphor: The Theory of Tropes in
Anthropology, ed. by J. Fernandez, 244-262. Stanford: Stanford University
Press.
Csordas, T. 1994. Embodiment and Experience. New York: Cambridge University
Press.
Dent-Read, C., G. Klein, & R. Eggleston. 1994. "Metaphor in Visual Displays
Designed to Guide Action". Metaphor and Symbolic Activity 9. 211-232.
, & A. Szokolszky. 1993. "Where Do Metaphors Come From?" Metaphor
and Symbolic Activity 8. 227-242.
D'Andrade, R. 1995. Cognitive Anthropology. New York: Oxford University
Press.
Emanation, M. 1995. "Metaphor and The Expression of Emotion: The Value of
Cross-Cultural Perspectives. Metaphor and Symbolic Activity 10. 163-182.
Fernandez, J., ed. 1991. Beyond Metaphor: The Theory of Tropes in Anthropo-
logy. Stanford: Stanford University Press.
Forceville, C. 1996. Pictorial Metaphor in Advertisements. London: Routledge.
Gibbs, R. 1994. The Poetics of Mind: Figurative Thought, Language, and
Understanding. New York: Cambridge University Press.
1996. "Why Many Concepts are Metaphorical". Cognition 61. 309-319.
, & H. Colston. 1995. "The Cognitive Psychological Reality of Image
Schemas and Their Transformations". Cognitive Linguistics 6. 347-378.
, J. Bogdonovich, J. Sykes, & D. Barr. 1997. "Metaphor in Idiom Compre
hension". Journal of Memory and Language 37. 141-154.
, & S. Nascimento. 1996. "How We Talk When We Talk About Love:
Metaphorical Concepts and Understanding Love Poetry". Empirical and
Aesthetic Approaches to Literature, ed. by R. Kreuz, & M. MacNealy,
291-308. Norwood, N.J.: Ablex.
, & N. Nayak. 1991. Why "Idioms Mean What They Do". Journal of
Experimental Psychology: General 120. 93-95.
, & J. O'Brien. 1990. "Idioms and Mental Imagery: The Metaphorical
Motivation for Idiomatic Meaning". Cognition 36. 35-68.
, L. Strom, & M. Spivey-Knowlton. 1997. "Conceptual Metaphors in

Mental Imagery for Proverbs". Journal of Mental Imagery 21. 83-110.
Gibson, J. 1979. The Ecological Approach to Visual Perception. Boston:
Houghton Mifflin.
Glucksberg, S., & B. Keysar. 1990. "Understanding Metaphorical Comparisons:
Beyond Similarity". Psychological Review 97. 3-18.
Glucksberg, S., M. Brown, & M. McGlone. 1993. "Conceptual Metaphors are not
Automatically Accessed During Idiom Comprehension". Memory &
Greeno, J. 1989. "Situations, Mental Models, and Generative Knowledge".
Complex Information Processing, ed. by D. Klahr, & K. Kotovsky,
285-318. Hillsdale, N.J.: Erlbaum.
Haley, J. 1973. Uncommon therapy. New York: Norton.
Holland, D., & N. Quinn, eds. 1987. Cultural Models of Language and Thought.
New York: Cambridge University Press.
Hutchins, E. 1995. Cognition in the Wild. Cambridge: MIT Press.
Jackson, J. 1994. "Chronic Pain and the Tension Between the Body as Subject and
Object. Embodiment and Experience, ed. by T. Csordas, 201-228. New
York: Cambridge University Press.
Johnson, S. 1997. Interface Culture. New York: HarperCollins.
Keysar, B., & B. Bly. 1995. "Intuitions of the Transparency of Idioms: Can You
Keep a Secret by Spilling the Beans?" Journal of Memory and Language
34. 89-109.
Kennedy, J. 1982. "Metaphor in Pictures". Perception 11. 589-605.
, C. Green, & J. Vervaeke. 1993. "Metaphoric Thought and Devices in
Pictures". Metaphor and Symbolic Activity 8. 243-255.
Kirmayer, L. 1992. "The Body's Insistence of Meaning: Metaphor as Presentation
and Representation in Illness Experience". Medical Anthropology Quarterly
6. 323-346.
Kvecses, Z. 1986. Metaphors of Anger, Pride, and Love. Amsterdam/Phi
ladelphia: John Benjamins.
1990. Emotion Concepts. New York: Springer.
(in press). The Force of Emotion: Why Metaphor Matters. New York:
Lakoff, G. 1987. Women, Fire, and Dangerous Things. Chicago: Chicago
University Press.
1990. "The Invariance Hypothesis: Is Abstract Reason Based on Im
age-Schemas? Cognitive Linguistics 1. 39-74.
, & M. Johnson. 1980. Metaphors We Live By. Chicago: Chicago Univer

sity Press.
, & M. Turner. 1989. More Than Cool Reason: A Field Guide to Poetic
Lazarus, R. 1991. Emotion and Adaptation. New York: Oxford University Press.
Leontiev, A. 1981. Problems in the Development of Mind. Moscow: Progress.
Low, S. 1994. "Embodied Metaphors: Nerves as Lived Experience". Embodiment
and Experience, ed. by T. Csordas, 139-162. New York: Cambridge
University Press.
Luria, A. 1976. Cognitive Development: Its Cultural and Social Foundations.
Cambridge: Harvard University Press.
McGlone, M. 1996. "Conceptual Metaphors and Figurative Language Interpreta
tion: Food for Thought?" Journal of Memory and Language 35. 544-565.
Murphy, G. 1996. "On Metaphoric Representation". Cognition 60. 173-204.
Nayak, N., & R. Gibbs. 1990. "Conceptual Knowledge in the Interpretation of
Idioms". Journal of Experimental Psychology: General 119. 315-330.
Ohnuki-Tierney, E. 1992. "Embedding and Transforming Polytrope: The Monkey
as Self in Japanese Culture". Beyond Metaphor: The Theory of Tropes in
Anthropology, ed. by J. Fernandez, 159-189. Stanford: Stanford University
Press.
Parkinson, B. 1994. Ideas and Realities of Emotion. London: Routledge.
Quinn,N. 1991. "The Cultural Basis of Metaphor". Beyond Metaphor: Thetheory
of Tropes in Anthropology, ed. by J. Fernandez, 56-93. Stanford, Calif.:
Stanford University Press.
Shannon, B. 1993. The Representational and Presentational: An Essay on the
Cognition of Mind. London: Harvester Wheatsheaf
Scheper-Hughs, N., & M . Lock. 1987. "The Mindful Body: A Prolegomenon to
Future Work in Medical Anthropology". Medical Anthropology Quarterly
1. 6-41.
Steen, G. 1994. Understanding Metaphor in Literature: An Empirical Approach.
London: Longman.
Stigler, J. 1984. "Mental Abacus: The Effect of Abacus Training on Chinese
Children's Mental Calculation". Cognitive Psychology 16. 145-176.
Strathern, A. 1996. Body Thoughts. Ann Arbor: University of Michigan Press.
Suchman, L. 1987. Plans and Situated Action. New York: Cambridge University
Press.
Sweetser, E. 1990. From Etymology to Pragmatics: The Mind-Body Metaphor in
Semantic Structure and Semantic Change. Cambridge: Cambridge Univer
sity Press.
Thelen, E., & L. Smith. 1994. A Dynamical Systems Approach to Development.

Cambridge: MIT Press.
Turner, M. 1991. Reading Minds: The Study of English in the Age of Cognitive
Science. Princeton N.Y.: Princeton University Press.
1996. The Literary Mind. New York: Oxford University Press.
Vygotsky, L. 1978. Mind in Society. Cambridge: Harvard University Press.
1986. Thought and Language. Cambridge: Harvard University Press.
Wilson, R. 1994. "Wide Computationalism". Mind 103. 351-372.
Winkler, C. 1994. "Rape Trauma: Contexts of Meaning". Embodiment and
Experience, ed. by T. Csordas. 248-268. New York: Cambridge University
Press.
METAPHOR
DOES IT CONSTITUTE OR REFLECT CULTURAL MODELS?
ZOLTN KVECSES
There are two notions that have become extremely influential in recent decades
in attempts to describe and characterize the human conceptual system: cultural
model and metaphor. Psychologists, anthropologists, and linguists have made
extensive use of the notion of cultural model under a variety of different names
(see, e.g., Holland and Quinn 1987; Lakoff 1987). The main idea on which the
notion is based is that we have a coherent organization of any kind of human
experience. However, cultural models are not only all-pervasive, they are also
utilized in all kinds of cognitive processes, such as reasoning. Equally important
is the notion of metaphor. As was shown by Lakoff and Johnson (1980),
metaphor is pervasive in language and it structures much of our thought. Given
these obviously highly important notions, it is legitimate to ask: what is the
relationship between metaphor and cultural models? Cultural models exist both
for concrete and abstract concepts. Clearly, the issue of the nature of the
relationship between cultural model and metaphor can only arise in the case of
cultural models for abstract concepts. Our concepts for chairs, balls, water, rock,
forks, dogs, etc. do not require metaphorical understanding (at least in our
everyday conceptual system and for ordinary purposes). Some scholars claim that
cultural models even for abstract concepts exist without prior metaphorical
understanding, that is, we have a primary literal understanding of them (e.g.,
Quinn 1991). Others, however, claim that cultural models for abstract concepts are
inherently metaphorical, that is, they are constituted by metaphor (e.g., Lakoff and
Johnson 1980; Lakoff and Kvecses 1987).
Thus, the more specific issue I will raise in this paper is the following: does
metaphor constitute or merely reflect cultural models? Since metaphors are
ordinarily used in connection with cultural models that structure abstract concepts,
the issue really becomes: do metaphors constitute abstract concepts (as structured
by cultural models) or do they simply reflect them?
Several answers are theoretically possible to this question. One can say (1)
that abstract concepts emerge literally, without any metaphors constituting them;
(2) that abstract concepts emerge literally from basic human (physical-bodily or
cultural) preconceptual experiences, still without any metaphors constituting them;
168 ZOLTN KVECSES
(3) that abstract concepts emerge metaphorically, with the help of concrete
concepts constituting them; (4) that abstract concepts emerge metaphorically, the
metaphors having some additional physical-cultural basis. Possibilities ( 1 ) and (2)
can be regarded as cases of what can be termed the 'literal emergence view,' while
(3) and (4) as cases of what can be termed the 'metaphorical emergence view.'
These possibilities can be represented diagrammatically as follows:
(l) (2)
METAPHORS METAPHORS
ABSTRACT ABSTRACT
CONCEPT CONCEPT
BASIC
EXPERIENCE
Figure 1 : Literal emergence
(3) (4)
ABSTRACT BSTRACT
CONCEPT CONCEPT
~7K
BASIC
EXPERIENCE
CONCRETE CONCRETE
CONCEPT CONCEPT
Figure 2: Metaphorical emergence

METAPHOR: DOES IT CONSTITUTE OR REFLECT CULTURAL MODELS? 169
In the remainder of the paper I will examine these four possibilities and ask
whether they are viable accounts of the relationship between conceptual meta
phors and abstract concepts as structured by cultural models.
1. Literal emergence (1)

According to the view of literal emergence (1), we assume that a new
abstract concept, or cultural model, arises as a new configuration of content and
structure (i.e., concept, or cultural model) independently of more concrete
configurations of content and structure. That is, meaningful elements for abstract
concepts are assembled into a new whole without being derived from well
established and more concrete ones.
Let us take an example of what this process of emergence might involve in
the case of an abstract concept. Consider such specific abstract concepts as
company, society, government, theory, mind, political system, economy, people,
history, human relationships, language, etc. All these specific concepts can be
subsumed under the generic heading of complex systems. The specific concepts
represent cases in which we have a variety of parts that interact in complex ways.
The concept of complex systems (or organizations) is an abstract and superordi-
nate one. In talking about the abstract concepts that it subsumes, we use language
such as the following: we talk about a society's ills, an ailing company or
economy, heads of states, the heart of a culture, building a solid relationship,
laying the foundations of modern science, constructing scientific knowledge,
someone's career being in ruins, setting in motion the legal machinery, the
workings of the mind, sowing the seeds of freedom in a country, a budding
romance, cultivating a relationship, a blossoming economy, and a nation,
language, or civilization that once flowered. These ways of talking about complex
systems, or organizations, suggest that we think of them as having the following
properties:
they can be in an appropriate or inappropriate condition (ills, ailing)

they have a structure (head, heart)
they can be created (build, construct)
they can be lasting (solid, in ruins)
they can function (work(ings), machinery)
they can develop (sow seeds, bud, blossom, flower)
Unlike the linguistic examples given, we take these properties to be literal, rather
than metaphorical. Given each of these properties, we possess a great deal of
additional knowledge about abstract systems in general. For example, given the
170 ZOLTN KVECSES
property "they can develop," we have additional detailed knowledge such as the
following:
(1) You prepare the development of an abstract complex system.

(2) You create and start the abstract complex system.
(3) The system has initial stages of development.
(4) The system goes through a number of stages.
(5) The complex system may develop appropriately or inappropriately.
(6) You take care of the complex system to insure its appropriate development.
(7) The complex system reaches its best stage in the course of its development.
(8) The complex system produces certain benefits.
(9) The system begins to decline and eventually it ceases to exist.
The more detailed knowledge is also taken to be literal, rather than metaphorical.
This conceptualization shows up in language in a variety of ways. The language
appears to be literal again. Thus, we often talk about companies developing, the
appropriate or inappropriate development of a society, a civilization or theory
reaching its peak and then declining, the benefits that a political or economic
system can yield, and so on.
Not all the complex systems mentioned above have all these properties
associated with them, but many such concepts are associated with most of them.
How did this particular conception of abstract complex systems emerge? We
could say that it emerged literally and that these are literal features of the concept.
The big question is if this particular assembly of these assumed literal features has
come about independently of more concrete concept(s). If we say that it has come
about independently of them, then we have a concept that is pretty much arbitrary
in its content and structure; it is "ungrounded" and "unnatural." We have no way
of accounting for why this particular content and structure was assembled for what
we would now consider the concept of COMPLEX SYSTEMS. (I will continue the
analysis of the notion of COMPLEX SYSTEMS in section 3.)
However, this arbitrary but literal assembly of content and structure could
be taken to be the basis for the selection of metaphor(s). Thus, on the view of
Literal Emergence (1), it could be suggested that this (partial) preexisting literal
cultural model serves as a basis that determines the selection of particular
metaphor(s), such as the metaphor COMPLEX SYSTEMS ARE PLANTS. In this view,
the literalness of a concept goes together with conceptual "unnaturalness," or
"arbitrariness" in the creation of the concept.
In general, on the Literal Emergence (1) account of the relationship between
abstract concepts and metaphor very little that is systematic can be said about why
we have the abstract concepts with their particular assembly of content and
structure. But perhaps no one would want to claim that abstract concepts emerge
in a completely arbitrary way. As we saw in the introduction, there seem to be

three ways of accounting for the non-arbitrary emergence of abstract concepts.
These are the options to which I now turn.
2. Literal emergence (2)

The Literal Emergence (2) view shares with Literal Emergence (1) the idea
that metaphors reflect a pre-existing literal cultural model, but adds to it the idea
that the literal cultural models emerge directly (i.e., without metaphor) from some
preconceptual experience. We can conceive of the Literal Emergence (2) view as
an improved version of the Literal Emergence (1) view in which the obvious
weakness of ungroundedness characterizing Literal Emergence (1) is remedied.
I will describe this view in more detail below, when I discuss Quinn's work on
American marriage.
2.1 Marriage
Quinn (1991) suggests that, contrary to a claim made by Lakoff and
Kvecses (1987), metaphor simply reflects cultural models. In contrast, Lakoff
and Kvecses claim that metaphors largely constitute the cultural model, or naive
understanding of anger, as based on their study of American English.
Implicit in Quinn's claim that metaphors simply reflect preexisting cultural
models are two very important further claims: one is that abstract concepts can be
understood in a literal way and the other is that the core of culture consists of
literally understood cultural models (for both concrete and abstract concepts).
The first claim arises from the fact that Quinn's generalization is based on
the examination of such abstract concepts as anger and marriage. Quinn suggests
that concepts such as marriage are understood literally by people. The concept of
marriage is one of several other concepts indicating human relationships.
Furthermore, she seems to think that anger, a prototypical emotion concept, can
also be literally understood. Both concepts of human relationships and emotions
are prime examples of abstract concepts. Indeed, Quinn (1991:64-65) makes a
more general claim about abstract concepts: "while I certainly agree that
metaphors play some role in the way we comprehend and draw inferences about
abstract concepts, I take issue with the claim that they or the schemas on which
they are said to be founded actually constitute the concepts." A little earlier in the
paper she states: "I will be arguing that metaphors, far from constituting
understanding, are ordinarily selected to fit a preexisting and culturally shared
model" (1991:60; my emphasis).
This is a general claim about the nature of the human conceptual system. My
discussion will focus on this particular issue. I will have nothing to say about the
second assumption; namely, that the core of culture consists of literally understood
172 ZOLTN KVECSES
cultural models. As regards this claim, I refer the reader to Bradd Shore's work
(Shore 1996), who claims, contrary to Quinn, that even the most basic notions of
a culture may be metaphorically constituted. Gibbs (1994) provides additional
criticism of Quinn's challenge.
On Quinn's view, the American conception of marriage can be characterized
by a set of expectations: marriage is expected to be shared, mutually beneficial,
and lasting (1991:67). She points out, furthermore:
that this particular constellation of expectations derives from the mapping

of our cultural conception of love onto the institution of marriage and the
consequent structuring of marital expectations in terms of the motivational
structure of love. Because people want to be with the person they love, they
want and expect marriage to be shared; because they want to fulfill the
loved person's needs and have their own needs fulfilled by that person, they
want and expect marriage to be beneficial to both spouses in the sense of
mutually fulfilling; and because they do not want to lose the person they
love, but want that person to go on loving them forever, people want and
expect their marriages to be lasting. (Quinn 1991:67)
In this view, marriage takes over several properties of love, which then come to
define it. The question of course becomes: where does the abstract concept of love
come from? Does it emerge literally or metaphorically? Quinn's answer is
straightforward. It emerges literally from certain basic experiences, and then these
experiences will structure marriage. The particular basic experiences that Quinn
suggests the American conception of love and marriage derives from involve early
infantile experiences between baby and the first caretaker. Here is the relevant
passage:
I speculate that the motivational constellation that is part of our understand

ing of love and that provides marriage with its structure itself makes sense
in psychoanalytic terms. Psychoanalysts since Freud, who characterized
adult love as a "re-finding" of infantile love for the first caretaker, have
theorized about the relation between the two. My claim is that Americans'
distinctive conception of marriage takes the particular shape it does and has
the force it does for us because of the cultural model of love mapped onto
marriage and, thus, indirectly because of an infantile experience that
Americans have shared and that underpins our conception of adult love.
(1991:67)
In other words, the picture that Quinn paints of the emergence of the concept of
marriage is subtler than the one depicted in the diagram Literal Emergence (2) and
can be given as Literal Emergence (2a):
MARRIAGE
LOVE |
INFANTILE
EXPERIENCE
BETWEEN BABY AND
FIRST CARETAKER
Figure 3 : Literal Emergence (2a)
However, the point remains: no metaphor is needed for abstract concepts to

emerge. The expectational structure of marriage derives from the motivational
structure of love, which in turn derives from the basic infantile experience
between baby and first caretaker.
Quinn, then, goes on to say that marriage has some additional aspects. In her
own words again: "The remainder of the cultural model of marriage reflected in
the metaphors for marital compatibility, difficulty, effort, success or failure, and
risk, derives from a contradiction that arises inevitably between the expectations
of mutual benefit and that of lastingness" (1991:67). She argues further that in
voluntary relationships, if one's needs are not fulfilled one is free to leave the
relationship. However, marriage is special in this respect: it is supposed to last.
She adds: "a variety of situations can initiate a felt contradiction between the
expectation of marital fulfillment and that of a lasting marriage."
If we characterize the essence of marriage, as Quinn does, as a set of
expectations that can be viewed as being literal, Quinn's major claim stands: the
core of the concept of marriage is literal, hence metaphors do not play a
constitutive role in its understanding. More generally, abstract concepts such as
marriage can exist without metaphors constituting them. This analysis would
support the Literal Emergence (2) view.
However, I believe that the analysis is incomplete and problematic. The
problem is that we cannot take the expectational structure of marriage to be literal.
Notice that Quinn's claim is that it is the motivational structure of love (i.e., that
we want to be with the person we love, we want mutual need fulfillment, and we
174 ZOLTN KVECSES
want love to be lasting) that provides the expectational structure of marriage.

What Quinn does not say is how the concept of love itself is structured over and
above its motivational structure. We should, therefore, first ask what love is before
we discuss its expectational structure. And, ultimately, the question we have to
face is whether the structure of the concept of love itself is derivable from the
basic infantile experiences that Quinn mentions. Can the concept of love emerge
literally from these basic experiences? My answer is that the basic infantile
experiences play an important role in the emergence of the concept but are not
sufficient for its detailed characterization. The insufficiency comes from the fact
that the infantile experiences lack the detailed content and structure that
characterize the concept of love in adults. (Details will be provided in section 4.)
In other words, the metaphorical source domain has structure and content that is
additional to that found in the basic experience. To take another example of a
similar situation, the bodily-physiological experiences associated with anger
cannot be said to define the content and structure of the concept of anger. For that,
we need conceptual metaphors (such as ANGER IS A HOT FLUID IN A CONTAINER).
With the help of basic experiences alone and without such conceptual metaphors
it is difficult to see how abstract concepts can emerge and, in emerging, how they
can acquire the detailed content and structure that they have. Nevertheless, the
basic experiences do play an important function in this process; namely, they can
be seen as motivating and constraining the conceptual metaphors that can
eventually provide the necessary content and structure of abstract concepts. I will
claim in section 4 that in general abstract concepts can emerge from basic
experiences through the mediation of metaphor only.
3. Metaphorical emergence (3)

It will be recalled from the introduction that in this view abstract concepts
emerge from concrete ones via conceptual metaphor. Let us now consider how this
Metaphorical Emergence view would work for the concept of COMPLEX SYSTEMS
that we briefly looked at in section 1.
The particular metaphor that could be suggested as producing the generic
concept of abstract complex systems is COMPLEX ABSTRACT SYSTEMS ARE
COMPLEX PHYSICAL OBJECTS (including the HUMAN BODY, BUILDINGS, MACHINES,
and PLANTS); in the conception under consideration (i.e., given as the properties
1-9 in section 1), the specific source domain would be PLANTS.
The COMPLEX SYSTEMS ARE PLANTS metaphor is based on a small number
of basic, constituent mappings, including the following:
a) the plant is the complex system

b) parts of the plant are parts of the complex system
c) the biological growth of the plant is the abstract non-biological development

of the complex system
We can illustrate these mappings with such metaphorical sentences as these:
(1) Please turn to the local branch of the organization.

(2) She has grown a lot as a scholar lately.
Sentence (1) demonstrates mappings (a) and (b), whereas sentence (2) is a
linguistic manifestation of mappings (a) and (c). The part of a plant can include
several things, for example a specialization in some discipline, as shown in
sentence (3) below:
(3) Laser equipment is expensive but it can be used in many branches of

surgery.
This sentence comes from Cobuild English Guides 7: Metaphor, which is a

dictionary of English metaphors for learners of English as a foreign language. The
series is based on "the Bank of English," a huge corpus of everyday English.
Indeed, in my characterization of the COMPLEX SYSTEMS ARE PLANTS metaphor
below, I will rely exclusively on this source of information.
As is well known in the cognitive linguistic view of metaphor, in some cases
of metaphor we have a great deal of rich knowledge about the elements in the
source, and consequently, we can make use of this knowledge in the comprehen
sion of the target. Two such pieces of knowledge include the following: when
plants grow, they become physically bigger, and plants are sometimes cut or
primed, which results in a smaller size. Now it seems that speakers of English
make use of this additional information in understanding certain features of
complex systems. We can represent these metaphorical entailments as submeta-
phors Of COMPLEX SYSTEMS ARE PLANTS as follows:
A COMPLEX SYSTEM BECOMING LARGER IS A PLANT GROWING BIGGER

- Only now, 21 years since he established his distinctive women's range, is
he branching out into men's clothing.
REDUCING COMPLEX SYSTEMS IS MAKING PLANTS SMALLER (PRUNING, CUTTING)

- They selectively pruned the workforce.
- Government and educational bureaucracies can and should be ruthlessly
pruned.
176 ZOLTN KVECSES
The features of complex systems in question in these cases are (1) complex
systems becoming larger and (2) the reduction of complex systems. Additional
rich knowledge concerning plants is utilized to capture these features.
However, most of the metaphorical entailments that derive from the PLANT
metaphor in relation to complex systems have to do with mapping (c) above:
biological growth in the source corresponding to some abstract development in the
target. As will be seen, a huge amount of detailed knowledge is carried over from
plants to complex systems relative to this mapping. Here are the ones that stand
out on the basis of the Cobuild Metaphor Dictionary:
PREPARING THE DEVELOPMENT OF A COMPLEX SYSTEM IS PREPARING THE

GROWING OF A PLANT
- The work will prepare the ground for future development.
- These two chapters prepare the ground for the critical argument that
follows.
- Now they have signed agreements that lay the ground for a huge growth in
trade and co-operation.
- Their positions had not changed but they had laid the ground for working
together and that was very encouraging.
TO START OR CREATE A COMPLEX SYSTEM IS TO SOW A PLANT

A seed of doubt may have been planted in your minds.
- He had the skill to plant the seed in Jennifer's mind that her problem was
not so important.
- The emphasis must now be on sowing the seeds of such a movement.
- ...debate that sowed the seeds of the welfare state.
THE INITIAL STAGES OF DEVELOPMENT ARE THE BEGINNINGS OF GROWTH

- Typically thefirst green shoots of recovery herald an increase in bankruptcy.
- There would, he added, be no green shoots of economic recovery until
interest rates came down.
- In this way, problems that can lead to depression and even illness can be
nipped in the bud.
- They will run a workshop for budding authors on how to make, write and
illustrate their own books.
- He is not particularly serious about his budding recording career.
- Our budding romance was over.
- Another equally outstanding design was germinating at Bristol.
- The new phase in the relationship between father and son had germinated
on the long drive from Toronto.
- The book is an account of the germination and fruition of ideas as

experienced through a full career.
TO MAINTAIN OR TAKE CARE OF A COMPLEX SYSTEM IS TO CULTIVATE A PLANT

- He always cultivated friendships with the ruling class.
- ...technical universities which boast well-organized courses and carefully
cultivated links with industry.
- He may have cultivated this image to distinguish himself from his younger
brother.
- This will make it more difficult to weed out people unsuitable for the
profession.
- The police may need to establish ways of weeding out lazy and inefficient
officers.
- Those in the motor trade who ignore women customers deserve to be
weeded out.
- The worst material was never shown. It was weeded out by the television
companies themselves.
T H E SUCCESSFUL OR APPROPRIATE DEVELOPMENT OF A COMPLEX SYSTEM IS THE

HEALTHY GROWTH OF A PLANT
Exports flourished, earning Taiwan huge foreign currency reserves.
- His career is flourishing again.
- As the king refused to educate the public, ignorance and prejudice
flourished.
- ...the ruins of a once flourishing civilization.
T H E UNSUCCESSFUL OR INAPPROPRIATE DEVELOPMENT OF A COMPLEX SYSTEM IS

THE UNHEALTHY GROWTH OF A PLANT
- The center parties have achieved spectacular by-election results in the past,
only to see their support wither again in general elections.
- They had been innocent sweethearts at a German University but their
romance withered when they came back to England.
- I could see her happiness withering.
- The changes are likely to cause severe disruption for all the countries as the
old system withers away.
- The union, which was once the most powerful in the country, has shrivelled
under his leadership.
- The sympathy made something in him shrivel, shrink away.
- It was the kind of rain that shrivels the hopes of holidaymakers.
- He visibly wilted under pressure.
178 ZOLTN KVECSES
- The look the president gave the reporter made that experienced journalist
wilt before his eyes.
THE BEST STAGE IN THE PROGRESS OR DEVELOPMENT OF SOMETHING IS THE

FLOWERING OF A PLANT
- The relationship blossomed. They decided to live together the following
year.
- It was not until he joined her for a skiing holiday that their romance
blossomed.
- She had studied, worked, traveled and blossomed into an attractive
intelligent young woman.
- His business blossomed when the railway put his establishment within reach
of the big city.
- As her career blossomed, she kept her personal and professional lives totally
separated.
- ... a blossoming, diverse economy.
- Greta is very much enjoying having the baby. She is blooming.
- Their friendship flowered at a time when he was a widower and perhaps felt
lonely in his personal life.
- ...the nation that had briefly flowered after 1918.
- They remembered her as she'd been in the flower of their friendship.
- I feel I can still do it even though I am no longer in the flower of my youth.
- This was in the seventeenth century when modern science was in its first
flowering.
T H E BENEFICIAL CONSEQUENCES OF A PROCESS ARE THE FRUITS OR THE CROP OF

A PLANT
- Now they've finished will they sit back and enjoy the fruit of their labors?
- American and Japanese firms are better at using the fruits of scientific
research.
They enjoy the fruits of success and live well.
- ...the first fruit of the union between IBM and Apple.
- Their campaign seems to be bearing fruit.
- Sooner or later our common efforts will bear fruit.
- Last week their labor bore fruit and most achieved good exam results.
- They were eager to continue the long and fruitful association.
- She returned home after her fruitless efforts to find a job.
- Unfortunately a plan to reprint the play never came to fruition.
- You have the capacity to bring your ideas to fruition.
- Employers reaped enormous benefits from cheap foreign labor.
Reynolds reaped the reward for his effort by taking sixth place.
Cecilia's records are not yet reaping huge profits.
...a TV film that's reaped a clutch of international awards.
We have thousands of ideas to harvest.
The harvesting of knowledge from space will be one of the great scientific
endeavors of the next century.
Now we can see that most of the assumed literal features of the superordinate
concept of complex systems as given in section 1 have their counterparts in the
PLANT metaphor:
(1) the preparation of the development the preparation of the growth

of a complex system of a plant
(2) to create and start a complex system > to sow a plant
(3) the initial stages of the development the beginning of the growth of a plant
(4) the best stages in the development the flowering of the plant
(5) the appropriate development of the healthy growth of a plant
the system
(6) the inappropriate development of the unhealthy growth of a plant
the system
(7) to take care of a system
to cultivate a plant
(8) the benefits that the system yields the fruits or the crops of the plant.
Apparently, then, the COMPLEX SYSTEMS ARE PLANTS metaphor utilizes most of
the metaphorical entailment potential associated with the concept of plant. This
is knowledge that ordinary speakers of English (as opposed to experts such as
biologists) have about plants. The vast amount of rich knowledge focuses on one
basic constituent mapping of the metaphor, the mapping according to which the
natural, biological growth of plants corresponds to the (abstract) progress or
development of complex systems. This elaborate knowledge about the growth of
plants structures much of our knowledge about the "developmental" aspects of
complex systems.
Now if we regard abstract concepts, such as abstract complex systems, as
arising independently of metaphor (such as COMPLEX SYSTEMS ARE PLANTS), we
have no possibility to account for the content and structure of these abstract
concepts; indeed, we are left with no possibility for explanation at all. All we can
say is that the abstract concept has emerged out of thin air. By contrast, the
metaphor account can provide a detailed and systematic explanation why a certain
assembly of content and structure constitutes an abstract concept; the particular
assembly of conceptual elements is provided by the source domain of plants.
180 ZOLTN KVECSES
However, this raises a further issue: we have no account of what motivates

the metaphor itself that in turn motivates the abstract concept. Although the
structure and content of an abstract concept is now motivated (by a metaphor, i.e.,
COMPLEX SYSTEMS ARE PLANTS), we do not have any clear experiential motivation
for the metaphor itself. This creates a situation that requires some explanation,
since all metaphors must have some experiential basis in order to be readily
understandable. Is it the case that COMPLEX SYSTEMS ARE PLANTS does not have
this? The problem is similar to what Grady (1997) mentions in connection with
THEORIES ARE BUILDINGS. He points out that on the surface this metaphor has no
experiential basis, but if we recognize the existence of what he calls "primary"
metaphors, the experiential basis is obvious. In this case, we would have two
primary metaphors that underlie THEORIES ARE BUILDINGS: ORGANIZATION IS
PHYSICAL STRUCTURE and PERSISTING IS REMAINING ERECT. These primary
metaphors do have a clear experiential basis; namely, ORGANIZATION IS PHYSICAL
STRUCTURE is motivated by the fact that a physical object with a structure is
characterized by logical and causal relations among its parts and PERSISTING IS
REMAINING ERECT is motivated by the fact that many physical objects typically
perform their functions in a standing position.
As I have previously pointed out (Kvecses 1995,1997), metaphors such as
COMPLEX ABSTRACT SYSTEMS (like THEORIES) ARE BUILDINGS and COMPLEX
ABSTRACT SYSTEMS ARE PLANTS are characterized by central mappings that
function as "simple" (or in Grady's terminology, "primary") metaphors. (On the
notion of "central mapping," see Kvecses, n.d. a.) In these two cases, they would
be ABSTRACT STRUCTURE IS PHYSICAL STRUCTURE (equivalent to Grady's
ORGANIZATION IS PHYSICAL STRUCTURE) and ABSTRACT DEVELOPMENT IS
PHYSICAL GROWTH. We can say that these simple, or primary, metaphors have
"internal motivation," as opposed to other cases in the literature that can be said
to be "externally motivated" (such as the motivation of increased body heat for
ANGER IS A HOT FLUID IN A CONTAINER). What this means is that given a physical
concept, the concept is characterized by certain abstract properties which are
projected into the mental space of abstractness and thereby become target
domains. As we saw above, in the case of ORGANIZATION IS PHYSICAL STRUCTURE
the properties projected from concrete to abstract space include logical and causal
relations among concrete parts, while in the case of ABSTRACT DEVELOPMENT IS
PHYSICAL GROWTH they would include some temporal progression from an initial
state to a desired final state, as specified in the mapping above. The resulting
picture for the Metaphorical Emergence (3) view would thus be somewhat
different from the diagram that was given at the beginning of the paper:
ABSTRACT
CONCEPT
ABSTRACT
PROPERTIES OF
CONCRETE OBJECT
PARTS AND
PROPERTIES OF
CONCRETE OBJECT
Figure 4: Metaphorical Emergence (3)
In other words, this is a case of metaphor where the metaphor is motivated, the
motivation being "internal" to the source domain. These cases are markedly
different from the motivation that can be found in many other instances of
metaphor in the literature (e.g., ANGER IS HOT FLUID IN A CONTAINER) and that can
be said to be "externally motivated," in the sense that the basic experience that
motivates a metaphor lies outside the metaphor's source domain (as body heat lies
outside of the source concept HOT FLUID IN A CONTAINER, at the same time bearing
some resemblance to it).
4. Metaphorical emergence (4)

Another instance of an externally motivated metaphor can be found in
connection with the concept of marriage. Let us now return to the Literal
Emergence (2) view of the concept of marriage and see what an alternative
metaphor-based explanation would be like. We can begin by observing that in her
discussion it remains unclear whether Quinn equates the expectational structure
of marriage with the concept of marriage itself. Nowhere does she describe or
define marriage itself in terms other than its "expectational structure." This leads
one to believe that marriage is conceptualized by people in terms of this structure
only. But is it? Don't people have an idea of what marriage is before they have an
expectational structure of it? One would think that they do, yet this aspect of the
concept of marriage does not show up in the paper. Marriage is presented by
Quinn as an expectational structure, and all the other aspects of it that she
discusses, such as compatibility, difficulty, effort, success and failure, and risk, are
given as consequences of this structure. What, then, does the notion of marriage
consist of before it acquires its particular expectational structure?
182 ZOLTAN KVECSES
First and foremost, marriage is some kind of abstract union between two
people. To illustrate this conception, consider some definitions of marriage in a
sample of American dictionaries:
marriage 1 the state of being married; relation between husband and wife;
married life; wedlock; matrimony 4 any close or intimate union
(Webster's New World Dictionary, Third College Edition)
marry 1 a) to join as husband and wife; unite in wedlock b) to join (a man) to a

woman as her husband, or (a woman) to a man as his wife vi. 2 to enter into a
close or intimate relationship; unite
(Webster's New World Dictionary, Third College Edition)
marriage 1 a: the state of being married b: the mutual relationship of husband and
wife; wedlock; c: the institution whereby men and women are joined in a special
kind of social and legal dependence for the purpose of founding and maintaining
a family
(Webster's Ninth New Collegiate Dictionary)
marry 1 a: to join as husband and wife according to law or custom 2 to unite in

close and usu. permanent relation vi 2 to enter into a close or intimate union
(these wines - well)
(Webster's Ninth New Collegiate Dictionary)
marriage 1. a. The state of being husband and wife; wedlock b. The legal union
of man and woman as husband and wife
(The Heritage Illustrated Dictionary)
marry 1. a. To become united with in matrimony

(The Heritage Illustrated Dictionary)
married 1. United in matrimony

(Funk & Wagnalls Standard Dictionary)
As these dictionary definitions show, a major component of the concept of

marriage is the-legal, social, emotional, etc.union of two people. This seems to
be a large part of the notion that is prior to the expectational structure associated
with marriage. In other words, the prototypical, or stereotypical, idea of marriage
must include the notion that it is an abstract union of various kinds between two
people.
As Quinn suggests, the concept of marriage is structured by a mapping of
the American cultural conception of love. However, she only finds this in the
expectational structure of marriage. But now we can see additional structure in

marriage that derives from love. This is the notion of unity involving two people.
As Kvecses (1988, 1991) showed, the concept of romantic love is, in large
measure, understood and structured by the metaphor LOVE IS A UNITY OF TWO
COMPLEMENTARY PARTS, as can be found in expressions like You belong to me
and I belong to you, Theirs is a perfect fit, We're as one, She's my better half
They broke up, They're inseparable, and They match each other perfectly. It is
largely the functional unity of two physical parts that may serve as the source
domain for the abstract target concept of marriage. But more generally, our
understanding of non-physical-social, legal, emotional, spiritual, psychological,
etc.-unions derives from physical or biological unions. This is a perfectly regular
way in which human beings conceptualize and, by conceptualizing, also build
their non-physical world. We have seen the same process at work for the various
specific concepts that belong to "complex systems."
In other words, in the terminology of the view of metaphor that Lakoff and
Johnson (1980) initiated, we have the conceptual metaphor NON-PHYSICAL UNITY
IS PHYSICAL UNITY. (It is significant that the etymological root of the words union
and unity is the Latin word unus meaning "one.") This is the metaphor that
underlies our conception of various social, legal, psychological, sexual, political,
emotional, etc. unities and explains the use of such expressions as "to join forces,"
"the merging of bodies," "the unification of Europe," "to be at one with the
world," "a union of minds," "a deep spiritual union with God," etc. Obviously, the
metaphor also applies to marriage as a non-physical unity between two people.
Some examples from the above dictionary definitions include "to join in
marriage," "a marriage union" "the legal union of man and woman," and "to be
united in matrimony"; hence the metaphor MARRIAGE IS A PHYSICAL AND/OR
BIOLOGICAL UNITY OF TWO PARTS. Not surprisingly, we find examples of this
metaphor in the data that Quinn presents. She names what we call the MARRIAGE
IS A PHYSICAL AND/OR BIOLOGICAL UNITY metaphor "two inseparable objects," as
in We knew we were going to stay together and "an unbreakable bond," as in That
just kind of cements the bond (Quinn 1991:68).
At this point it might be objected that my analysis is largely based on
dictionary data and that Americans may not conceptualize marriage according to
the unity metaphor. We have some evidence that they do. The evidence is both
direct and indirect. In 1992 at Rutgers University, New Jersey, in an informal
experiment I asked students in an Introduction to Anthropology course to write
down linguistic expressions about marriage. They came up with dozens of
different expressions, including the ultimate bond, She's my ball and chain, They
are a match made in heaven, They 've tied the knot, She's my better half They
broke up, I can't function without her, They're getting hitched, They dissolved
184 ZOLTN KVECSES
their union, and others. These are all UNITY metaphors or at least closely related
to this metaphor. They suggest that the notion of UNITY is not alien to many
Americans when they talk and think about marriage. The indirect evidence comes
from a set of interviews that a student of mine, Ted Sablay, conducted concerning
romantic love in the summer of 1996 at the university of Nevada, Las Vegas. The
interview subjects were seven male and seven female students from roughly the
same white middle-class background. What the interviews reveal about romantic
love should be taken seriously in dealing with marriage because, as Quinn herself
claims, marriage is in many ways structured by our understanding of love. In his
report on the project, Ted Sablay found that the most frequent metaphor for love
is the unity metaphor for his interview subjects. This gives us some reason to
believe that, at least for some Americans, the conception of marriage is still built
on the idea of forming a unity with another and that this notion is not just an
antiquated dictionary definition.
What is the relationship between the idea of marriage-as-a-non-physical-
unity and the expectational structure of marriage that Quinn describes? We can
suggest that the conception of marriage as a unity between two people is the basis,
or the foundation, of its expectational structure; namely, that marriage is expected
to be shared, beneficial, and lasting. The reason that marriage is expected to be all
these things is that it is conceptualized as a unity of a particular kind: the physical
unity of two complementary parts, which yields the metaphor MARRIAGE IS THE
PHYSICAL AND/OR BIOLOGICAL UNITY OF TWO COMPLEMENTARY PARTS. The
details of the UNITY metaphor for marriage can be given as a set of mappings:
(1) the two physical parts the married people

(2) the physical joining of the parts the union of the two people in marriage
(3) the physical/biological unity the marriage union
(4) the physical fit between the parts the compatibility between the married people
(5) the physical functions of the parts. the roles the married people play in
in the unity the relationship
(6) the complementariness of the the complementariness of the roles of
functions of the parts the married people
(7) the whole physical object the marriage relationship
consisting of the parts
(8) the function of the whole object the role or purpose of the marriage
relationship
What we have here is a source domain in which there are two parts that fit each
other and form a whole, where the particular functions of the parts complement
each other and the parts make up a larger unity that has a function (or functions).
This source schema of a physical unity has parts that are additional to the basic
experience between baby and first caretaker. Unlike the infantile experience, here
two originally separate parts are joined, or put together; unlike the infantile
experience, there is a pre-existing fit between the parts; unlike the infantile
experience, the whole has a function that is larger than, or goes beyond, the
functions of the individual parts. What corresponds to these in the target domain
of marriage is that two separate people who are compatible join each other in
marriage with some life goal(s) in mind. It is this structure that appears in the way
many people (Americans) think about marriage. But this way of conceptualizing
marriage is simply a special case of the larger process whereby non-physical
unities in general are constituted on the analogy of more physical ones. It is
important to see that the physical unity metaphor characterizes not just marriage
but many other abstract concepts where the issue of non-physical union arises, that
is, abstract concepts that have union as one of their dimensions, or aspects. This
dimension of non-physical union emerges from the content and structure of what
we called "physical unity (of two complementary parts)" source domain. In this
sense, abstract concepts that acquire the dimension of non-physical union can only
be metaphorical. This is for the simple reason that the dimension inevitably
emerges from the physical source of physical unity. The application of this simple,
constitutive metaphor to marriage is both transparent and important. Its signifi
cance lies in the fact that in the concept of marriage non-physical union is a core
dimension. Indeed, it is so fundamental that, as we will see shortly, the
expectational structure that Quinn identified derives from it.
In Quinn's view, the basic experiences constitute cultural models (like those
of abstract concepts in general and that of the concept of marriage in particular)
and the cultural models select the fitting conceptual metaphors. In my view, it is
the basic experiences that select the fitting conceptual metaphors and the meta
phors constitute the cultural models. As we saw above, there are differences bet
ween what the basic experiences and what the conceptual metaphors can yield
relative to abstract concepts. Basic experiences in themselves could not account
for the content and structure of the concepts of love and marriage (just as they
could not account for the cultural model of anger). The more that is needed is
provided by such constitutive metaphors as NON-PHYSICAL UNION (in love and
marriage) IS PHYSICAL UNITY and ANGER IS A HOT FLUID IN A CONTAINER.
This metaphorically structured understanding of marriage forms a definition
of marriage and provides its expectational structure. The definition could be given
as follows: marriage is a union of two people that are compatible with each other.
The two people perform different but complementary roles in the relationship.
Their union serves a purpose (or purposes) in life. This is, of course, a generic-
level definition that can be filled out with specific details in individual cases.
186 ZOLTN KVECSES
The expectational structure arises from the definition in the following way:
because a part by itself is not functional, people want to share their lives with
others in marriage. Because only one or some parts fit another part, people want
compatible partners in marriage. Because (to get a functioning whole) a part must
perform its designated function, people want to fulfill their designated roles in a
marriage relationship. Because wholes have a designated function to perform,
marriage relationships must be lasting.
As can be seen, this is similar to Quinn's expectational structure. One
difference, though, is that in our characterization compatibility is a mapping in the
UNITY metaphor, while in hers it is a consequence that follows from the expecta
tional structure. Another difference is more substantial. It is that we have given the
expectational structure of marriage as a consequence of a certain metaphorical
understanding of marriage, one that is based on the metaphor NON-PHYSICAL
UNITY IS PHYSICAL UNITY. It is in this sense that we claimed that the concept of
marriage is metaphorically constituted. (A similar argument is presented in
Kvecses, n.d. b.)
In sum, what Quinn calls the expectational structure results from a particular
metaphorical understanding of marriage. Thus, marriage is not a literally concei
ved abstract concept, although the metaphor that yields the expectational structure
is based on certain bodily experiences.
5. Conclusions
To attempt to answer the question whether metaphors constitute or simply
reflect cultural models requires an answer to the question how abstract concepts
emerge. We have considered several possibilities for the emergence of abstract
concepts: (1) Literal Emergence, (2) Literal Emergence from some basic expe
rience, (3) internally motivated Metaphorical Emergence, and (4) Metaphorical
Emergence motivated by some external experiential basis. We have argued that
Literal Emergence (1) is not a viable way of thinking of the emergence of abstract
concepts because it provides no account whatsoever of why abstract concepts have
the particular content and structure that they do. We have also argued against
Literal Emergence (2), a view which maintains that abstract concepts emerge
directly - without the mediation of metaphor - from basic human experience. In
particular, we pointed out that Quinn's analysis of American marriage leaves out
of consideration a large and significant portion of this concept - the part which is
metaphorically conceived and from which the expectational structure of marriage
derives. The notion of marriage, on our analysis, is partially based on and con
stituted by the generic metaphor NON-PHYSICAL UNITY IS PHYSICAL UNITY. Given
this metaphor, we can naturally account for why marriage has the expectational
structure that it has, as well as for the fact that the same metaphor applies to many
domains that are seemingly unrelated to marriage or love.
I suggested in this paper that a number of abstract concepts can only emerge
metaphorically: either via Metaphorical Emergence (3) or Metaphorical Emer
gence (4). Conceiving of their emergence this way, we can offer a systematic and
integrated account of not only such specific concepts as anger, love and marriage
but abstract concepts in general.
References
Deignan, Alice, ed. 1995. Cobuild English Guides 1: Metaphor. London: Harper
Collins.
Gibbs, Raymond W. 1994. The Poetics of Mind: Figurative Thought, Language,
and Understanding. New York: Cambridge University Press.
Grady, Joseph E. 1997. "THEORIES ARE BUILDINGS revisited". Cognitive Linguis-
tics 8. 267-290.
Holland, Dorothy, & Naomi Quinn, eds. 1987. Cultural Models in Language and
Thought. New York: Cambridge University Press.
Kvecses, Zoltn. 1988. The Language of Love. Lewisburg: Bucknell University
Press.
1991. "ALinguist's Quest for Love". Journal of Social and Personal Rela-
tionships 8.11-91.
1995. "American Friendship and the Scope of Metaphor". Cognitive Ling-
uistics 6. 315-346.
1997. A Student's Guide to Metaphor. Manuscript.
n.d.a "The scope of metaphor". Metaphor and Metonymy at the Cross-
roads, ed. by Antonio Barcelona. Berlin: Mouton.
n.d.b Metaphor and Emotion. Language, Culture, and Body in Human
Feeling. New York: Cambridge University Press.
Reveal about the Mind. Chicago: The University of Chicago Press.
, & Mark Johnson 1980. Metaphors We Live By. Chicago: The University
of Chicago Press.
, & Zoltn Kvecses 1987. "The Cognitive Model of Anger Inherent in
American English". Cultural Models in Language and Thought, ed. by D.
Holland, & Naomi Quinn, 195-221. Cambridge: Cambridge University
Press.
Quinn Naomi. 1991. "The Cultural Basis of Metaphor". Beyond Metaphor: The
Theory of Tropes in Anthropology, ed. by J. Fernandez, 56-93. Stanford:
Stanford University Press.
18 8 ZOLTN KVECSES
Shore, Bradd. 1996. Culture in Mind: Cognition, Culture, and the Problem of
Meaning. New York: Oxford University Press.
METAPHORS AND CULTURAL MODELS AS PROFILES
AND BASES
ALAN CIENKI
Emory University
For it is in metaphor - perhaps more dramatically than in any other form of

symbolic expression - that language and culture come together and display
their fundamental inseparability. A theory of one that excludes the other
will inevitably do damage to both. - Basso (1976)
None of the points I will make is radically new and revolutionary. Most of
them may seem obvious to anthropologists, working as they do on the
'messy' side of the language/culture boundary that linguists avoid as best
they can. - Keesing (1979)
1. Introduction: The problem to be considered

While it may be true that the issues raised in this paper will not be
revolutionary to anthropologists, perhaps this research will at least advance ling
uistics one more step toward the "messy", cultural side of language. This work
is motivated by interrelated research from several theoretical perspectives. One
of them is the view of meaning (semantic structure) as a kind of conceptual
structure. In cognitive grammar, for example, "...semantic structures are regarded
as conceptualizations shaped for symbolic purposes according to the dictates of
linguistic convention" (Langacker 1987:98). Another is a basic premise from
which many, if not all, who consider themselves 'cognitive linguists' work - that
linguistic meaning is encyclopedic in nature. As Langacker (1987:163) expresses
it, "The entity designated by a symbolic unit can therefore be thought of as a point
of access to a network. The semantic value of a symbolic unit is given by the
open-ended set of relations - simple and complex, direct and indirect - in which
this access node participates." The present study is also motivated by research on
the integral relations between metaphor and cultural models (e.g., Lakoff and
Kvecses 1987, Quinn 1987, Sweetser 1987a, Emanatian 1996). Here I will
follow Shore's ( 1996) explication of cultural models as falling into a large number
of genre forms, including linguistic and non-linguistic, spatial, social, diagnostic,
and others. While metaphorical models represent one kind of cultural model,
190 ALAN CIENKI
higher level conceptual metaphors, in particular, certainly play a role in various

kinds of cultural models. A final motivating issue is the problem of the interre
lation between shared, cultural models and individual cognitive or mental models.
As Shore (1996:47) notes, "Cultural models are constructed as mental representa
tions in the same way as any mental models with the important exception that the
internalization of cultural models is based on more socially constrained experi
ences than is the case for idiosyncratic models."
This complex of issues brings out various aspects of language as both a
cognitive and a social phenomenon. The dilemma to be addressed here is how to
handle these complex relations practically when trying to analyze specific lexical
items; in this paper, the semantics of terms for two different moral concepts in
Russian will be considered. If meaning is encyclopedic, and involves metaphors
which are integrally tied into various cultural, and cognitive models, where does
one draw the line on what constitutes the semantics of a particular word?
2. Previous approaches to this problem

In some respects, this is not the first time that this problem has been
confronted in semantic analysis. Keesing (1979), for example, discusses how
cultural knowledge is drawn upon to understand the meanings of particular lexical
items, and thus how adequate semantic representations "may have to be articulated
with, predicated on, and constructed with reference to such cultural knowledge"
(Keesing 1979:20). Yet when presenting a semantic analysis, how can one avoid
constantly repeating the presupposed cultural knowledge in the explanation of the
semantics of individual lexical items?
Keesing suggests several ways for linguistic description to begin to incor
porate and make reference to a cultural analysis. One is to admit key concepts
from the given cultural system into the system one is using for semantic des
cription. Another is to key the semantic analysis to culturally salient distinctions,
more specifically, to prominent symbolic oppositions in the culture. For speakers
of the language Kwaio on the Solomon Islands he notes some oppositions which
are also prominent in European cultures, such as male:female, up:down,
culturemature, and others which are less prominent, such as sacred:(mola:)pol-
luted and ancestors:humans. This is reminiscent of what Ivanov and Toporov
(1965) call 'semiotic modeling systems', pairs of oppositions which often co-align
in a coherent way within a culture, and what van Leeuwen-Turnovcov (1991,
1994, and elsewhere) discusses as 'cultural paradigms'.
Palmer's (1996) discussion of how cognitive linguistics is especially well-
suited to help bridge some theoretical and methodological gaps with anthropology
is also instructive. Palmer presents an approach to semantics which incorporates
cultural analysis by relying on the distinction made in cognitive grammar between
METAPHORS AND CULTURAL MODELS AS PROFILES AND BASES 191
a 'profile' and a 'base'. Langacker (1987) observes how the entities, processes,
and relations designated by language are merely highlighted parts of implicit,
larger wholes. In the framework of cognitive grammar, Langacker refers to the
highlighted substructure as the 'profile' and the whole to which it relates, the
scope of predication, as the 'base'. (Note the forerunner in Talmy's [1983] appli
cation of the Gestalt psychology distinction between 'figure' and 'ground' to the
study of language.) As Langacker (1987:183) explains, "The semantic value of
an expression resides in neither the base nor the profile alone, but only in their
combination; it derives from the designation of a specific entity identified and
characterized by its position within a larger configuration." Palmer extends the
application of this notion by observing that a semantic relation can be thought of
as the relation profiled against a culturally modeled base. Recasting Sweetser's
(1987a) analysis of lie, Palmer (1996:186) notes how a speech act can only be
judged a lie (the conceptual profile) relative to a certain cultural model of
communication (the conceptual base).
3. Terms for two moral categories as a case study

We will consider how to practically take these factors into account by
analyzing two terms for moral categories in Russian, one described by the root
chestn-1 (as in the words chestnosf, chestnyj, chestno (roughly translatable as
"honesty", "honest", and "honestly"), and the other characterized by the root
porjadochn-, as in the words porjadochnosV, porjadochnyj, porjadochno
(roughly: "honesty/decency", "honest/decent", and "honestly/decently"). (Note
that historically, the root chestn- is derived from chest' "honor" whileporjadochn-
comes from porjadok "order" which is derived from rjad "row".) The terms are
of particular interest because, although they are closely related, they are
characterized in terms of contrasting metaphors.
The data consist of over 20 videotaped conversation-interviews with pairs
of Russian university students conducted in Moscow in May of 1996.2 The data
were gathered as part of a larger study on the semantics, metaphors, and cultural
models related to terms for "honesty" as used by Russian and American university
students. The students were only told beforehand that I was interested in differen
ces in academic practices between American and Russian universities. The stu
dents were given three written questions to discuss with each other which con
cerned how they would describe how students take exams at their university, what
it means to conduct oneself honestly {chestno) during an exam, and how they
would feel if they found out that a classmate had received a higher grade on an
exam than they did as a result of cheating. They then read a brief scenario descri
bing how a couple of fictitious students found a problem set that was going to be
on an upcoming math exam, and how one of the students actually used them to
192 ALAN CIENKI
prepare for the exam. The interviewees were asked whether they thought the
students acted honestly {chestno) or not. For the last part of the interview I asked
them a few additional questions including, given all that they had talked about,
what they thought honesty (chestnost') was, and how they would compare it with
porjadochnost' The procedure, therefore, involved both implicit and explicit eli-
citation of the speakers' thoughts on the use and meanings of these words. Most
of the conversation-interviews lasted 25 to 40 minutes. Participants were infor
med of the full objectives of the project in a debriefing after completion of their
conversation-interview. As the analysis of the videotapes is ongoing, the findings
presented here are preliminary; nevertheless they provide interesting insights into
metaphors viewed against the background of cultural models.
3.1 Metaphors for chestnost'

Consider the following examples of how chestnosf was described.3
(1) [23] A - Chestnosf - ona bolee prjamolinejna, chrno-belogo cveta, takogo.

"Chestnosf - it's more straight-lined, like black and white.''
(2) [15] B - Nu, chestnosf - to zhstche, chem porjadochnost', vottakvot.

"Well, chestnost' is more rigid than porjadochnost', like that."
(3) [23] B - U nas kak-to, votja ne znaju. Menja .. ostanavlivaet voobshche kategorija
chestnostfi/nechestnosti.] [Ja ne- ne ponimaju. ]
A- [Delo v torn, chto-] Da. [Uni- Universi- UJniversitet, vs-taki, ne shkola,
kogda prepodavateV stremitsja vo to by chto ni stalo, vot... sognat' studentov
v kakie-to ramki.
B - "Here somehow, I just don't know. I'm .. basically stumped by the category
of chestnost' [/nechestnost'.] [I don't- don't understand.]"
A- ["The thing is that- ]Yeah. [The uni- universi- u]niversity isn't grade
school, after all, when the teacher tries no matter what, well... to round up the
students into some kind of boundaries (frames)."
Chestnosf was described, for example, as prjamolinejna ("straight-lined"), as an

absolute category {chrno-belogo cveta "black and white"), zhstche ("harder,
more rigid"), and as involving ramki ("limits, bounds") - in general, as an object
that is straight and rigid or as a space or path with definite boundaries. These, and
other, metaphorical expressions provide evidence of the following conceptual
metaphors among these speakers of Russian:
CHESTNOST' IS STRAIGHT (LINEAR)

CHESTNOST' IS RIGID
CHESTNOST' IS A BOUNDED SPACE.
However, the evidence for these metaphors is not limited to verbal expressions.
One major reason that these conversation-interviews were videotaped was to
gather data on the gestures that co-occur with speech. As previous research
(Calbris 1990; McNeill 1992; Cienki 1998) shows, evidence for conceptual
metaphors can also be found in spontaneous gesture, supporting the view that
metaphors are not restricted to expression only in spoken, signed, or written
language.
The gestures observed in this data provide another indication that these
metaphors reflect ways speakers are thinking about this abstract concept. In some
examples, the metaphor expressed by a gesture is not explicit in speech, and in
other examples, the gesture provides a missing lexical item, missing because the
gesture made the point better for the speaker than words could. Examples (4) and
(5) reflect the proposed metaphor CHESTNOST' IS STRAIGHT (LINEAR). 4
(4) [9] A - Dlja menja chestnost' to nekaja absoljutnaja kategorija. Kogda vot est' situacija,
seichas vostupit' chestno ftak}.
"For me chestnost' is a kind of absolute category - when there's this situation,
then (you need) to act honestly {like this!."
(5) [19] A - Antonim (chestnosti) - to Izhivost' i obman ...

kogda luchshe solgat' v blago zhe tarn chego-to, chem skazat' pravdu v Hi ho.
"The antonym (of chestnost') is falsity and deception ...
when it's better to lie for the good of something or other than to say the truth to
someone's {face}."
In example (4), the speaker displays a rigid, flat right hand, palm facing left, while
saying "like this". In (5) the speaker makes a straight-line motion with her open
right hand from in front of her face toward my face, as I am sitting facing her. As
the last example shows, metaphors characterizing chestnost' as 'straight' in Rus
sian seem to focus more on the negative aspects that this entails (directness as
rude) as opposed to the more positive evaluations of honesty as 'straight' and
'proper' found in the comparable conversation-interviews conducted with Ameri
can students.
Finally, it should be noted that chestnost' was characterized as something
abstract, that is, not as easily related to one's practical, lived experience:
(6) [23] B - V bytu ono ne ispol'zuetsja, vot, vot s- v moej srede ne ispol'zuetsja v bytu,
to est' to ponjatie iz oblasti ideaVnogo...
"In everyday life it isn't used, so in my environment it's not used in everyday life,
that is it's a concept from the realm of the ideal..."
194 ALAN CIENKI
3.2 Metaphors for porjadochnost'

Porjadochnost', however, was characterized via quite different metaphors.
(7) [23] A- Jadumaju, chtoporjadochnost' -boleeshirokoeponjatie.

"I think that porjadochnost' is a broader concept."
(8) [23] B - Porjadochnost' - razmytoe bolee ponjatie, gumanizirovannoe chto-li,...

"Porjadochnost' is a more washed out concept, humanized sort of, ..."
(9) [7] B - A votporjadochnost' okazyvaetsja bolee prizemlnnym ponjatiem.

"But porjadochnost' turns out to be a more down-to-earth concept."
Again, the gestures reveal the metaphorical way in which the speaker is thinking
about these concepts, which sometimes is expressed verbally, and sometimes not.
In (10) the speaker characterizes chestnosf by holding her hands stiff in front of
her, palms facing each other, rhythmically moving them up and down on the
stressed syllables (what McNeill [1992] calls 'beats') while speaking the
underlined words of the first sentence. In the second sentence, when discussing
porjadochnost', she spreads her two hands out flat in front of her, palms facing
down, moving them away from her body and outwards, away from each other, as
if over the surface of a table.5
(10) [23] B - Skazhem tak, chestnosf - to bolee formaliz.ovannoe sobliudenie kakix-to norm.
A porjadochnost', ona vkljuchaet sjuda ochen' ochen' boVshuiu is- (?)
moraVnye} plasty. Ona bolee shirokaja.
"Let's put it this way, chestnosf is a more formalized observation of certain norms.
But porjadochnost'', it includes here a very very big {(s- ?)
moral} layers. It's broader."
In (11) the speaker contrasts porjadochnost' with chestnosf and its rigidly fixed
category borders. The gesture again involves the two hands rigidly facing each
other in front of the speaker, and slightly curved, as if holding a large bowl. She
makes small beats with her hands in that position during the speech in curly
brackets.
(11) [23] B - Porjadochnost' - razmytoe bolee ponjatie, gumanizirovannoe chto-li;

ne otnositsja k ... k vot {k zhstko fiksirovannym tarn ..}...
"Porjadochnost' is a more washed-out concept, humanized sort of;
it doesn't relate to- ..to like (to rigidly fixed there ..}..."
The verbal metaphorical expressions for porjadochnost' included bolee shirokaja

("broader") and bolee razmytoe ponjatie ("a more washed out concept"), but also
prizemljonnoe ponjatie ("a down-to-earth concept"), and reflected it as a less
clearly delimited space or object, but also one that relates more closely to the
complexities of real life. These and the metaphoric gestures support at least the
following metaphors for porjadochnosf among these speakers:
PORJADOCHNOST' IS A BROAD SPACE

PORJADOCHNOST' DOES NOT HAVE CLEARLY DEFtNED BOUNDARIES.
Porjadochnost' is not conceived by these individuals as a mathematically defined,

rigidly bordered space defined by ideal rules, but a human space
(gumanizirovannoe "humanized"), more relevant to the real behavior of people in
society.
4. The wider context of usage

Many interviewees equated chestnost' with "telling the truth", as in
examples (12) and (13).
(12) [15] A - Chestnost' - to govorit' pravdu. B - Ne govorit' lozh'.

A - "Chestnost' is telling the truth." B - "Not telling a lie."
(13) [21] A - to nazvat' veshchi svoimi imenami.

"It's calling things by their own names."
The metaphorical characterization of chestnost' as a straight path is coherent with

what is often described as the CONDUIT metaphorical model of (truthful)
discourse. Consider the reinterpretations of Reddy's (1979/1993) CONDUIT meta
phorical model of speech commonly assumed in American and many other
cultures. Vanparys (1995) includes among its components the metaphor that COM
MUNICATION is TRANSFERRING WORD/SENTENCE/TEXT-CONTAINERS, and Grady
(1998) proposes TRANSMISSIONS TRANSFER. The most felicitous transfer is pre
sumed to occur in a straight fashion with no obstructions (cf. Sweetser [1987b]
andEmanatian [1996] on discourse journeys). Cienki (1998) offers evidence for
the metaphor TO SPEAK IN A MAXIMALLY INFORMATIVE WAY (truthfully, honestly)
IS TO TRANSFER WORDS ALONG A STRAIGHT PATH among both English and Russian
speakers. However, the positive or negative evaluation of honest speech is depen
dent on context. So while telling the truth is seen as good, especially from the
adult perspective if children do it, it can also be rude (cf. skazat' pravdu prjamo
v lico "to say the truth straight to someone's face") given cultural models of
politeness that not all truths need to be spoken out loud.
Sweetser ( 1992) and Emanatian ( 1996) provide convincing evidence for the
following metaphors relating to thought and reasoning among speakers of Ame-
196 ALAN CIENKI
rican English. The present data from Russian indicate that these metaphors are,
indeed, more widespread.
IDEAS ARE LOCATIONS (POINTS) IN SPACE

THINKING IS MOVING FROM ONE IDEA-SPACE TO ANOTHER
LOGICAL THOUGHT IS STRAIGHT
DECIDING IS CHOOSING BETWEEN MULTIPLE PATHS
To tie in the present data, metaphorically MOVING (that is, acting) in one
DIRECTION (manner) along the STRAIGHT PATH of chestnosf does not require deci
sion-making; it involves simpler rules, doing things 'straight'. Porjadochnosf,
however, is more complex, involving social norms that depend on the situation.
MOVEMENT (action) in the less clearly delimited SPACE of porjadochnosf does
require thoughtful action and can entail a variety of consequences. This may also
explain why many of the Russian interviewees associate chestnoe povedenie
("chestnyj behavior") with children, and porjadochnoe povedenie ("porjadochnyj
behavior") as something adults are more conscious of. Consider example (3) in
which the speaker deemphasizes the importance of chestnosf' at the university
versus in grade school. In that same interview, the other speaker says the
following (14):
(14) [23] B -Ja dumaju, chto to zavisit ot togo - vot, raznicu m/ v ponimanii tix.. veshchej,v
kakoe vremja v nashej zhizni v obshchem vxodjat ti ponjatija. Chestnosf - to
(?) tak skazaf v nas, nachinaja uzhe v sadik, tam "to nechestno, ne xorosho tam,
ne krasivo". Porjadochnosf - to uzhe na urovne ... pojavljaetsja tam v
podrostkovom vozraste, kogda bolee slozhnye moral' nye kategorii pojavljajutsja.
"I think that it depends on - the difference b/ in the understanding of these.. things -
on what time in our life in general these concepts come in. Chestnosf - it's (?), so
to speak, in us, beginning in kindergarten already, like 'That's not chestno, not good
there, not nice.' Porjadochnosf - it's already on a level ... appears there in
adolescence, when more complex moral categories appear."
The metaphors for chestnosf and porjadochnosf are also integrally related to
other cultural models which are not necessarily metaphorically-based, but are
shared by Russians who have gone to Russian schools, especially if they have also
attended a Russian university. For example, Russian students are normally
admitted to specific departments of study, and go through the university in groups
in each department, taking the same classes together. As the interviewees indica
ted, the bonds between students in a group are very strong, and members of a
group rely on each other a great deal, both academically and socially. Treating
each other right is consequently viewed as more important than following the rules
imposed by the institution, the university. (Contrast this with the American system
in which college students choose their own classes, with students majoring in the
same subject only sometimes sharing classes together. Studying independently,
rather than working together, is more the norm.) The larger opposition in Russian
culture - what Ivanov and Toporov (1965) call a semiotic modeling system, and
what Shore (1996) might call a foundational schema - of svoj ("one's own") ver
sus chuzhoj ("other") ties in with the system of groups, such that acting in accor
dance with the interests of one's own group is deemed very important. This rela
tes to the positive evaluation of empathy for members of one's group. One stu
dent, speaking about porjadochnost', said:
(15)[15] B - Da, vsegda stavit' sebja na ix mesto. Samoe glavnoe. to tochno

samoe glavnoe ... u cheloveka. To, chestnost', uzhe - ne objazatel'no.
"Yeah, always putting yourself in their place. The most important thing. It's
definitely the most important thing ... with a person. But chestnost', well - not
necessarily."
Similarly, the positive evaluation of porjadochnost' fits in with the importance

placed on observing certain culturally shared norms of behavior in society, as
mentioned by the speaker in (16).
(16) [13] A - Porjadochnost' - naverno sobljudenie obshcheprinjatyx ustanovlennyx kakix-to

norm v obshchestvennyx.. to est', .. tika tam, povedenie v obshchestvennyx
mestax.
"Porjadochnost' is probably the observation of some kind of established generally
accepted norms in society .. that is, .. like ethics, behavior in public places."
Indeed, the history of relations in Russia between 'the common people' and autho
rity figures, the communist authorities in particular, the power differential between
them, and the fact that the student-interviewees are still part of a generation raised
under the communist system, must not be ignored. The importance of not telling
on others, especially on members of one's own group (despite institutio
nal/university rules to the contrary) should not be underestimated. Several stu
dents mentioned the importance of not telling on fellow students who were getting
answers from others during an exam. The phrase Ne donosi ("Don't inform") was
mentioned more than once as a guiding principle of behavior. This, too, favors the
positive evaluation of the metaphorical flexibility of porjadochnost' over the
rigidity of chestnost' in regard to rules of behavior.
The Russian students' valuing of porjadochnost' over chestnosf in the
context of taking exams also fits in with the interpersonal nature of how Russian
198 ALAN CIENKI
university students take most exams, as a one-on-one meeting with the professor
or with a committee of professors during which the student gives an extensive
narrative (oral) answer to one or more questions chosen at random beforehand
from slips of paper. One student concluded,
(17) [20] B - Voobshche i ona, chestnost' nasha, ne formal'no ne polagaetsja. Eto ix

(prepodavatelej) sovershenno ne interesuet. Ix naverno interesuet kak-to obmen
znanija, a uzh chestnost' -nechestnost' - v zadnem plane.
"In general it, our chestnost', is not formally expected. It doesn't interest them
(instructors) at all. They're probably interested in some exchange of knowledge,
but uh chetnost'-nechestnost' (honesty-dishonesty) is in the background."
The format of the exam, choosing a slip of paper with the question on it that you
are required to answer, probably contributes to the generally held view among
students that it is like a lottery, and that the exam will not necessarily reflect what
the student really knows. The word for the slip of paper, bilet, is also the same as
that for "ticket". The metaphor of an exam as a lottery frames it as something left
up to chance rather than being a competition between participants.
(18) [23] A - I kogda on prixodit na kzamen, on, putm zhrebija ili loterei, vybiraet sebe bilet,
v kotorom dva ili tri voprosa iz vsego kursa.
"And when he comes to an exam, as in a drawing or lottery, he chooses a ticket for
himself on which there are two or three questions from the entire course."
(19) [23] A-No inogda byvaet ob"ektivno nu popal neschastlivyj bilet.

"But sometimes, objectively, it happens, well, you get an unlucky ticket."
(20) [16] B - kzamen - nekaja fortuna.

"An exam is a kind of fortune."
Helping fellow students by sharing answers, either by copying during a written

exam or sharing information while preparing an oral answer for the question on
the bilet one has chosen, is technically against university rules, and therefore
nechestno ("dishonest"). However, since most exams (other than entrance exams
to the university) are not a matter of competition, acting porjadochno and helping
one's fellow students, rather than chestno and not helping, is generally considered
by students the more socially acceptable behavior.
Returning to Palmer's (1996) recasting of Langacker's (1987) distinction
between a profile and a base, such that a semantic relation can be thought of as the
relation profiled against a culturally modeled base, we can see how various
cultural models provide different bases according to which metaphors, as profiles,
may be interpreted and evaluated as positive or negative (table 1). This overview
of the metaphors for chestnost' and porjadochnost' reveals interesting parallels

with what Lakoff (1996) calls the 'strict father' and 'nurturant parent' models of
morality; however, detailed comparison with them must remain a subject for
future study.
some aspects profiled

by the metaphors chestnost' porjadochnost'
spatial: linear (straight) planar / broad space
clearly bounded fuzzy boundaries
force-dynamic: unidirectional motion multidirectional motion
rigid flexible
realm of existence: ideal human
type of order/rules involved: simple complex
type of behavior involved: follow the ideal rule consider many interpersonal
factors/relations
evaluation of metaphors for
chestnost' and porjadochnost'
some of the bases within these bases
cultural model of discourse tell the truth (respect others in your group)
cultural model of action act individually conforming act with respect to others
to the institutional rules in-group
cultural models of politeness possibly rude polite
cultural model of evaluating officially proper po-chelovecheski (human/
behavior humane)
cultural models of age- important for children important for adults
appropriate behavior
cultural model of exam- do own work, don't cheat help others in your group get
taking (student perspective) through exams
(students') cultural model of irrelevant why not help others win too?
exam as lottery
cultural models of relations Ne donosi! ("Don't report"/
to authority (non- don't squeal on
authority perspective) others, esp. in-group)
Table 1 Metaphors for chestnost' and porjadochnost' profiled against the bases of
various cultural models
While recognizing the role of metaphor in our understanding of abstract concepts

is one way of enriching semantic analysis, acknowledging how metaphorical and
other cultural models interrelate can take the study of semantics even further
toward its "messy" side. As Palmer (1996:292) states, "Most linguistic usages and
discourses exist somewhere on a continuum from conventional, mutually
presupposed and traditionally situated meanings to meanings emergent in unique
confluences of society, culture, history, and discourse itself." The study of this
200 ALAN CIENKI
relationship between metaphorical conceptualizations of abstract domains, such

as moral categories, and how they are contextualized with respect to different
cultural models can give us further insight into meaning as something both
situated and emergent. Some cultural models of behavior are more relevant and
prominent in speakers' minds in certain situations than in others. As Shore
(1996:313) expresses it, "the mind is best thought of as having a polyphonic
structure, and... knowledge of cultural models is always layered as different kinds
of knowledge, at different degrees of distance from focal awareness." How the
metaphor CHESTNOST'IsSTRAIGHT, for example, is interpreted in a given context
will depend in part on the cultural model according to which it is being framed:
as "proper", conforming to (metaphorically STRAIGHT) institutional rules of
behavior; or as "rude", according to the model of polite behavior which frames
honest speech as metaphorically STRAIGHT and RIGID, and therefore capable of
injuring someone in a metaphorical (emotional) sense just as a straight, rigid
object (like a knife) could injure someone in a physical sense.
The differing spatial/force-dynamic gestalts profiled metaphorically through
the roots chestn- (straight, bounded, rigid, etc.) and porjadochn- (broad, fuzzy
boundaries, flexible) might be thought of as distinct "foundational schemas" (see
Shore 1996) in Russian culture, general patterns which organize or link up
"families" of related cultural models. Shore (1996:53) notes that the image
schematic patterns discussed by Johnson and Lakoff, such as CENTER-PERIPHERY,
CONTAINER (inside-outside), and PATH (source-path-goal) may be instantiated in
numerous models shared by many members of a group, and so can provide
coherence to a shared worldview. Similarly, the linear/rigid and broad-
space/flexible patterns discussed here are repeatedly highlighted as contrasting
metaphoric characterizations of various culturally-modeled domains.
5. Conclusion: Approaches advocated in this study

Even this brief analysis has hopefully shown that a combination of
approaches in the study of metaphor and semantics can provide richer insights.
Future stages of this project will involve analysis of conversation-interviews with
American students, and ultimately the comparison between different types of
cultural models of honesty employed by Americans and Russians. The present
project has tried to focus attention on the following areas as a way of studying
metaphors as profiles against the bases of cultural models.
One area is the analysis of metaphorical expressions in context, especially
in one of the most prototypical contexts of language use, that of two people
talking - the conversational dyad. One topic to be explored further is how the
profile/base approach can be applied to the study of creative versus conventional
metaphoric expressions within a given body of discourse. A hypothesis to be
explored is that a more creative expression highlights a new framing of the topic
under discussion, thereby giving it new salience, and allowing it to serve as a new
profile against the base of what has already been discussed.
Another approach advocated here is to integrate the study of metaphorical
expressions from actual discourse data with what is being said in between the
metaphorical expressions to help reveal the cultural models of which they are a
part. In connection with this, it is also important to distinguish between what is
being said versus what is tacit, assumed cultural knowledge (the bases against
which semantic relations are being profiled). This assumed knowledge may be
made explicit if, for example, someone from outside the cultural group is present
(such as the American interviewer with the Russian students), and the speaker is
aware of the information gap between them (perhaps prompted by the inter
viewer' s questions). Future research in this project will give special consideration
to which cultural notions and practices are considered by interviewees to be more
versus less conventionally known, and how this factor relates to interviewees' use
of metaphorical expressions.
Finally, this study advocates further research on the relations between verbal
and non-verbal (such as gestural) expression, in part because of how gesture can
offer additional evidence of metaphorical thought. Future analysis of the
videotaped conversation-interviews will concern the extent to which gestural data
support the reality of the underlying conceptual metaphors proposed on the basis
of the verbal, spoken data.
Notes
I gratefully acknowledge the comments received from members of the audience when an
earlier version of this work was presented at the Fifth International Cognitive Linguistics
Conference in Amsterdam in July, 1997. I am also grateful to Gerard Steen, Ray Gibbs, and an
anonymous reviewer for their comments on an earlier draft. Address for correspondence: Dept.
of Russian and East Asian Languages and Cultures, Emory University, Atlanta, GA 30322, USA
(e-mail: lanac@emory.edu).
1. In the transliterated Russian words, the spelling ch represents a voiceless palatal affricate
(as in English spelling), they is a voiced palatal glide, and the apostrophe indicates that the
preceding consonant is palatalized.
2. This research was supported by a Short-Term Travel Grant from the International Research
and Exchanges Board (IREX).
3. Numbers in square brackets indicate the interview number; the letters A and B indicate the
two different speakers in each interview; metaphorical expressions are highlighted in bold
face; text in square brackets aligned vertically between two speakers indicates overlap;
slashes (/) indicate restarts by the speaker; given the length of the examples, translations,
rather than morpheme-by-morpheme glosses, have been provided.
4. Speech co-occuring with relevant gestures will be underlined, with curly brackets { }
placed around the speech that aligns with the main "stroke" portion of each gesture. Words
202 ALAN CIENKI
in parentheses, unnecessary in the context of the actual discourse, have been added to make
the relevant references clear.
5. The question mark in parentheses indicates an unintelligible word.
References
Basso, Keith H. 1976. '"Wise Words' of the Western Apache: Metaphor and
Semantic Theory". Meaning in Anthropology, ed. by K. Basso, & H. Selby,
93-121. Albuquerque: University of New Mexico Press.
Calbris, Genevive. 1990. The Semiotics of French Gestures. Bloomington, Ind:
Indiana Unuversity Press.
Cienki, Alan. 1998. "Metaphoric Gestures and Some of their Relations to Verbal
Metaphoric Expressions". Discourse and Cognition: Bridging the Gap, ed.
by Jean-Pierre Koenig, 189-204. Stanford, Calif.: CSLI.
1998. "STRAIGHT: An Image Schema and its Metaphorical Extensions".
Cognitive Linguistics 9. 107-149
Emanatian, Michele. 1996. "Conversational and Intellectual Paths". Paper
presented at the conference "Conceptual Structure, Discourse, and Language
n", Buffalo, N.Y.
Grady, Joe. 1998. "A Reassessment of Metaphors for Mental Processes".
Discourse and Cognition: Bridging the Gap, ed. by Jean-Pierre Koenig,
205-218. Stanford, Calif.: CSLI.
Ivanov, Vjacheslav, & Vladimir Toporov. 1965. Slavjanskie jazykovye
modelirujushchie semioticheskie sistemy [Semiotic Modeling Systems of the
Slavic Languages]. Moscow: Nauka.
Keesing, Roger. 1979. "Linguistic Knowledge and Cultural Knowledge: Some
Doubts and Speculations". American Anthropologist 81. 14-36.
Lakoff, George. 1996. Moral Politics: What Conservatives Know that Liberals
Don't. Chicago: University of Chicago Press.
, & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of
Chicago Press.
, & Zoltn Kvecses. 1987. "The Cognitive Model of Anger Inherent in
American English". Cultural Models in Language and Thought, ed. by
Dorothy Holland, & Naomi Quinn, 195-221. Cambridge: Cambridge
University Press.
Langacker, Ronald W. 1987. Foundations of Cognitive Grammar. Vol. 1:
Theoretical Prerequisites. Stanford, Calif.: Stanford University Press.
McNeill, David. 1992. Hand and Mind: What Gestures Reveal about Thought.
Palmer, Gary B. 1996. Toward a Theory of Cultural Linguistics. Austin, TX:
University of Texas Press.
Quinn, Naomi. 1987. "Convergent Evidence for a Cultural Model of American

Marriage". Cultural Models in Language and Thought, ed. by Holland, &
Quinn. 173-192. Cambridge: Cambridge University Press.
Reddy, Michael J. [1979]/1993. "The Conduit Metaphor: A Case of Frame
Conflict in our Language about Language". Metaphor and Thought, second
edition, ed. by Andrew Ortony, 164-201. Cambridge: Cambridge University
Press.
Shore, Bradd. 1996. Culture in Mind: Cognition, Culture, and the Problem of
Meaning. Oxford: Oxford University Press.
Sweetser, Eve. 1987a. "The Definition of lie: An Examination of the Folk Models
Underlying a Semantic Prototype". Cultural Models in Language and
Thought, ed. by Dorothy Holland, & Naomi Quinn, 43-66. Cambridge:
1987b. "Metaphorical Models of Thought and Speech: A Comparison of
Historical Directions and Metaphorical Mappings in the Two Domains".
Proceedings of the Thirteenth Annual Meeting of the Berkeley Linguistics
Society, ed. by J. Aske et al., 446-459. Berkeley, Calif.: Berkeley Linguistics
Society.
1990. From Etymology to Pragmatics: Metaphorical and Cultural Aspects
of Semantic Structure. Cambridge: Cambridge University Press.
1992. "English Metaphors for Language: Motivations, Conventions, and
Creativity". Poetics Today 13. 705-724.
Talmy, Leonard. 1983. "How Language Structures Space". Spatial Orientation:
Theory, Research, and Application, ed. by Herbert L. Pick, & Linda P.
Acredolo, 225-282. New York: Plenum Press.
1988. "Force Dynamics in Language and Cognition". Cognitive Science
12.49-100.
Van Leeuwen-Turnovcov, Jirina. 1991. "KRUM und DREHEN im Kultur
paradigma der ORDNUNG". Znakolog 3. 131-66.
1994. "TAUB, BLIND, STUMM und DUMM: Zur Semantik der Abweichung
vom Gewohnten". Zeitschrift fr Slawistik 39. 545-59.
Vanparys, Johan. 1995. "A Survey of Metalinguistic Metaphors". By Word of
Mouth: Metaphor, Metonymy and Linguistic Action in a Cognitive
Perspective, ed. by Louis Goossens et al. 1-34. Amsterdam/Philadelphia:
John Benjamins
CONGRUENCE BY DEGREE
ON THE RELATION BETWEEN METAPHOR & CULTURAL MODELS
MICHELE EMANATIAN
1. Introduction
This paper explores the plausible idea that the conceptual linkings in
conventional metaphors should occur elsewhere in the conventional imaginative
repertoire.1 If systematic, everyday metaphor provides us with a way to think about
something through mapping it to another conceptual domain, wouldn't it make
sense for the same mapping to be found in our extra-linguistic symbolic lives?
Claims in the literature notwithstanding (Lakoff 1987; Quinn 1991;
Kvecses 1990), I believe that we really do not yet know much about how
metaphors which serve as schemas fit with non-metaphorical cultural models.
Conclusions are, of course, dependent on definitions of cultural models, on where
one model is felt to leave off and the other begin.
We know that there may be both linguistic (metaphorical, metonymic,
grammatical) and non-linguistic aspects to the shared schemas of a culture, those
received yet emergent and mutable patterns we may absorb, use, change, reject,
and pass on. In such cases, do we count the linguistic aspects as a separate cultural
model, and the other aspects (for example, action sets, kinesthetic schemas, visual
image models, etc. - Shore 1996) as models in their own right? Or does it make
more sense to speak of a single cultural model, which is manifest in a variety of
different 'modalities'? As always, how we slice things depends on our purposes.
In this paper I will sometimes speak as if metaphor (the type of linguistic model
I am interested in here) is part of a 'larger' cultural model, and at other times
speak as if it constitutes a model separate from other models for the same thing.
What matters for present purposes is the extent to which the mappings found in
metaphor resemble, echo, or differ from the mappings found elsewhere.
My goal is to argue that the relation of metaphor to cultural models varies,
and that no simple statement about its priority, centrality, or epiphenomenality
will suffice. To this end, I would like to draw attention to one of the relationships
between a metaphorical model and non-metaphorical cultural models, that of
congruence. The cross-domain mapping of a conceptual metaphor and the cross-
domain mapping that might be observable in other symbolic resources available
206 MICHELE EMANATIAN
to members of a culture may be congruous in that they may exhibit parallelism,

harmony, or appropriateness - a correspondingness in character.
Congruence across mappings will be illustrated through the specific example
of lust and sex in Chagga, a Bantu language of Tanzania. Chagga metaphors for
lust and sex will be shown to be congruous with non-linguistic symbols and rituals
of sexual coming of age. The expanses of congruence, however, mask points of
incongruence: there is some measure of discord if the fine details of mappings are
taken into account.
2. Metaphors of sex & lust in the Chagga language

I would now like to illustrate the complexity that I feel is common in
congruent relationships between metaphorical models and non-metaphorical,
especially non-linguistic, models for the same target domain or concept.
Before we examine the systematic conceptual metaphors for lust and sex
that are used by Chagga speakers, the limitations of this study must be made
explicit. The data and native speaker commentary were elicited from a single male
speaker outside the Chagga speech community, over the course of several years.
They are therefore both sex-skewed and de-contextualized. I can make no
observations about the kinds of situations which provoke these expressions, about
the use or appropriateness in encounters, nor about expected responses to them.
I cannot vouch for the representativeness among Chagga men of this speaker's
means of expressing his feelings of lust and about sex. I do, however, believe that
the body of metaphors analyzed here constitute a conventional Chagga perspec-
tive on sex and lust, while raising interesting questions for further research.
On first glance, the similarity with English is dramatic. There is a very small
number of source domains, and these English has too: heat, eating, and to a lesser
extent, animals. These are systematically employed in expressing lustful feeling
and in articulating sentiments about sexual matters, such as assessments of the
qualities of potential partners. As in English, inferences may be drawn about the
target concepts (e.g. a sexual partner) on the basis of the understanding of the
source concepts (e.g. a type of food).
We can begin with the domain of eating. In Chagga copulation is metaphori
cal eating. For a man to say ngikundiimlya, as in (1), is to express the desire to
copulate with a woman.
(1) ng kundmlya [I want to eat her]

= "I want to have intercourse with her."2
Men feel hungry, that is, lustful; they look for snacks; enjoy feasting; salivate at
the thought of good food; and feel satiated - all in the sexual domain. They crave
CONGRUENCE BY DEGREE 207
tasty, or sweet, partners, and if they find one, may never have to eat again.
Consider (2) through (12).
(2) ngi'ichuo nja (ia mndu mka) [I feel hunger (for a woman)]
= "I'm desirous (of a woman)."
(3) ng ndpfl wundo w lyo [I'm going to look for a little something to eat]
= "I'm going to find a sexual partner."
(4) ng km'meTia ma 0ta [I swallow saliva for/because of her]

= "I really desire her (as a sex partner)."
(5) kengonja h, me [Give me just a taste (of) there, woman]

= "Let me try you out, sexually."
(6) ngm 'soki ch kndo [I'll descend on her like a feast]

= "I'll have sex with her enthusiastically and exhaustively."
(7) nkewtsu [She satisfies]

= "She's sexually satisfying to me/men."
(8) ng 'm 'ly ngchk ng sepfo [If I eat her, I won't have to eat again]
= "If I have sex with her, I'll be satisfied forever."
(9) ng m 'lya ngchpf pfo [If I eat her, I' 11 live forever]
= "If I have sex with her, I'll be satisfied forever."
(10) kelo ta ? [Does she taste good/sweet?]

= "Is she a good sexual partner?"
(11) nkelota ch 'skri 'ki [She tastes sweet/good as sugar honey ]

= "She's a good sexual partner."
(12) ngalo tio [I was "tasted good", i.e., I experienced sweet taste]
= "I've been exquisitely satisfied, sexually."
Utterances like (8) and (9) show that it is possible to use the source domain to
reason about the target domain. The structural relationship, of ingesting food and
thereby feeling satiated, is maintained in the target domain of sex.
The second conceptual metaphor for the sexual sphere in Chagga is a heat
metaphor. Desirable sexual attributes are correlated with heat, so that a warm
woman, one who roasts or burns, as in (13)-(15), or who has a "heaven of fire",
as in (16), is highly desirable.
(13) nkekya [She roasts]

(14) nkeh [She burns]
(15) nwo m'rike [She has warmth]
= "She is sexually desirable."
(16) nwo 'shngu lo mo [She has a "heaven" of fire]

= "She has desirable sexual attributes (skills, natural endowments, interests)."
(Ushangu 'heaven' is colloquial and euphemistic for 'vagina'; it is a shortened

form of the phrase ushanguni lo ruwa 'stand before god', lit., 'in the face of god'.
ruwa 'god' also means 'sun'.)
Sexy women are metaphorical heat-producers, such as ovens or fires (17),
which give off smoke (18). Cold women are not of interest (19).
(17) kyambuya rik lilya [Look at that oven]

= "Look at that sexy woman."
(18) ngi wni m'tsu [I see smoke]

= "I can tell she's sexually exciting."
(19) nkechlli [She's cold]

= "She lacks desirable sexual attributes."
It should be noted that examples thus far are from a man's point of view; this is
an artefact of fieldwork which I will return to below.
The third source domain for sexual matters in Chagga is animals. Certain
kinds of animals map metaphorically to potential partners with a variety of sexual
attributes. See examples (20)-(23).
(20) ni kite [She is a dog]

= "She is promiscuous."
(21) kapf 'iko ? [How are you, my bull?]

= "How are you, my stud?"
(22) kiambuya ly [Look at that rooster]

= "Look at that sexy young guy."
(23) ap 'taw ngileyetsi [Wow, a fattened heifer]

= "Wow, a sexy young woman!"
For instance, framed sexually a woman may be a dog, meaning she's promiscuous ;
or a man may be a bull, meaning he's a 'substantial' male lover. A rooster is a
young, good-looking man, while a fattened heifer is a nubile young woman.
Sexual partners are linked to animals in these and other expressions, via certain
attributes, which are themselves not mentioned. Animal behaviors, such as
stalking prey, may also be ascribed to people, once they are framed sexually, as
in (24).
(24) ni k k owlia ih ? [What are you stalking there?]

= "Who are you pursuing sexually?"
Note the coherence between the heat metaphor and the eating metaphor. In each,
the sexual attributes of a woman are conceptualized as perceptible features: heat
and taste. Cooked food is, of course, hot. There is also some coherence between
the animal metaphors and the eating metaphor. Stalking is hunting behavior,
engaged in by both humans and some kinds of animals to obtain sustenance. In
both metaphors the desired person is framed as ultimately being consumed as
food.
A comparison with the metaphorical 'structuring' of the sexual domain in
as many languages as data is readily available for, indicates that the domains of
eating and heat, at least, are cross-culturally common as metaphorical sources for
sexual feeling, behavior, and attributes. It is not common to find sexual feelings,
attributes, or behaviors metaphorically conceived as activities like farming or
grooming, or as qualities like wetness or bendability (Emanatian 1995, 1996).
Observations about pan-cultural aspects of metaphorization are important and
provocative in this era in which cultural relativism has attained such prominence.
Yet, as has been argued elsewhere, it is equally important to note that there is
conventionality here; that is, there is cultural specificity to the metaphor: in how
the domains are framed; in their productivity; in the fine details.
3. Metaphorical correspondences outside language?

Returning now to questions of the relationship between metaphor and wider
cultural models, it must be acknowledged that in Chagga conventional metaphors
do not exhaust the notions of sexual attraction, sexual desirability, lust, the
goals/purposes/functions of sex, etc. available to members of the culture. Other
dimensions of culturally grounded understandings of this domain might also
include (at least) aesthetic values; notions of responsibility; of social appropriate
ness and normalcy (e.g. whether 'rape' is accepted or not; what actions count as
'rape'; whether multiple partners are sanctioned; etc.); gender distinctions and
roles; where sexual power resides; cosmological understandings about the
regeneration of life; beliefs about sexual maturation and what being an adult is;
the symbolization and enactment of these in artefact and ritual; the
institutionalization of some of these ideas in child-rearing practices, law, and
religion; and so on. (One culture may or may not give salience to dimensions of
a cultural model that are quite prominent in another; for example, conventional
visual images are very important in American cultural models of sex, but this does
not seem to be the case in Chagga.)
The degree of harmoniousness between the metaphors and non-linguistic
aspects of the Chagga 'take' on things sexual (the latter as recorded and
interpreted in Moore's 1977 ethnography) is quite striking. In the symbolism of
ritual events and artifacts there are correspondences between heat, eating, and
animals, on the one hand, and sex, on the other hand. These correspondences are
outside language. These specific linkings of sex with non-sex are widely
established in Chagga culture.
Moore describes cosmological correspondences between life and death,
between food and sex: "there was a symbolic and ritual preoccupation with food
and sex, with eating and fertility as the basic means of preventing death, both
immediate and eternal, and perpetuating life" (1977:47). The body was a
container, in both feeding and reproduction. Feeding in the mouth maintained life,
"feeding" in the vagina in intercourse produced new life. Aspects of traditional
daily life reflected these particular culturally significant correspondences, and to
some extent this is still the case. Bananas, a staple food on Kilimanjaro, are
proscribed in raw form for adults. Married women and men eat only bananas
which have been peeled and cooked, or circumcised and having been sexually "in
the fire". Children eat only raw or unpeeled roasted bananas. The "separation of
sexual life of the generations was extended to food" (Moore 1977:51). Pregnant
women were not supposed to eat bananas, nor drink beer, as these were and are
associated with men, symbolically and economically.
In ritual we find conventional symbolic representations of the order of the
universe. Part of the instruction to Chagga boys undergoing the circumcision and
initiation ritual was a likening of the fetus to a trapped or arrow-pierced animal
struggling to get free. Intercourse was likened to hunting (and thus the achieve
ment of procreative manhood). Initiate boys were allowed to "hunt" birds, plants,
and gentle mammals (such as gazelles) with bows and arrows, referring to their
prey as "elephant", "buffalo", "leopard", or "lion" (Moore 1977:59). After
initiation there was a licentious period, in which boys were allowed to roam in
groups and rape any woman they chose (1977:60).
In the female circumcision and initiation ritual, small animals, such as
grasshoppers and tadpoles, were captured. "Hunting had the implication of
seeking out progeny, a metaphor for procreation".3 The ritual also included
instruction to the girls in which a firebrand stood for the penis, ashes for the vulva,
and embers for the seeds in a woman's body out of which children grow (Moore
1977:62). Freshly initiated girls, when betrothed, were traditionally sequestered
for three months, during which time they were "lavishly fed" (1977:61, 63). Both
women and cows were and are considered a man's property.
In ritual symbolism, then, there are mappings between sex and food,
between the search for a mate and hunting, and between sex and heat or fire. The
mappings are parallel to the metaphorical correspondences found within
contemporary colloquial Chagga usage. As has been noted by others, this sort of
congruence contributes to an outsider's sense of how a culture works (Strauss &
Quinn 1994; Shore 1996).
Yet, a closer look at these symbolic associations reveals a measure of
discord: the congruence between linguistic/conceptual mappings and extra-
linguistic/conceptual mappings is a matter of degree.
First, the symbolism in the boys' rituals is somewhat different in fine detail
from the symbolism in the girls' rituals. The boys hunt for sexual partners, akin
to talk of stalking and the metaphorical search for something to eat. But the girls
are hunting for progeny - in contrast to the metaphors described to this point.
Cosmology shows a similar difference of perspective along gender lines:
according to Moore, "feeding" in the vagina produces new life. Yet in the
metaphors we have examined, it is the man who is fed in intercourse.
Interestingly, the male informant on which the metaphor study is based
offered two expressions used by Chagga women which shed some light on this
seeming incongruence. Consider (25) and (26).
(25) ngimangari [I'm thirsty]

= "I want to have more intercourse."
(26) napf l m Yuwa [She's searching for milk]

= "She's 'in heat'; she's desirous of sex."
A woman may express a desire for sexual intercourse by saying she's 'thirsty' ; this
functions as a plea that intercourse not be interrupted (as it might be to avoid
conception). A woman may also 'search for milk', as does a female animal in heat.
It is relevant that semen in Chagga is 'male milk'. Male milk and female blood
combine in the child (Moore 1977:62). (Even today new mothers are given a
potent mixture of cow's blood and butter, called mlaso.) There is reason to believe
that this thirst metaphor is an expression not of feminine lust, but rather of the
desire to conceive a child. The status of a woman in Chagga society increases as
she bears children (particularly if they are male). The existence of these
212 MICHELE EMANATTAN
expressions alongside the progeny-oriented initiation rituals suggests an

interesting difference of perspective among Chagga women and men, and may not
in fact be incongruence (of the sort we have been discussing) at all.
A comparison of the symbolic associations for heat with the metaphorical
meaning of heat in the sexual domain provides a second instance of partial
incongruence. The avoidance by unmarried men and women of roasted bananas -
bananas that have been "in the fire" - accords with the contact of male with hot
female in the metaphors. In the girls' initiation ritual, however, it is the penis
which is the hot object (a firebrand), while the vulva are only ashes. The 'product'
of conception is figured here too, as embers. Again, it is conspicuous that
conception is not part of the system of linguistic-conceptual metaphors collected
from a male consultant.
It seems, then, that the striking parallelism between conceptual metaphors
for lust/sex and extra-linguistic symbolism in Chagga is in actuality threaded with
incongruity. To my mind, though, the existence of disparity does not erase the
basic congruity within these different parts of a cultural model. We might want to
say instead, for instance, that the conception-oriented metaphorical and symbolic
correspondences are part of a different cultural model, one more available and
useful to females, and created by females, in this highly gender-differentiated
society. The sexual domain is one area where there are clear differences of
perspective according to gender; where there are different construals of the whole
ensemble of experiences (and their offshoot concepts) of life-desire-sex-
procreation-death. Given Chagga social structure, we might expect there to be two
different cultural models and two different systems of conceptual metaphor. A
complementary study of situated use of metaphor, and in particular, of women's
ways of talking about sexual matters, may well undo apparent instances of
incongruence between linguistic and non-linguistic mappings. In any case it
remains remarkable that the same non-sexual domains are drawn from as sources
for both symbol and metaphor. This larger scale resonance does not disappear
upon the discovery of smaller scale incongruence, such as we have for the
metaphor and symbol of heat.
4. Discussion
The Chagga example shows a high but not total degree of congruence
between metaphor and non-linguistic symbolism; it involves complex cultural
models in which metaphor is only one 'constituent' or dimension. In looking more
generally at the issue of the relatedness of metaphorical and non-metaphorical
models, the notion of a scale of congruence will be useful to us.
At one end of the scale, we find congruence across many dimensions of a
cultural model, metaphorical and otherwise. If a model is general enough, it may
have broad applicability across many target domains, cross-cutting other cultural
models. Such is the case with "foundational schmas". An example is the
modularity schema, discussed by Shore (1996) and Martin (who calls it "flexible
specificity", 1994). In the contemporary U.S., the modularity schema is manifested
as a design strategy underlying real physical structures like furniture, shopping
malls, and workbooks; more abstract realities, like work or the immune system;
and highly abstract concepts that are largely our creations, such as "lifestyle", or
knowledge itself. A conceptual metaphor, Adaptable Is Flexible, is one of the
many tentacles of the modular foundational schema (Emanatian, in prep.).
Contemporary flexibility - metaphorical and otherwise - is often understood as
modular, with salient part-whole structure and notions of fit and interchangeabi-
lity. Metaphorical objects, such as a body of law, a child care policy, or a personal
identity, may be modularly flexible, h flexible investment strategy, for example,
achieves its flexibility (bendability) - its adaptability - via its modular structure,
its ever-shiftable and largely interchangeable parts. The wide ranging application
of the modularity schema in American culture is paralleled by the increasingly
common association between the modular brand of metaphorical flexibility and
the highly desirable American trait of adaptability. That is, the broadly relevant
cultural model of modularity is echoed in a prevalent linguistic-conceptual
metaphor. High degrees of congruence would seem to obtain between metaphor
and other aspects of cultural models in the case of foundational schmas. While
other instances of foundational schmas have been mentioned in the literature
(Indian purity and pollution, American self-reliance - Strauss & Quinn 1994;
Samoan mana - Shore 1989), the relation between conceptual metaphor and these
'larger' cultural models has yet to be systematically studied.
At the other end of the congruence scale, we find co-existence of contradic
tory cultural schmas, that is, incongruence. Although such co-existence may be
peaceful (Kay 1987; Turner & Fauconnier 1995), recent research has highlighted
the "collision" of contradictory models (Quinn 1996; Shore 1996; Strauss &
Quinn 1994), raising the questions of how, in a given context, we select one model
from among a set, and of how we deal with conflict when more than one is
evoked.
Similarly, within linguistic studies of metaphor, attention has been drawn
to varying degrees of coherence among metaphors for the same concept (within
a culture). (I use "congruence", rather than extending the term "coherence"
(Lakoff & Johnson 1980), "co-alignment" or "co-orientation" (Sweetser 1993),
to give prominence to relations of harmoniousness or correspondingness, and to
de-focus the notion of a logical or orderly relationship that "coherence" can
connote.) As Johnson points out, for example, "... our most basic concepts, such
as law, freedom, and rights are defined by multiple, often inconsistent conceptual
metaphors..." ( 1995:162). Conflicting entailments of related metaphors have been

drawn out, for example, in considering the war metaphor for argument and a
hypothetical dance metaphor (Lakoff & Johnson 1980), or the different metaphors
for thought within analytic philosophy (Lakoff & Johnson 1999). (See also Lakoff
1996a, 1996b; and Pesmen 1991 on prohibitions against mixed metaphor.) As
metaphor analyses become more socialized - in the sense of being broadened to
encompass the fuller cultural models the metaphors are part of - the issues of
coherence, selection, and conflict resolution are likely to become more salient and
more interesting.
On the scale of congruence, then, we have at one extreme high congruence,
as in the case of a foundational schema. At the other, we have low congruence, as
in the case of colliding cultural models for 'the same thing'. In-between, I am
claiming, lie the perhaps more abundant instances of partial congruence, such as
we have seen illustrated by the domain of lust and sex in Chagga. Of course, field
research into ways women talk about their sexual experience might reveal a
metaphoric system in perfect congruence with the non-linguistic models
underlying girls' initiation practice and beliefs about conception; then again, it
might not. My own cautious inclination would be not to assume congruence until
'the facts' are in.
We might wonder whether there might ever be a relationship of non-
congruence between metaphor and cultural model, that is, not where the mappings
are in conflict, but rather where they miss each other entirely. Non-congruence
would seem to be an accurate description for the two limiting cases: either where
there is little or no metaphor to a cultural model (consider the American schema
of going to a laundromat), or where a metaphor comes close to exhausting a
cultural model. With highly abstract notions, such as that of logical or coherent
thought or discourse, there appears to be little to our cultural understanding that
is not metaphorical. For example, it is difficult to define coherent thought without
using path and journey metaphors, such as direct, straightforward, ox follows
(Emanatian 1996). As Grady (1995) puts it,
[C]ertain metaphors are so deeply entrenched in our thought processes that

... they can hardly be the basis for expressing novel conceptualizations ...
[but rather are] part of our conceptual architecture. It is probable that many
of these 'invisible' metaphors are based on generic-is-specific type
mappings from physical experience to more abstract concepts, which may
actually have little if any structure in the absence of the metaphors, (p. 15;
italics mine)
In either of these limiting cases, then, a relationship of non-congruence seems to

obtain between metaphor and the larger schematic patterns of understanding
provided by a culture.
Again, the significance of the scale of congruence is that the one thing we
know (to my mind) about the relation between metaphorical cultural models and
non-metaphorical cultural models is that it is highly variable. Claims that
metaphorical understandings are primary or central - or epiphenomenal - are at
best premature.
5. Concluding remarks
I have tried to argue in this paper that the relation of conventional and
systematic conceptual metaphor to non-linguistic cultural models is not simple,
nor can it be specified in advance of looking at the details of particular cases. To
this end we have examined rather closely an instance of partial congruence across
different aspects of conventional conceptualizations, metaphorical and otherwise.
Partial congruence across modalities should not be surprising to metaphor
analysts, since it is paralleled by the partial coherence found across the metaphors
themselves. In the case of Chagga metaphors of lust and sex, the fact of partial
congruence with ritual and symbolism is mirrored by the degrees of coherence in
the metaphors. Points of coherence across the eating, heat and animal metaphors
have already been mentioned, but there is incoherence as well. For instance, while
a woman's desirable sexual attributes are mapped to pleasurable sensory
experience in both the eating and heat metaphors (sweet tastiness and warmth), the
basic ontologies of these metaphors are partly at odds: in the one case the woman
is an oven that can roast food, while in the other she herself is food. Another
example is the coherence-with-incoherence of the animal metaphors and the Sex
Is Eating metaphor, mentioned earlier: both animals and humans are known to
hunt for other animals, in order to consume them for food. Yet not all the human-
animal correspondences pattern like this; for example, neither bulls nor men hunt
cows. Also, some of the animals that appear in the metaphors are neither hunted
nor consumed for food in real life (e.g. dogs).
I have speculated about the extent to which 'cross-modal' congruence is a
ubiquitous phenomenon. In close studies of how spontaneous gesture is related to
the speech it accompanies, McNeill (1992) and others have found that metaphori
cal gestures often express apart of the speaker's overall conception which is not
being conveyed in the accompanying speech. There is a relationship of
complementarity between them. The Chagga example discussed here shows a high
but not total degree of congruence between metaphor and non-linguistic
symbolism; it involves complex cultural models in which metaphor is only one
'constituent' or dimension. Other, limiting cases of overlap between metaphor and
cultural models suggest that we cannot generalize about how exhaustive a role
metaphor plays in the conventional schematization of concepts.
The somewhat muddier analyses resulting from investigating congruence in
metaphor and non-metaphorical models can nevertheless be of value. For
example, the Chagga women's thirst metaphor may be put into perspective and
made more intelligible to us through the examination of traditional female
initiation practices. The partial congruence found across metaphorical vs. non-
linguistic conceptualizations of sexual activity in Chagga may spur us to explore
the women's side of the story. As Corradi Fiumara has said, "To make explicit the
ramifications of our dominant metaphors is to engage in a practice which brings
us inexorably close to our inner life and which thus enhances unforeseen shifts in
our axes of culture. The general picture changes: from clusters of changeless
talking heads to a historical community of living creatures" (1995:85).
Acknowledgments
I appreciate helpful comments from the editors and two anonymous reviewers. Special
thanks go to David Delaney and Alan Cienki, for many productive exchanges; to Alan again, for
inviting me to the theme session, "On the Place of Metaphors in Cognitive and Cultural Models";
and to Joe Grady, for graciously delivering my paper at the conference in Amsterdam.
Notes
1. See Johnson (1987) and Shore (1996) on how elements of thought and culture can be both
conventional and imaginative.
2. Chagga (Chaga, KiChagga) is spoken by about 70,000 people, on Mt. Kilimanjaro in
Tanzania. I describe the KiVunjo dialect, of Central Kilimanjaro.
For the examples I have used an orthography common in Africanist linguistics. Exceptions
include: T for a retroflex flap, #t for a slightly fricated alveolar approximant, and r for an
alveolar trill. High tone ' , falling tone , and downstep ' are marked; low tone is left
unmarked.
Following the Chagga is a literal translation in brackets, while an approximation to the
conveyed meaning appears below in double quotes. These latter glosses are often
infelicitous or awkward, but a free translation into English is avoided because these
sometimes employ a different metaphor, making for confusion.
3. Note that Moore's use of "metaphor" is broader than mine. I prefer to keep metaphor and
symbolism distinct, in part because not all symbols are cross-domain mappings. Some may
be metonymically based (see Shore 1996), as for example, the use of a lion's mane to stand
for bravery (thought also to be a lion's trait).
References
Corradi Fiumara, Gemma. 1995. The Metaphoric Process: Connections between
Language and Life. New York: Routledge.
Emanatian, Michle. 1995. "Metaphor and the Expression of Emotion: The Value
of Cross-Cultural Perspectives". Metaphor and Symbolic Activity 10. 163-
182.
1996. "Everyday Metaphors of Lust and Sex in Chagga". Ethos 24. 1-42.
1996. "Intellectual and Conversational Paths". Conference presentation,
CSDL n, Buffalo, N.Y., April 1996.
I n preparation. Why We All Want Flexibility: A Study of a Rhizomic
Metaphor.
Grady, Joe. 1995. Special Field Exam, UC Berkeley Linguistics Dept.
Johnson, Mark. 1987. The Body in the Mind: The Bodily Basis of Meaning,
Reason and Imagination. Chicago: University of Chicago Press.
1995. "Introduction: Why Metaphor Matters to Philosophy". Metaphor
and Symbolic Activity 10. 157-162.
Kay, Paul. 1987. "Linguistic Competence and Folk Theories of Language. Two
English Hedges". Cultural Models in Language and Thought, ed.by D.
Holland, & N. Quinn, 67-77. Cambridge: Cambridge University
Kvecses, Zoltan. 1990. Emotion Concepts. New York: Springer-Verlag.
Reveal about the Mind. Chicago: University of Chicago Press.
1996a. Moral Politics: What Conservatives Know that Liberals Don't.
1996b. "Sorry, I'm Not My self Today: The Metaphor System for Conceptu
alizing the Self'. Spaces, Worlds, and Grammar, ed. by G. Fauconnier, &
E. Sweetser, 91-123. Chicago: University of Chicago Press.
s & Mark Johnson. 1980. Metaphors We Live By. Chicago: University of
Chicago Press.
, & Mark Johnson. 1999. Philosophy in the Flesh: The Embodied Mind and
Its Challenge to Western Thought. New York: Basic Books.
Martin, Emily. 1994. Flexible Bodies. Tracking Immunity in American Culture -
from the Days of Polio to the Age of AIDS. Boston: Beacon Press.
McNeill, David. 1992. Hand and Mind. Chicago: University of Chicago Press.
Moore, Sally Falk. 1977. The Chagga of Kilimanjaro. Pt.l of The Chagga and
Meru of Tanzania, by S. Falk Moore, & P. Puritt. Ethnographic Survey of
Africa, East Central Africa 18, ed. by W.M. O'Barr. London: International
Africa Institute.
Pesmen, Dale. 1991. "Reasonable and Unreasonable Worlds; Some Expectations
of Coherence in Culture Implied by the Prohibition of Mixed Metaphor".
Beyond Metaphor: The Theory of Tropes in Anthropology, ed. by J.W.

Fernandez, 213-243. Stanford: Stanford University
Quinn, Naomi. 1991. "The Cultural Basis of Metaphor". Beyond Metaphor: The
Theory of Tropes in Anthropology, ed. by J.W. Fernandez, 56-93. Stanford:
Stanford University
1996. "Culture and Contradiction: The Case of Americans Reasoning
About Marriage". Ethos 24. 391-425.
Shore, Bradd. 1989. "Mana and tapu": Developments in Polynesian Ethnology,
ed. by A. Howard, & R. Borofsky, 137-173. Honolulu: University of
Hawaii.
1996. Culture in Mind: Cognition, Culture, and the Problem of Meaning.
New York: Oxford University
Strauss, Claudia, & Naomi Quinn. 1994. "A Cognitive/Cultural Anthropology".
Assessing Cultural Anthropology, ed. by R. Borofsky, 284-297. New York:
McGraw-Hill.
Sweetser, Eve. 1993. "'Inside/Outside, Upside Down' - Co-Orientation of
Metaphorical Mappings in Literary and Everyday Language". Ms. Linguis
tics Dept., UC Berkeley.
Turner, Mark, & Gilles Fauconnier. 1995. "Conceptual Integration and Formal
Expression". Metaphor and Symbolic Activity 10. 183-204.
SUBJECT INDEX
A conceptual integration, 5, 105, 106, 111,

absolute metaphors, 15 113, 115, 118
abstract experiential domains, 47 conceptual metaphor, 7, 1-6, 13, 18, 20-22,
abstract metaphorical meanings, 34 57-61, 66, 68, 73, 79, 81, 86, 89, 90, 97,
analogical reasoning, 72, 76, 77 101, 102, 110, 112, 120, 121, 123,
analogy, 10, 12-14, 22, 43, 66-74, 76, 97, 145-148, 150, 151, 153, 158, 159, 161,
111, 112, 114, 120, 185 162, 174, 183, 205, 207, 212, 213, 215
conceptual metaphor theory, 58, 66, 89, 90,
B 101, 102, 110, 120, 121
base, 2, 6, 191, 198,200,201 conceptual representations, 147
blended space, 103, 106, 107, 114, 115, CONDUIT metaphor, 99, 203
119,120 conflicting entailments, 214
blending theory, 72, 101,102,110,120,121 congruence, 8, 6, 205, 206, 211-216
bodily experience, 34, 47-49, 55, 148, 154 conventional language, 146, 151
bodily orientation, 14 conventional mappings, 91, 118
body-world interactions, 156 conventional metaphors, 80, 84, 98,
109-112,205,209
C conventional symbolic representations, 210
Chagga, 206-212, 214-217 conventionality, 44, 48, 80, 96, 209
cognitive grammar, 95, 189-191, 202 Conyers, 126-128, 130, 133, 136, 138-141,
cognitive linguistics, 5, 1,2,4-7, 10, 25, 26, 144
42-45, 55-59,65,99,101,123,124,128, correlation metaphors, 89, 90, 93, 95-97,
140, 145-147, 149, 153, 154, 156, 161, 122
163, 164, 187, 190,201,202 counterdetermination, 21
cognitive models, 15, 19, 121, 128, 140, counting metaphors, 7, 47
146, 153, 154, 162, 190 cultural model, 7, 115, 127, 140, 143, 153,
cognitive semantics, 22, 29, 34, 47, 55 154, 167, 169-173, 185, 189, 191, 199,
cognitive topology, 38, 149 200, 203, 205, 210, 212-214
cognitive webs, 146
coherence, 63, 157,163,200,209,213-215, D
217 directionality, 17, 20, 95, 98, 110, 117,118,
comparison, 10, 19, 59, 65-74, 85, 95, 117, 120
134, 199, 200, 203, 209, 212 discourse, 7, 6, 11, 25, 48, 49, 53, 55,
comparison statement, 68, 69, 71 57-66, 68, 71, 73, 75, 76, 99, 108,
complex metaphorical blends, 112 123-125, 129, 135, 136, 139, 140, 142,
complex systems, 169, 170, 174-176, 179, 144, 150, 151, 158, 195, 199-202, 214
180 domain hypothesis, 18
componential analysis, 35
comprehension, 3, 4, 38, 64, 66, 75, 76, E
146, 162, 175 eating metaphor, 209, 215
conceptual blending, 5, 6,72,101,107, 111, elicited vs. non-elicited accounts, 134-136
120-123 embodied experiences, 148, 150, 154, 155
220 SUBJECT INDEX
embodiment, 6, 29, 34, 153-156, 161, localist theory of grammar, 10

163-166 long-term memory, 146, 152, 156
emergent structure, 103, 107, 122
empirical research, 151, 161 M
entrenchment, 106 mental models, 55, 164, 190
Evangelical Christians, 129 mental representations, 87, 101, 147, 152,
experiential motivation, 79, 81, 84, 97, 180 153, 156, 157, 190
experimental research, 50, 51, 55, 72 mental spaces, 102, 107, 111, 123
external representations, 152, 156, 157 metaphor analysis, 6, 59, 64, 65
metaphor focus identification, 60, 61, 65,
F 68,73
face-to-face narratives, 133, 134, 137 metaphoric blends, 113, 114, 117, 122
foundational schema, 197, 213, 214 metaphoric(al) gestures, 195, 215
framing, 109, 115, 117,201 metaphorical emergence, 168, 174, 180,
181, 186,187
G metaphorical extensions, 39, 41
gender, 210-212 metaphorical ICMs, 19, 21
generic space, 103, 104 metaphorical idea identification, 62, 65, 67,
68,73
H metaphorical scope, 34, 41
heat metaphor, 207, 209 metaphorology, 15-18
metonymie tightening, 108, 113
I metonymy, 44, 85, 99, 146, 157, 187, 203
idealized native speaker, 3 moral categories, 191, 196, 200
idiomaticity, 148
idioms, 2, 44, 145, 148-151, 155, 159, 161, N
164 narrative, 124, 126,129,135,137,144,159,
image donor, 18, 22 198
image fields, 18-21 non-biological metaphors, 133, 134
image recipient, 18, 22 non-visual metaphors, 125, 131, 133-136,
image schemas, 47, 48, 124, 148, 151, 152, 139
154, 163 nonliteral analogy identification, 67-69, 73
image-schematic structure, 148 nonliteral comparison identification, 66-68,
implicit metaphors, 61, 64, 65 73
incongruence, 206, 211-213 nonliteral mapping, 66, 67, 71-73
individualistic view of cognition, 151 nonliteral mapping identification, 71, 73
invariance principle, 38 novel conceptualizations, 101, 214
L O
language and agency, 128 off-loading, 160
linguistic metaphor, 1, 2, 4, 6, 9, 16, 21, on-line narratives, 133-135, 137, 138
57-61,65 onomasiological approach, 22
linguistic reality, 20 ontology, 96
linguistic/conceptual mappings, 211 optimality principles, 107
literal, 3, 6, 18, 21, 44, 59, 61-66, 69-71, 73,
89, 108, 123, 149, 150, 167-173, 179, P
181,186,216 perception verbs, 29, 42, 43
literal emergence, 168-173, 181, 186 perceptual metaphors, 7, 125, 131-140
SUBJECT INDEX 221
perceptual vocabulary, 133, 134, 136-138, S
141 scale of congruence, 213-215
pilgrim, 139 secular experience, 133, 136, 138, 139
poetic metaphor, 44, 76, 99, 165 semantic change, 29, 43, 165
polysemy, 2, 29, 43 semantic extensions, 34, 38, 41-43
primary metaphors, 82, 84-86, 90, 93, semasiological approach, 22
95-99, 122, 123, 180 semiotic modeling system, 197
production, 3, 4, 38, 72 sensory verbs, 132
profile, 6, 149, 191, 198, 200, 201 sexual intercourse, 211
property selection processes, 37, 39 similarity, 12, 13, 66, 67, 76, 77, 88, 89, 96,
propositional analysis, 61-65, 67, 75 97, 111, 112, 120,206
propositional metaphor analysis, 64, 65 similarity theory, 89, 97
propositions, 62, 66-69, 76, 134, 142, 152 situational cognition, 152
proverbs, 91, 145 social/cultural artifacts, 162
psycholinguistic evidence, 145, 151 symbolism, 210-212, 215, 216
psycholinguistics, 3, 59, 75, 145, 148, 151
psychology, 43, 45, 55, 63, 133, 145-148, T
152, 154, 161, 163, 165, 191 target concepts, 82, 83, 86, 96,97,118,122,
public representations, 157, 160, 162 206
target domains, 5, 17, 34, 40,102,131,148,
R 149, 180,213
religious narratives, 127, 130, 133, 134, tenor, 61, 62
136-139 therapeutic discourse, 158
representation, 3, 4,14, 66, 90-93,102, 111, topic, 42, 62, 71, 73, 86, 117, 200, 201
113, 147, 152, 159, 164
resemblance hypothesis, 97 U
resemblance metaphors, 87, 93-97, 122 unidirectionality, 9, 13, 17, 20, 95
ritual, 126, 129, 142-144, 210-212, 215
V
visual metaphors, 125, 131, 133-136, 139
NAME INDEX
A Clark, A., 163

Ahlstrom, R., 36, 42 Classen, C, 36, 42
Allbritton, D., 145, 162 Clements, C, 40, 43
Alverson, H., 159, 163 Cohen, L., 58, 75
Anderson, J., 10, 24 Colby, B., 157, 163
Arendt, H., 10, 24 Coleman, L., 129, 139, 142
Colston, H., 5, 148, 149, 163
B Comrie,B.,45, 129, 142
Badia, P., 35, 42 Corradi Fiumara, G., 216, 217
Balaban, V., vii, 6, 125, 126, 141 Coulson, S., v, vii, 5, 101, 117, 118, 120,
Barr, D., 146, 150, 163 121, 123
Barwise, J., 42 Csordas, T., 153-155, 163-166
Basso, K.H., 189,202
Bates, E., 129, 142 D
Beardsley, M., 60, 75 Deignan, A., 187
Berglund, R., 42 DeLancey, S., 129, 130, 142
Berglund, U., 42 Demecheleer, M., 48, 55
Bharati, A., 154, 163 Dornseiff, F., 10, 11,23,24
Black, M., 75 Dougherty, J.W.D., 127, 142, 143
Blake, R., 35, 42, 44 Dowty, D., 42, 43
Bliss, E.L., 127, 142 Draaisma, D., 49, 55
Blumenberg, H., 7, 9, 12, 15-17, 22-24 Duranti, A., 142
Bly, B., 37, 44, 164
Boers, F., vii, 5, 6, 47-50, 55 E
Bogdonovich, J., 146, 150, 163 Eggleston, R., 160, 163
Bovair, S., 62, 63, 75 Emanatian, M., viii, 6, 189, 195, 202, 205,
Brandt, P., 121, 122 209,213,214,217
Bral, M., 20, 24 Engen, T., 36, 43
Britt, M., 63, 76
Britton, B., 62, 63, 75 F
Brown, M., 146, 164 Fauconnier, G., 76, 93, 101, 102, 107, 111,
Brugman, C., 42, 81, 111, 123 114, 115, 117, 120, 121, 123, 124, 213,
Buck, CD., 36, 42 217,218
Bhler, K., 10, 11,24 Fernandez, J., 153, 163, 165, 187, 218
Fillmore, C, 115, 123
C Fodor, I , 35, 43
Cacciari, C , 4, 7 Forceville, C., 160, 163
Cain, W.S., 36, 42 Frege, G., 42, 43
Calbris, G., 193, 202
Cann, R., 41,42 G
Cas sirer, E., 10, 11, 14, 25 Gamble, E., 42, 43
Cienki, A., vii, 6, 189, 193, 195, 202, 216 Garnham, A., 63, 75
NAME INDEX 223
Geertz, C , 127, 129, 142 Johnson, M., 1, 7, 9-11, 13, 14, 17-19,
Gehlen, A., 24 21-22, 25, 26, 29, 34, 37, 41-44, 47, 49,
Gentner, D., 40, 43, 47, 49, 55, 72, 76, 122, 55, 57, 76, 79, 81, 83, 85, 89, 96, 98, 99,
123 101, 112, 122, 123, 131, 132, 140, 143,
Gerrig, R., 145, 162 145, 147, 148, 159, 164, 165, 167, 183,
Gibbs, R.W., Jr., v, vii, 1, 3-7, 40, 41, 43, 187,200,202,213,214,216,217
75, 81, 86, 99, 139, 142, 145, 146, Johnson, S., 164
148-151, 155, 156, 161, 163, 165, 172, Johnson-Laird, P., 10, 26
187,201 Jones, S., 137, 143
Gibson, J., 160, 164
Givon, T., 129, 142 K
Glucksberg, S., 40, 43, 92, 98, 99, 146,164 Kant, L, 7, 9, 12-14, 22, 23, 25, 114, 116,
Goldberg, A., 99, 121, 123 117, 121
Goodman, N., 10, 24 Katz, A., 7
Grady, J., v, vii, 5, 79, 81-83, 85, 96, 98,99, Katz, J.J., 43
101, 112, 122, 123, 180, 187, 195, 202, Kay, P., 213, 217
214,216,217 Keesing,R., 189, 190,202
Green, C., 164, 176 Keil, G., 12, 26
Greeno, J., 152, 164 Kennedy, J., 160, 164
Grudin, J., 49, 55 Keysar, B., 37, 40, 43, 44, 92, 98, 99, 115,
123, 146, 164
H Kieras, D., 62, 75
Haley, J., 158, 164 Kirmayer, L., 155, 156, 164
Harding, S., 129, 142 Klein, G., 160, 163
Hartung, J., 10, 11,23, 24 Kvecses, Z., vii, 6, 30, 42, 44, 48, 56, 147,
Henning, H., 43 153, 156, 164, 167, 171, 180, 183, 186,
Herder, J., 9, 25 187, 189,202,205,217
Hermann, P., 10, 11,26 Kryk, B., 42, 44
Herz, R., 37, 43
Hoffman, R., 25 L
Holland, D., 7, 127, 142, 143, 153, 164, Labov,W., 129, 143
167,187,202,203,217 Lakoff, G., 1, 2, 5, 7, 9, 11, 13, 14, 17-19,
Honeck, R., 25 21, 22, 26, 29, 34, 37, 41, 42, 44,47,48,
Hopper, P.J., 129, 142 56-58, 72, 76, 79, 81, 87-89, 91-93,
Howes, D., 42 95-99, 101, 104, 109, 111, 112, 121,
Hutchins, E., 152, 164 123, 129, 131, 132, 140, 143, 145,
Hymes, D., 129, 143 147-149, 153, 159, 164, 167, 171, 183,
187, 189, 199, 200, 202, 205, 213, 214,
I 217
Ibarretxe-Antunano, L, 5, 29, 38, 42, 43 Langacker, R.W., 116, 129, 130, 143, 189,
Ivanov, V., 190, 197, 202 191,202
Lawler, J., 26
J Lazarus, R., 155, 165
Jackson, J., 155, 164 Lea, M., 137, 144
Jkel, O., vii, 6, 9, 13, 14, 19-21, 23, 25 Leech, G., 42, 44
Jaynes, J., 10, 25 Lehrer, A., 31,44
Jeziorski, M., 72, 76 Leontiev, A., 152, 165
Liebert, W.-A., 23, 26, 55
224 NAME INDEX
Lindvall, T., 42 181-187, 189, 202, 203, 205, 211, 213,

Lock, M., 155, 165 217,218
Locke, J., 9, 11-13,26
Low, S., 54, 76, 81, 155, 165, 214, 216 R
Luria,A., 152, 165 Rappaport, R., 128, 143
Reddy, M., 203
M Reingold, H., 143
MacLaury, R., 25, 27 Reinhart, T., 58, 60-62, 64, 67, 69-72, 76
MacWhinney, B., 129, 142 Rice, S., 4, 7
Maglio,P., 121, 124 Richards, L, 60, 76
Mandelblit, N., 121, 124 Rigotti, F., 48, 49, 56
Margalith, A., 58, 75 Robert, A., 25-27, 121, 124
Martin, E., 25, 213,217 Roediger, H.L., 47, 56
Matlock, T., 5, 121, 124, 132, 143 Rogers, A., 42, 44
McGlone, M., 146, 164, 165 Rosaldo,R., 127, 143
McKoon, G., 145, 162
McNeill, D., 193, 194, 202, 215, 217 S
Miller, A.L,10, 26, 49, 56, 58, 64, 66-69, Sandra, D., 4, 7
71,72,76 Schiffman, S.,44
Morgan, P., 99, 123 Searle,J., 81,89, 99, 115
Mller, F., 10, 11,26 Sekuler, R., 35, 42, 44
Murphy, G., 4, 7, 37, 44, 147, 165 Shannon, B., 152, 165
Shore, B.,127,128,143,172,188,190,197,
N 200,203,205,211,213,216-218
Nascimento, S., 145, 163 Slater, C, 128, 143
Nayak,N., 145, 163, 165 Slobin, D., 129, 144
Smith, L., 152, 166
O Spears, R., 137, 144
Oakley, T., v, vii, 5, 101, 120, 124 Sperber, D., 127, 144
Obeyesekere, G., 128, 143 Steen, G., v, vii, 1, 3, 5-7, 41, 57, 61, 64, 74,
Ortega y Gasset, J., 10, 26 76, 154, 165, 201
Ortony, A., 24, 26, 40, 44, 71, 75-77, 99, Sternberg, R., 40, 45
203 Stigler, J., 157, 165
Strathern, A., 154, 165
P Strauss, C , 127, 129, 140, 143, 211, 213,
Palmer, G.B., 190, 191, 199, 202 218
Parkinson, B., 159, 165 Strom, L., 145, 164
Perfetti, C., 63, 76 Stromberg, P., 126-130, 144
Perry, J, 42 Suchman, L., 152, 165
Pesmen, D., 214, 217 Sweetser, E., 5, 8, 29-31, 33, 41, 42, 44, 81,
Peters, S., 42, 43 99, 120, 121, 131, 132, 140, 144, 145,
Porzig, W., 10, 11,26 165, 189,195,203,213,217,218
Postal, P., 35, 43 Sykes,J, 146, 150, 163
Synnott, A., 42
Q Szokolszky, A., 163
Quinn, N., 6, 7, 127, 128, 140, 142, 143, Szuchewycz, B., 129, 144
153, 154, 156, 164, 165, 167, 171-174,
NAME INDEX
Van Oosten, J., 129, 130, 144

Talmy, L., 203 Vanparys, J., 99, 195, 203
Taub, S., 99, 123 Veale, T., 103, 124
Taylor, J., 144 Vervaeke, J., 164
Thelen, E., 152, 166 Viberg, A., 30, 42, 45
Thompson, S.A., 129, 142 Vico, G., 9, 27
Toporov, V., 190, 197, 202 Vosniadou, S., 71,76, 77
Tourangeau, R., 40, 45 Vygotsky,L., 152, 166
Traugott, E.C., 144
Trier, J., 10, 11, 18,23,27 W
Turner, E., 4, 7, 42, 44, 45, 57, 72, 76, 79, Wall, R.E.,42, 43
81, 82, 87-89, 91, 92, 96-98, 101, 104, Weinrich, H., 7, 9, 17-23, 27
107, 111, 114, 115, 117, 120, 121, 123, Wennberg, A., 42
124, 144, 145, 149, 165, 166, 213, 218 Whorf,B., 10, 11,27
Turner, M., 45, 76, 166 Williams, J., 3, 8
Turner, V., 144 Wilson, R., 152, 166
Winkler, C, 155, 166
Wllner, F., 9, 10, 23, 27
Van Dijk, T.A., 58
Van Leeuwen-Turnovcova, J., 190, 203

Gibbs & Steen (Eds.) - Metaphor in Cognitive Linguistics PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Gibbs & Steen (Eds.) - Metaphor in Cognitive Linguistics PDF

Caricato da

Copyright:

Formati disponibili

METAPHOR IN COGNITIVE LINGUISTICS

AMSTERDAM STUDIES IN THE THEORY A N D

Series IV - CURRENT ISSUES IN LINGUISTIC THEORY

Advisory Editorial Board

Raimo Anttila (Los Angeles); Lyle Campbell (Christchurch, N.Z.)

Raymond W. Gibbs, Jr. and Gerard J. Steen (eds)

Metaphor in Cognitive Linguistics

RAYMOND W. GIBBS, JR.

JOHN BENJAMINS PUBLISHING COMPANY

Library of Congress Cataloging-in-Publication Data

Raymond W. Gibbs, Jr, University of California, Santa Cruz

Congruence by Degree: On the Relation between Metaphor

GERARD STEEN & RAYMOND W. GIBBS, JR.

Metaphor in cognitive linguistics is a two-way affair: it can go from linguistic

IDEAS ARE FOOD

IDEAS ARE MONEY

Conventional linguistic metaphors reflect pervasive conceptual metaphors and are

"anger," such as ANGER IS HEATED FLUID IN A CONTAINER and ANGER IS ANIMAL

(a) a commitment to seek general principles governing all aspects of human

As a result, the cognitive linguistic study of metaphor is related to the study of

possesses the same conceptual metaphors to the same degree of detailed

Some of these issues have been discussed, and various psychological

to experimentally test falsifiable hypotheses using more objective methods under

cognitive commitment mentioned above. If metaphor is just one specific case of

Sweetser, E. 1990. From Etymology to Pragmatics: Metaphorical and Cultural

Our perception proceeds in part by way of the senses, in part by means of

2. KanVs concept of analogy and his "symbolical sensualisation"

3. Blumenberg's philosophical "metaphorology"

... on closer inspection they are clearly oriented around a metaphorical

In addition to such metaphorical models Blumenberg (1960:69) speaks of

Their content determines behavior by providing orientation. They give

Many of Blumenberg's examples look familiar, too, as they remind us of

directionality of metaphorical transfer from concrete source domains to abstract

4. Weinrich's linguistic theory of metaphorical "image fields"

Is every metaphor rooted in an image field? That would be too much to

Again, Weinrich's exposition equals that of the cognitive theory of metaphor.

Black (1977), is a graphic confirmation of the basically constructivist view which

This "linguistic reality" of conceptual metaphor is up for investigation by modern

proponents of the cognitive theory of metaphor. Probably because they do

A metaphor is never a simple word, but always a piece of text, however

Thus by its use in a particular context, the original meaning of a lexeme is

5. Summary and conclusion

Kant analogy symbol 0

Blumenberg background metaphor cognitive model 0 0

Weinrich image field metaphor (hypothetical) image donor image recipient

Table 1: The central terminological correspondences between the cognitive theory of

In the explanation of the structure of these metaphors of perception, Sweetser

- Bad smell to indicate bad character or 'dislikeable' characteristics

(1) He is a stinker (Sweetser 1990:37)

- Detection of such characteristics

(2) I smell something fishy about this deal (Sweetser 1990:37)

2. The sense of smell: Extended meanings

The meanings discussed in this study do not represent a comprehensive typology

(4) Mary can smell trouble a mile off (OSD)

(5) Sailburuaren kontuakzuzenak ez zirela

(6) Kanturako haren zera ikusiz, mutrikuarra zela

(7) Me huele que ella est detrs de todo sto

(8) Juan ya se ha olido la broma

(11) Txakurra usnaka zebilen erbiaren bila

(12) Bere gauzetan usnaka ibili ondoren, bera

(13) El perro estaba olfateando el suelo en busca de

(14) Le han ordenado que husmee las

3. The sense of smell: Property selection process