CHAPTER 1 INTRODUCTION
1.2 The Current Study
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
gestures are generated from the semantics of lexical items. The third one maintains that the
information in gesture originates from the representations based on the on-line interaction of
spatial thinking and speaking. Kita and Öyzürek (2003) have conducted research on the
cross-linguistic expressions of motion events to look at the three hypotheses. They focused on
the informational coordination between iconic gestures and their corresponding lexical
affiliates. Likewise, the present study investigates the relationship between language and
gesture, but we will discuss the hypotheses from the perspective of metaphorical expressions.
1.2 The Current Study
The purpose of the thesis is to discuss: (i) people’s habitual expressions of metaphors to
conceptualize concepts in daily communication, and (ii) the collaboration of language and
gesture in expressing metaphors with regard to the hypothesis of speech-gesture production.
Concerning people’s habitual expressions of metaphors, five questions are addressed. What
are the metaphor types people usually convey in daily communication? What are the source
and target domains of the metaphors? What sources can be used to conceptualize multiple
targets? What targets can be realized by multiple sources? Do the metaphors concurrently
occurring in language and gesture and the metaphors occurring in gesture exclusively express
similar or different metaphor types, source/target domains, and source-to-target
correspondences? To discuss the cooperation between language and gesture, a question is
addressed. What is the temporal patterning of speech and gesture in presenting metaphors?
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The thesis is organized in the following order. Chapter 2 reviews previous studies
concerning the issues about conceptual metaphor and the theoretical hypotheses about speech
and gesture production. Chapter 3 introduces the data used in this study and the methodology
adopted to examine the metaphoric expressions. Chapter 4 reports the analysis of the
cross-modal manifestation of metaphors. Chapter 5 shows the general discussion and
compares the present study with the previous research. Chapter 6 is a conclusion of the thesis.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
CHAPTER 2
LITERATURE REVIEW
The thought that metaphor is not restricted to the realm of literature has been widely
accepted since Lakoff and Johnson’s study of conceptual metaphor in 1980. After that, a large
number of studies on metaphor in language as well as metaphor in gesture provide evidence
to support the conceptual and embodied view of metaphor. This chapter contains three main
sections. In Section 2.1, the notion of conceptual metaphor proposed by Lakoff and Johnson
and the past studies on the metaphoric expressions in language are discussed. In Section 2.2,
the previous studies on the gestural manifestations of conceptual metaphors are reviewed. In
Section 2.3, three theoretical hypotheses concerning the cognitive process underlying speech
and gesture production are introduced. Section 2.4 is the summary.
2.1 Conceptual Metaphor in Language
In Lakoff and Johnson’s framework, the word metaphor refers to the “metaphorical
concept” in thought and is presented in a form with small capital letters, for example, LOVE
IS A JOURNEY (Lakoff & Johnson 1980c: 6). The term metaphorical expression refers to the
surface manifestation of a metaphorical concept. Language is an essential modality for us to
understand the metaphors. Although we are not normally aware of our conceptual system, we
can explore the system by studying language, since communication shares the same system
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
we use in thinking (Lakoff & Johnson 1980c). In addition, metaphor is central to our ordinary
language semantics due to the fact that everyday abstract concepts like time, state, and
change can be metaphorical (Lakoff 1993). The research on the metaphoric expressions in
language provides a way for us to look at what conceptual metaphor is like. An overview of
conceptual metaphor is presented in Section 2.1.1. Studies on metaphors in language which
allow us to see the nature of conceptual metaphor are reviewed in Section 2.1.2. The studies
on the embodiment of conceptual metaphor are discussed in Section 2.1.3.
2.1.1 Overview of Conceptual Metaphor
According to Lakoff and Johnson, “[t]he essence of metaphor is understanding and
experiencing one kind of thing in terms of another” (1980a: 455, 1980c: 5). Namely,
metaphor can be conceived as a conceptual mapping from one domain to another domain
(Lakoff 1993). The conceptual domain used to understand another domain is called source
domain. The conceptual domain that is comprehended is called target domain. The source
domain is typically concrete, physical, and delineated; on the contrary, the target domain is
typically abstract, non-physical, and less delineated (Lakoff & Johnson 1980c). For example,
JOURNEY is the more concrete source domain and LOVE is the more abstract target domain in
the metaphor LOVE IS A JOURNEY. There are also instances that concrete ideas can be
understood in terms of metaphors, such as LAND AREAS ARE CONTAINERS (Lakoff &
Johnson 1980c: 29) and INANIMATE OBJECTS ARE PEOPLE (Kövecses 2002: 58). A
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
metaphor—a conceptual mapping—is a set of correspondences (Lakoff 1993). When we refer
to LOVE IS A JOURNEY, the mapping covers a set of correspondences: the lovers correspond
to the travelers, the lovers’ goal to the destination of the journey, and so on. Metaphorical
correspondences obey the Invariance Principle that “[m]etaphorical mappings preserve the
cognitive topology (that is, the image-schema structure) of the source domain, in a way
consistent with the inherent structure of the target domain” (Lakoff 1993: 215). The inference
pattern of a target follows the inference pattern of a source.
There are four false views of metaphor: metaphor is a rhetorical device—a matter of
language; metaphor is based on similarity; all concepts are literal not metaphorical; and
concepts are disembodied (Lakoff 1993; Lakoff & Johnson 2003). Lakoff and Johnson’s
Conceptual Metaphor Theory argues against these views. First, metaphor is in thoughts not in
language. Conceptual metaphors can be realized in both linguistic and non-linguistic ways
such as cartoons, rituals, sculptures, gestures, and so on. The cultural ritual that a baby is
carried upstairs to pray for success manifests the metaphor STATUS IS UP (Lakoff 1993:
241-242). The sculptures of oversized heroes manifest the metaphor SIGNIFICANT IS BIG
(Kövecses 2002: 58). The locus of metaphor is not in language but thought, which indicates
that metaphors can be conveyed by the modalities other than linguistic expressions. Second,
metaphor is typically based on source-to-target correlations in our experience not inherent
similarities. For instance, the metaphor MORE IS UP is grounded in the experience that adding
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
more water into a container will lead to the water level rising. The event of adding more
water is not similar to the event of the water level raising. Moreover, a metaphor may be
based on “perceived structural similarity” (Kövecses 2002: 71). For example, the source of
the metaphor LIFE IS A PLAY is not inherently like its target, yet we still perceive some
similarities between the LIFE and the PLAY. The relationship between a role and his/her
performing ways in a play is similar to the relationship between a person and his/her action in
real life. It is the use of a metaphor that creates perceived similarities (Lakoff & Johnson
2003). Third, metaphor is indispensable in our conceptualization of the world. Even the
mundane concepts like TIME, QUANTITY, and STATE are understood via metaphors. Since
most our everyday concepts to define our world are metaphorical, how we think and what we
do would be associated with metaphors (Lakoff & Johnson 1980a, 1980c). Fourth, conceptual
metaphor can be shaped by our body experiences. For example, the metaphor ANGER IS
HEAT is grounded in the experience that a person feels hot when he/she is angry. In addition,
there are metaphors which depend on our social and cultural practice (Lakoff & Johnson
1980c, 2003). VIRTUE IS UP is a metaphor with social and culture basis because virtue is the
standard embedded in the culture and set by the society.
To sum up, metaphor is the conceptual correspondence between two domains in the way
that the logic of a source is used to understand the inference of a target. The Conceptual
Metaphor Theory maintains four significant views about metaphors: metaphor is in thoughts;
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
metaphor is based on the correlations or the structural similarity between two domains;
metaphor helps to structure our ordinary conceptual system; and metaphor can be grounded
in the body or socio-cultural experiences.
2.1.2 The Nature of Metaphor
In this session, we will focus on the nature of conceptual metaphor through its
manifestation in linguistic modality. The partial characteristic of conceptual metaphor is
discussed first. The target domain of a metaphor is normally more abstract, and the source
domain of a metaphor is usually more concrete. Most of the time, a single abstract concept is
not completely or exactly defined by a single concrete concept. Lakoff and Johnson (1980b)
suggested that an abstract concept is normally understood in terms of more than one concrete
concept. A cluster of metaphors are used to understand an abstract concept, and each
metaphor partially defines the concept. For instance, the focuses are different in the following
two expressions: Life is empty for him (Lakoff & Johnson 1980c: 51) and He’s holding all the
aces (ibid.). In the former expression, life is comprehended as a container, and the emphasis
is on the content of life. In the latter one, life is understood in terms of a gambling game, and
we pay attention to how people live rather than what is in life. These examples denote that
“abstract concepts are not defined by necessary and sufficient conditions” (Lakoff & Johnson
1980b: 200). The partial nature of metaphor shows that a concept may be reasoned in terms
of different sources which profile different sematic aspects. Based on this notion, the present
‧
study examines what the targets are that people tend to conceptualize through various
sources.
Next, we will focus on the direction of metaphorical correspondences. Kövecses’s
(2002) finding supports the notion that metaphorical correspondences usually go from the
more concrete and delineated domains to the abstract and less delineated domains. He
collected the linguistic expressions of metaphors in dictionaries (e.g., Cobuild Metaphor
Dictionary) to survey the common sources and targets in his qualitative study. The frequent
sources given by him are arranged in Table 1.
Table 1. Common source domains of metaphorical expressions in English (Kövecses 2002: 16-20)
Source Example
The human body the heart of the problem
Health and illness a healthy society
Animals He is a donkey.
Plants the fruit of her labor
Buildings and construction She constructed a theory.
Machines and tools conceptual tools
Games and sports He tried to checkmate her.
Money and economic transactions (business) Spend your time wisely.
Cooking and food What’s your recipe for success?
Heat and cold a warm welcome
Light and Darkness She brightened up.
Forces I was overwhelmed.
Movement and direction He went crazy.
These sources are concrete in general, and they are what we are familiar with.
Kövecses also mentioned that basic entities and the properties of the entities are common
source domains as well. The sources suggest that metaphor is grounded in our bodily
experience or socio-cultural practices. For instance, heat/cold and light/darkness associate
‧
with our perceptual experience. Forces and movement relate to our motor experiences.
Games and commercial activities are the socio-cultural practice we perform in everyday life.
Table 2. Common target domains of metaphorical expressions in English (Kövecses 2002: 21-24)
Target Example
Emotion He was bursting with joy.
Desire She is hungry for knowledge.
Morality that was a lowly thing to do
Thought I see your point.
Society/ Nation What do we owe society?
Politics the president plays hardball
Economy the growth of the economy
Human relations the built a strong marriage
Communication That’s a dense paragraph.
Time Time flies.
Life and death Grandpa is gone.
Events and action She has reached her goals in life.
Religion the God’s sheep (sheep = follower)
The common targets given by Kövecses are arranged in Table 2. He roughly classified
the most frequent targets into psychological states and events (emotion, desire, morality, and
thought), social groups and process (society, nation, politics, economy, human relationship,
and communication), and personal experiences and events (time, life, death, and religion).
Compared to the frequent sources, these targets are more abstract and less delineated. We
may say the healthiness/illness of the society, yet we do not commonly talk about the society
of healthiness/illness. Kövecses then concluded that metaphors are normally unidirectional;
that is, the corresponding direction between the sources and targets is asymmetry.
Furthermore, some metaphors are universal. Expressions in Chinese are compared with
expressions in English in several qualitative studies on metaphors in Chinese (including
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Mandarin and Southern Min). In Yu’s (1998) study, metaphors in Mandarin were collected
from dictionaries to compare with the expressions of emotion metaphors, time as space
metaphors, and event structure metaphors in English. Lin (2003) obtained data from
dictionaries and compared the body-part metaphor in Southern Min with the ones in
Mandarin and English. Lai (2011) used data in pop songs to explore the sources for the
metaphor about LOVE in Southern Min. Metaphoric expressions of love in Southern Min,
Mandarin, and English are compared in her study. She also found that multiple source
domains (e.g., FOOD, PLANT, and GAMBLING) map to the target-domain concept of LOVE.
The above studies investigated metaphors from cross-language perspective, and it was found
some metaphors are universal and some are culture specific. In Wang’s (2010) quantitative
study, he gathered data from pop songs to examine love metaphors in Mandarin. He also
found love can be realized by different types of metaphor. Results show that the event
structure metaphor and the ontological metaphor are the most frequent types of metaphors to
express LOVE.
Kövecses’s (2002) study in English supported that source domains are more concrete,
that target domains are more abstract, and that the corresponding direction is asymmetrical.
The cross-linguistic research (Yu 1998; Lin 2003; Lai 2011) revealed there are some
universal metaphors shared by different cultures. Wang’s (2010) and Lai’s (2011)
investigations on love metaphors provided evidence for the notion that an abstract concept
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
can be defined by different metaphors which highlight different semantic elements of the
concept. The findings also showed that a single target-domain concept can map to multiple
sources. Except for Wang’s study of LOVE metaphor, all the studies mentioned above only
employ qualitative analysis. The present study attempts to analyze the data in both
quantitative and qualitative ways to gain dependable information about the habitual
expressions of metaphors and the correspondences between the two domains.
2.1.3 The Embodiment of Metaphor
Embodied cognition holds the view that cognition is rooted in the body’s interaction
with the world (Wilson 2002; Gibbs 2005). Such view rejects that “cognition is computation
on amodal symbols in a modular system” but proposes that cognition is grounded in
simulations, situated action, and, bodily states (Barsalou 2008: 617). With regard to language
comprehension, empirical evidence demonstrates that language is grounded in bodily action.
In Glanberg and Kaschak’s (2002) study, participants decided whether a sentence is sensible
after reading a sentence of which the implied direction was manipulated (e.g., open the
drawer implies action toward body). Results revealed that bodily action can facilitate or
interfere with our understanding of a sentence. Hauk et al. (2004) used event related fMRI to
record brain activity in a passive reading task. It was found words referring to actions of
different body-parts would activate different brain areas which may be activated by actual
movement of the tongue, fingers, or feet. The findings implied that the meaning of action
‧
words correlate with the physical body action. According to the Conceptual Metaphor Theory,
“which metaphors we have and what they mean depend on the nature of our bodies, our
interactions in the physical environment, and our social and cultural practices” (Lakoff &
Johnson 2003: 247). Metaphoric expression in language also provides evidence for the
embodied cognition.
Johnson (1987) and Lakoff (1987) introduced the notion of “image schema” to
metaphorical projection. An image schema is “a recurring dynamic pattern of our perceptual
interactions and motor programs that gives coherence and structure to our experience”
(Johnson 1987: xiv) and following is a more specific definition of image schemata:
[T]hey are not Objectivist propositions that specify abstract relations between symbols and objective reality...they do not have the specificity of rich images or mental pictures... A schema consists of a small number of parts and relations, by virtue of which it can structure indefinitely many perceptions, images, and events. In sum, image schemata operate at a level of mental organization that falls between abstract propositional structures, on the one side, and particular concrete images, on the other” (Johnson 1987: 28-29)
Image schemas are studied as the embodied bases for metaphoric extensions in past research
(Johnson 1987; Lakoff 1987, 1993; Lakoff & Johnson 1999; Kövecses 2002; Gibbs 2005,
2006; Johnson 2007). Three important aspects of image schema relate to the grounding of
meaning (Johnson 2005: 21-22, 2007: 139). First, image schema is part of the thing that
makes bodily experience to have meaning for us. Second, image-schematic structure has a
logic which makes it possible for us to make sense of our everyday experiences. Third, image
‧
schema is not merely mental or merely bodily, but a contour of body-mind. Johnson (1987)
proposed a selective list of image schemas which he thought to be the more important image
schemas (see Table 3). It should be noted that “the image schema list has never constituted a
closed set” (Hampe 2005: 2). If one defines schema more loosely than Johnson does, it is
possible to extend the list.2 Nevertheless, the core of the standard inventory of image
schemas is on the basis of Johnson’s list of schemata and many of the image schemas offered
by Johnson are frequently discussed in other studies.
Table 3. Image schemas proposed by Johnson (1987: 126)
Image Schema
CONTAINER BALANCE COMPLUSION
BLOCKAGE COUNTERFORCE RESTRAINT REMOVAL
ENABLEMENT ATTRACTION MASS-COUNT
PATH LINK CENTER-PERIPHERY
CYCLE NEAR-FAR SCALE
PART-WHOLE MERGING SPLITTING
FULL-EMPTY MATCHING SUPERIMPOSITION
ITERATION CONTACT PROCESS
SURFACE OBJECT COLLECTION
The metaphor MORE IS UP is an instance based on the SCALE schema which is
grounded in the experience that when we add more water into a container, the water level
rises. Such an image-schematic based metaphor is treated as a member of the “primary
metaphor” category, which shows a straightforward correlation to our ordinary embodied
experience (Lakoff & Johnson 1999; Gibbs 2005, 2006). Unlike primary metaphor,
2 Some image schemas were not included in Johnson’s list: FRONT-BACK (Lakoff 1987), UP-DOWN (Lakoff 1987), SELF MOTION (Mandler 1992), LEFT-RIGHT (Krzeszowski 1993; Clausner & Croft 1999), FORCE (Cienki 1997;
Clausner & Croft 1999), IN-OUT (Clausner & Croft 1999), and SPACE (Clausner & Croft 1999).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
“compound” or “complex” metaphor does not suggest a direct experiential basis (Lakoff and
Johnson 1999; Gibbs 2005, 2006). The complex metaphor A PURPOSEFUL LIFE IS A
JOURNEY does not directly relate to the image schema but it is still embodied. This metaphor
is built up by the primary metaphors (PURPOSE ARE DESTINATIONS and ACTIONS ARE
MOTIONS) and the cultural knowledge that people are believed to have purposes in life and to
act to achieve the purposes (Lakoff and Johnson 1999: 60-61).
Empirical evidence also reveals that people’s understanding of metaphorical language
is based on non-linguistic experience. In the experiment made by Gibbs et al. (1997), people
made lexical decisions to letter-strings related to metaphor (e.g. heat) faster after they read
the idioms (e.g., blew his stack) with metaphorical implications. The result showed that
people could compute embodied representations as soon as they understand the idioms with
metaphorical meaning. Research made by Gibbs et al. (2004) investigated people’s
interpretation of metaphors realized in language (based on DESIRE IS HUNGER) about human
desires. Their research tests participants’ acceptability of different linguistic expressions of
desire. Results implied that embodied knowledge could be the primary source of
metaphorical meaning and understanding. In Wilson and Gibbs’s (2007) experiment, people’s
reaction time for comprehending the metaphors was faster in the matching prime condition
when they practiced or imagined the priming actions which relate to the metaphors. The
results demonstrated that real and imagined body movements associated with metaphorical
‧
phrases can facilitate people’s immediate comprehension of the metaphors.
2.2 Conceptual Metaphor in Gesture
So far, we have discussed the studies on linguistic expressions of metaphors. However,
some scholars suspect that the language-based analysis can directly reflect the pattern of our
some scholars suspect that the language-based analysis can directly reflect the pattern of our