Content uploaded by Bálint Forgács
Author content
All content in this area was uploaded by Bálint Forgács on Dec 15, 2021
Content may be subject to copyright.
1
The Pragmatic Functions of Metaphorical Language
1
Bálint Forgács
Abstract
Figures of speech have been suggested to play important pragmatic roles in language. Yet the
nature of these pragmatic functions has not been specified in detail, and it is not clear what
particular social-communicative purposes metaphors fulfill. I propose that metaphors are
utilized in two distinct ways in communication. First, similarly to indirect speech, they can be
utilized in social bargaining: by expressing intentions, beliefs and desires in a veiled manner,
they put the burden of interpretation on the hearer, which makes them revocable and thus a
great tool for negotiations. Secondly, metaphors can be used to transform the literal meaning
of words to describe phenomena and refer to concepts that do not have a lexical entry, by
transferring some abstract sense figuratively. This latter use is not only a tool of verbal creativity
but a means of linguistic change as it adds novel senses to words. Metaphor does not seem to
be a mere example of loose language use, but a sophisticated communicational tool, either to
deliberately create ambiguity in a deniable manner, or to extend word meaning beyond the
public lexicon, which puts some fundamental mechanisms of abstract thought to figurative use.
1
This chapter is a deeper elaboration and further development of ideas laid out initially in the very first scientific
paper I have written, in Hungarian (Forgács, 2009), as a PhD student guided by Professor Csaba Pléh. I would like
to express my eternal gratitude for his helpful, careful, and truly transformative mentorship and for introducing
me, through his most welcoming and witty manner, to the excitement and joy of cognitive science, in hope of
following in his footsteps towards the heights of the science of the mind. Csaba has been a sharp and open minded
mentor from the classical school, encouraging investigations outside of his main area of interest, which shows in
the diversity of his students’ research. As a brilliant scientist, excellent speaker and outstanding mentor there is a
lot to learn from him, but the Pléh-superpower seems to be beyond reach: Csaba is able to recall and recommend
practically any author, book and idea from the recorded history of psychology and philosophy alike. Particularly
noteworthy is the smaller havoc that erupted in the PhD room when someone accidentally printed his full
bibliography instead that of the past five years: the printer could not be stopped. His productivity and intellectual
freedom combined with his friendliness and organization skills not only enabled him to establish cognitive science
in Hungary, but to bring fresh air to the study of mind and language at large – which is radiating ever further
through the several generations of students all over the globe he helped spread their wings.
This work was funded by a NKFIH Young Researcher grant (125417) to Bálint Forgács.
Please cite this chapter as: Forgács, B. (2022). The Pragmatic Functions of Metaphorical Language. In G. Csibra,
J. Gervain, & K. Kovács (Eds.), A Life in Cognition - Studies in Cognitive Science in Honor of Csaba Pléh (pp.
41–57). Springer.
The final version of this preprint is available at: https://doi.org/10.1007/978-3-030-66175-5_4
2
1 Introduction
Why do we speak figuratively, if everything could be told in a simple, direct, literal manner?
Metaphorical language has been fascinating scholars since the time of the ancient Greek
philosophers, but the question of why and how we use it gathered considerable attention again
only in the 20th century. The chapter is first going to review two radically different, but
prominent views of metaphor that propose different roles for them in language and cognition,
Lakoff and Johnson’s (1980) Cognitive Metaphor Theory and Sperber and Wilson’s (2008)
Relevance Theory based deflationary account. After taking stock of prior models, the pragmatic
functions of metaphors are going to be discussed by highlighting two important roles they play
in communication beyond rhetoric: revealing hidden properties and covering up risky topics.
Finally, a novel model, Abstract Conceptual Substitution is proposed, which unifies the unique
semantic and specific pragmatic aspects of metaphor.
There is no commonly accepted definition of what metaphor is, but it can be thought of
as an expression linking two distinct concepts and thereby creating a comparison or an analogy,
often via an identity or category statement (e.g., ‘Odysseus is a lion’). This process of qualifying
the target or base of the metaphor (‘Odysseus’) using a distant concept creates for the source or
vehicle expressions (‘lion’) a meaning beyond its primary or literal interpretation (the ‘king’ of
the animals), and it is this figurative sense (‘strength, bravery, pride’) that most metaphor
theories try to account for.
The first known scholarly discussion is from Aristotle (322 BC/1952, 335 BC/1952),
who suggested that metaphor is an ornament of language, indispensable in prose and poesy, but
its expressive power makes it a particularly important rhetorical tool as well. Its skilled use is
“the mark of genius, for to make good metaphors implies an eye for resemblances.” Aristotle’s
view and interpretation dominated thinking about metaphor for over two millennia, and it has
not been challenged profoundly until the last century. Bréal (1899) was the first to suggest that
metaphors are not only poetic or rhetoric tools, but they serve as important means of linguistic
innovation and change, and that they are widespread in everyday language; Black (1962) noted
that they work on a conceptual level. Figurative language has been recognized to play an
important role in communicational pragmatics and consequently lie beyond typical language
processing (Pléh, 2000; Van Lancker Sidtis, 2006). Some suggest that metaphors are unique
and special, either because of the way we use them (Aristotle, 335 BC/1952; Grice, 1975) or
because of their role in cognition (Lakoff & Johnson, 1980; Mithen, 1996; Pinker, 2010), while
others hold that they are mundane and just one of many possible tools of communication, and
3
neither have a special place in language (Sperber & Wilson, 2008), nor require specific
comprehension mechanisms (Carston, 2010; Wilson & Carston, 2007).
2 Conceptual Metaphor Theory
Lakoff and Johnson (1980) claim that metaphors are not only extensively used in everyday
language but structure thought. They demonstrated that a vast number of expressions in simple
conversations are figurative in nature (e.g., ‘above all things’), often without being recognized
as such, and suggested that this is because metaphors reflect a deeper organization of the
conceptual system and cognition at large. Conceptual Metaphor Theory (CMT) (Lakoff &
Johnson, 1980; Lakoff, 1987) proposes that metaphorical mappings between conceptual
domains form the backbones of and provide structure for concepts. We understand abstract
notions (e.g., THEORIES)
2
by systematically mapping more concrete, experiential, real world,
imageable ideas on them (e.g., BUILDINGS), which thereby provide internal relational structure
and content to otherwise vague concepts. Everyday expressions, like ‘It is a well-founded
theory’ or ‘His model collapsed’ are expressions of the deeper conceptual metaphor THEORIES
ARE BUILDINGS. Embodied cognition (Lakoff & Johnson, 1999; Gallese & Lakoff, 2005) is a
theory that claims that all cognitive functions have a bodily, experiential basis, and cognition
is founded on metaphorical mappings, which transmit perceptual, sensorimotor information
towards unimageable abstract ideas. The fundamental building blocks are primary metaphors,
which are motivated by perception and have a clear sensorimotor basis (e.g., ORGANIZATION IS
PHYSICAL STRUCTURE; PERSISTING IS REMAINING ERECT), and which combine into complex
metaphors (e.g., THEORIES ARE BUILDINGS) (Grady, 2005). Cognition and experience associates
in schemas of experiential gestalts where the whole seems more basic than the parts; for
example, experiencing physical warmth and smiling faces together in our childhood makes the
expression ‘warm smiles’ eventually self-explanatory (Lakoff & Johnson, 1980).
Embodiment (à la Lakoff & Johnson, 1999; Evans et al., 2007) and CMT (Lakoff, 2014;
Lakoff & Johnson, 1980) argue that the function of metaphors is to structure and organize
thought on a conceptual level. No special communicational or conversational role is assigned
to them: they permeate language and cognition. The reason Lakoff (1993) provides for utilizing
metaphors in everyday communication is that whenever we leap away from concrete, physical
experiences and start to talk about abstract entities or emotions we rely on metaphorical
2
Following the convention in cognitive science I use small capitals for concepts, single quotes for words and
expressions, double quotes for quotations proper, and italics for meanings, senses, referents, and any semantic
values (the word ‘dog’ expresses the concept DOG and refers to, means, and expresses the property of being a dog).
4
mappings, because that is the way we conceptualize them and otherwise we would have no
words to express them.
Overall, however, the central claims of Embodiment, that metaphorical mappings shape
thinking, language, and cognition by way of sensorimotor processes have not received
unequivocal empirical support in the decades following its conception (e.g., Mahon &
Caramazza, 2008), and embodied theories often offer axioms and presuppositions rather than
falsifiable hypotheses regarding cognitive processes (Evans et al., 2007). Embodiment is
fundamentally an Empiricist conception of cognition – with some eerie resemblances to the
ideas of Giambattista Vico (1668-1744), the empiricist philosopher (Nuessel, 2006) –, where
the heavy lifting from the concrete, perceptual to the abstract is carried out by metaphors. Yet,
neuroscientific experiments aimed at the processing of metaphors found that metaphors are not
processed like concrete but more like abstract language (Forgács & Pléh, 2019; Forgács et al.,
2015; Forgács, 2020), and consequently it is hard to see how metaphorical mappings can serve
as the fundamental mechanism of cognition being transmitters of sensorimotor information to
abstract domains once themselves are processed as abstract rather than concrete.
There are some further, theoretical problems with CMT as well (Bowdle & Gentner,
2005; Cameron & Deignan, 2006; McGlone, 2007), especially with its potential to explain
conceptual organization. Mappings are always partial (Kövecses, 2002), but it is not clear why
and which elements of a source domain are left out (e.g., WINDOWS or ROOFS are not parts of
THEORIES), why and which are transferred, and how they find their place in the target domain.
CMT also fail to address why one target domain can have (or require) multiple, rather different
source domains (LOVE IS A JOURNEY but also A PATIENT, MADNESS, MAGIC and WAR), or why a
source domain can contribute to multiple target domains (LOVE IS A JOURNEY just as well as
CAREERS and LIFE in general are) (Lakoff & Johnson, 1980). Lakoff and Johnson’s (1980)
collections of metaphorical expressions provide fascinating examples of structure in language,
but if so many non-systematic groups of conceptual mappings are necessary to understand each
abstract and emotional concept, and concrete concepts can contribute to so many, arbitrarily
distinct and unrelated notions, the comport of the theory is rather doubtful. If any basic element
can be related to any complex element, structure is lost, and the model can hardly provide
predictions beyond rationalizations.
3 Relevance Theory
On the other end of the spectrum of enthusiasm for metaphors lies an approach in stark contrast
with that of Lakoff, Sperber and Wilson’s (1986; Wilson & Sperber 2004) Relevance Theory
5
(RT). These authors argue that metaphors play no role in cognition, are not unique in language,
and do not form a natural kind of theoretical importance. In their “deflationary account”
(Sperber & Wilson, 2008) they suggest that we understand metaphors by the same inferential
mechanisms that we apply to any other forms of language, and metaphor is simply one of many
kinds of loose language use. Loose language broadens the denotation of decoded meanings
during meaning construction to include senses beyond the narrow lexical entry (Wilson &
Sperber, 2002); for example, when we hear ‘Holland is flat’ we do not expect ‘flat’ to strictly
and literally mean the flatness of a table but to include trees and houses.
Sperber and Wilson’s (1986) RT follow Grice (1975) in his footsteps when proposing
that human communication is inferential in nature: intended meanings are established not
merely based on information content (cf. Broadbent, 1958), but on the assumption that
communicative partners strive to change our thoughts. Utterances are utilized as pieces of
evidence for inferring what others wish to make us believe. Linguistic communication is borne
out of mutual attributions of mental states and intentions, and meaning is established based on
the “cognitive environment”, which includes the broader context, local common ground, mental
states, situation models, common history, etc. Therefore, mentalization is proposed to be an
integral part of inferring linguistic meaning and necessary for successful communication.
The way we accomplish comprehension is by striving to maximize relevance (Sperber
& Wilson, 1986) by maximizing cognitive effects and minimizing cognitive efforts. A cognitive
effect is a change in our representations, for example, a true conclusion based on an inference
using new information, either by reinforcing or confirming a previous assumption, or by
refuting a present assumption (Wilson & Sperber, 2004). In linguistic communication relevance
maximization is carried out in two steps. Initially, the logical form of the utterance is developed
(based on morphology, semantics, syntax), which creates a non-propositional frame. To
develop the proposition itself, relevance is necessary: based on the given cognitive
environment, we establish explicatures and implicatures. Explicatures are what follow from the
form of the utterance, for example, from the statement ‘it is cold here’ it could follow that a
window is open. Implicatures are purely inferential, implicit premises and conclusions, implied
and included in utterances, for example, that the window should be closed. Meaning is
constructed in parallel through these procedures, based on expectations and hypotheses.
Explicatures are beyond Grice’s (1975) notion of “what is said”. The surface meaning might
not be intuitively so clear or common sensical and itself is a subject matter of decoding and
inference, which means disambiguation, enrichment, and fine tuning of the decoded meaning
(Wilson & Sperber, 2002). Importantly, Grice (1975) argued that the default interpretation of
6
“what is said” is the literal meaning and suggested that, for example, metaphors violate it. With
the introduction of explicatures Wilson and Sperber (2002) relieved themselves from a
commitment to a literal layer of word meaning.
3.1 Loose Language Use
According the RT (Sperber & Wilson, 1986) and the lexical pragmatist approach
(Wilson & Carston, 2007) meaning is always construed on-line by means of narrowing and
broadening decoded meanings. When someone says, ‘I have a temperature’ or ‘I’m not
drinking’ we narrow meaning to include only fewer and alcohol, respectively; and when we
hear ‘it’s boiling hot outside’ we broaden ‘boiling’ to include temperatures well below 100 °C.
Loose language use is created by broadening, which is often employed to deliver extra
communicational effects, including poetic and rhetoric effects: “Juliet is the sun.” When
speaking loosely, speakers optimize relevance by not strictly speaking the truth: implicatures
are weak, which means that the audience cannot be entirely certain what the speaker could have
had in mind, which leads to more cognitive effort, which, in turn, yields more cognitive effects
(e.g., ‘Should a president be grilled?’).
Loose language use is sometimes a communicational necessity. A statement is never
fully identical with the thoughts of the speaker, and the audience needs to develop some kind
of interpretive hypothesis regarding the speaker’s informative intention (Wilson & Sperber,
2002). We might struggle as speakers as well to find the exact words with close enough
meanings that could express a complex thought of ours. In fact, Sperber and Wilson (1998)
suggest that since we have a lot more concepts than words, but we need to rely on the words
available to us and our interlocutors, we always construe ad hoc meanings. Instead of using
word prototypes we narrow and broaden the meaning of lexical entries to express personal
concepts, for example, for a particular tingling, or simply to adjust to the local context.
Therefore, sometimes we speak loosely out of necessity, perhaps unintentionally, yet relatively
unambiguously, by picking the words we have close at hand to express ourselves meaningfully.
However, we could also speak loosely intentionally, when we want to evoke higher
cognitive effects in our audience on purpose. Some of these enhanced effects might be poetic
or rhetoric in nature – tropes, i.e. poetic tools, such as simile, metonymy, hyperbole, irony, etc.
(Nuessel, 2006) are good examples –, which might attract attention in everyday conversations
as well. Loose use can be employed both unintendedly and strategically. The question is, then,
why speaking ambiguously loosely on purpose? Some cognitive effect might not stem merely
from a higher processing effort due to semantic but to social factors. Multiple layers of meaning
7
can, for example, function similarly to indirect speech. Indirect speech is a way of phrasing and
framing requests or statements to create interpretational ambiguity that allows for social
bargaining, for example, by offering a bribe covertly: ‘Is there another way to solve this?’ (Lee
& Pinker, 2010; Pinker et al., 2008). Thus, an additional reason for expressing ourselves loosely
besides not having enough words, or making an emphasis by creating a strong effect, might be
speaking ambiguously to open a way for relationship negotiations and social bargains.
3.2 Metaphors in Relevance Theory
Cognitive effects are generated in everyday communication, but Sperber and Wilson
(2008) suggest there is no profound difference between metaphorical and non-metaphorical
language. Literal meaning is just as underdetermined as figurative, and literal interpretations
are not the default. In fact, audiences never take entirely narrowly, strictly, or literally what
speakers say. We speak loosely for the sake of optimal relevance, because sometimes using
figuratively the words both we and our interlocutor have is the most effective way to hint at a
thought we have. Metaphorical effects might not be extraordinary, as such interpretations can
be simply more relevant than literal ones (Tendahl & Gibbs, 2008). For example, when we say,
‘his head is empty’ we do not instantly think that it literally contains nothing, yet we do not
need much extra effort and might not derive much extra effect from such ‘empty headedness’.
There is, however, some tension in RT’s conception: metaphors require not only
broadening (of their lexical meaning) but some narrowing as well (of their literal senses).
Carston (2010; Wilson & Carston, 2007) and the lexical pragmatists propose that we create ad
hoc concepts during conversations, where we use broadening to expand denotations beyond
encyclopedic meaning. For example, when we say, ‘Boris is a giraffe’, we simultaneously
broaden and narrow meaning, to refer only to Boris’ overall height, not to the length of his neck.
This is how the explicature of metaphors are born, but there are two concerns here. First, it is
left underspecified, which are the features that are excluded by narrowing, which are the
features that are included in the broadening, and which are the features that are excluded from
the broadening. Giraffe is broadened to include humans, narrowed to not include neck length
or any features apart from height, and from the broadening all humans apart from Boris are
excluded. The theory provides a clear model to describe cases of metaphorical language and
meaning construction in general but does not explain why a particular dynamic unfolds for one
particular word in one particular context.
Secondly, by proposing a lexical or an encyclopedic meaning for the basis of narrowing
and broadening, lexical pragmatists and RT let ‘literal’ back through the back door. Wilson and
8
Sperber (2002) argue that Grice’s (1975) literal meaning cannot be the basis of comprehension
because it assumes an encoding-decoding of information and treats literalness in terms of
truthfulness – and they break away from this tradition of truth-value semantics. Indeed, the
neurocognitive processing of metaphors involves neither the recognition of a violation of a
literal meaning nor an ensuing paraphrasing by transformation into a simile in a serial fashion
(Blasko & Connine, 1993; Forgács et al., 2014; Forgács et al., 2012; Ortony et al, 1978). Instead
of literal meaning, Wilson and Sperber (2002) claim that it is the decoded, the encyclopedic or
lexical meaning that should matter, however, such entries start with the most ‘basic’ sense,
which typically turns out to be a literal one. In some cases, the figurative sense could be the
first entry, for example, for conventional metaphors (e.g., the word ‘cool’), but would that mean
that using them literally requires narrowing? Literal is not expelled but gains a different sense
in RT: some sort of literal meaning remains the basis of narrowing-broadening and even if
lexical (literal) meaning can be subjected to narrowing, sometimes it labels the narrow end of
the continuum (Sperber & Wilson, 2008). Sperber and Wilson (2008) suggest that literalness is
somewhat preserved with narrowing, but not with broadening – their point being that narrowing
and broadening are not functionally distinct uses of language, and meaning construction is
independent of figurativeness. Nevertheless, quite indirectly and not in a Gricean sense,
literalness seems to play a role in RT, if not a definitional but an interpretive or anchoring one.
3.3 Indirect Speech
While narrowing and broadening primarily considers lexical pragmatics and the
construction of meaning on the word level, loose meanings can arise on the sentence level as
well. The strategic speaker theory of Lee and Pinker (2010) puts forward the idea that indirect
speech is not exclusively a polite or eloquent manner of expressing desires (‘Could you please
close the window?’), but it has a structure that allows for negotiations by way of ambiguity.
Polite indirectness has been analyzed as preemptive face saving strategy when risking the
rejection of a request (Brown & Levinson, 1987). Consequently, some of its forms (“off-record
indirect speech”) creates ambiguity on purpose, which opens a way for safe backtracking. For
example, when attempting to bribe a police officer, by asking ‘Is there another way to solve it?’
two possible interpretations are on offer: a safe, neutral one that pretends some sort of ignorance
of rules or procedures and a risky one offering a bribe that could have serious consequences.
The two, simultaneously available interpretations allow for backtracking via the plausible
deniability of the risky offer by exploiting the neutral surface. The police officer might accept
the bribe or reject it but has no strong evidence to prove the offer itself. Therefore, a safe
9
negotiating situation is established, where both interlocutors can choose between the two
possible interpretations, and they can reach an agreement without openly expressing anything
legally binding. The above basic structure of indirect speech is utilized in a great variety of
social situations to express socially risky beliefs and desires in order to negotiate relationships,
from sexual offers to resolving conflicts (Pinker et al., 2008).
Many forms of loose language use allow for plausible deniability. Sarcasm, irony, and
joking in general are regularly offered as ways of backtracking from offensive or otherwise
harsh statements (‘I didn’t mean it’). Another often utilized excuse is ‘speaking figuratively’.
Since some metaphors can have both a figurative and a literal interpretation, they can be used
beyond simply denying what was meant and to open up a second layer across sentences (e.g.,
journey planning as relationship planning and vice versa), which could allow for flexible
negotiations much like indirect speech proper. Indirect speech is not always metaphorical, and
metaphor is not always indirect speech, but the dual nature of metaphor endows it with unique
pragmatic functions.
4 The Pragmatic Functions of Metaphor
Metaphor appears to play various roles in human communication, which makes it
unique among the tropes and other forms of loose language use. Its expressive power has been
recognized for millennia and it has been utilized by poets and rhetors ever since. It also plays
important roles in everyday communication, but instead of merely being either a conceptual or
lexical tool to establish word meaning, it seems to serve particular communicative purposes to
establishing social meaning as well. Here I propose that metaphor has two main pragmatic
functions: first, we can speak metaphorically with the aim of creating two parallel layers of
meaning, a figurative and literal one, which allows for social bargains and veiled conversations,
similarly to indirect speech. Second, we use metaphors to extend the meaning of words with a
particularly protruding conceptual property in a figurative direction to express a novel notion
via cross-domain similarity that we have no words for in the public lexicon. In the former case
we intend to increase ambiguity, so beliefs and desires can be communicated in a
comprehensible, but deniable manner, while in the latter case we intend to decrease ambiguity
by providing specifications and analogies using vivid words for ideas we have no common way
to discuss yet.
The duality of metaphorical and literal layers of meanings together with the combining
of conceptual domains place metaphor in a unique position among other forms of loose
language use. Hyperbole is when a notion or property, for example, coldness, is expressed in
10
an exaggerated manner (e.g., ‘What an Arctic summer night!’) but the figurative content
remains in the conceptual domain of the literal meaning it exploits, that is, COLD. It is the same
case with approximation (e.g., ‘Holland is flat’ – not entirely, but still) and category extension
(‘Google it!’ – which is still a search). Metonymy (‘The theater erupted in applause’) and
synecdoche (‘Black Lives Matter’) both involve mappings but within domain part-whole
relations (Lakoff & Johnson, 1980; Lakoff, 1987). Irony states just the opposite of what the
speaker has in mind (‘What a pleasant blizzard’), which is intimately related, as an antonym.
Synesthesia involves two domains but both are perceptual (‘purple song’). Simile involves two
distinct domains (‘this meal is like a Bildungsroman’), but due to the comparison, some literal
senses remain active (Glucksberg & Haught, 2006), which could risk the deniability of the
figurative content. A metaphorical meaning is not only figurative, but it transfers only non-
literal meanings to a different conceptual domain. Such a transmutation through identification
– or categorization (Glucksberg, 2003) – can create a tension between the literal and the
figurative senses of the vehicle and mutually exclusive interpretations. It is in this sense that
among all tropes only metaphors possess a kind dual layer that allows for negotiations during
social bargains and are able to catapult meaning into entirely different dimensions.
Historically, the most recognized use of metaphors is that of rhetors and poets as
described by Aristotle (350 BC / 1952): a good metaphor has strong evocative power and the
similarity it elucidates through an analogy, based on vivid imagery, makes it a great
communicational tool, even beyond being a mere ornament of speech. Rhetoric metaphors are
certainly generating enhanced cognitive effects (Sopory & Dillard, 2002), but not necessarily
through ambiguity. A distinction between two kinds of broadening is useful here: broadening
in order to create multiple possible meanings and broadening to include an unusual but
particular meaning. Poetic metaphors play with ambiguity and multiple layers of senses, and as
characteristic examples of loose use, they belong to the first class by leaving the interpretation
for the audience. Rhetoric metaphors, however, often strive to hit unambiguously upon a
felicitous example, image or parallel, therefore they seem to belong to the second class, where
broadening serves specificity.
4.1 Sensualizing: Broadening for a Particular Novel Figurative Meaning
The cases where broadening is targeted could be labelled “sensualizing”, and it is one
of the paradigmatic uses of meaning broadening as suggested by Sperber and Wilson (1998),
one that involves a concrete source domain. When we wish to express a novel or a personal
concept for which there is no word in the public lexicon, we bend the meaning of existing
11
words. When we create a novel metaphor for that end, we often use some kind of a tangible
notion, based either on a sensorimotor experience or on a vivid image, which is easy to relate
to for the audience. For example, when physicists discovered qualitatively different kinds of
quarks, to express their distinctiveness, they coined the term “quark flavors.” Instead of creating
obscurity, the expression instantly transmits the idea that quarks differ from one another just
lake tastes do. Such metaphors elucidate one particular aspect of the topic or target domain of
the metaphor (QUARKS) – a feature, a relation, or a structure – by way of a prominent part,
feature, relation or structure of the vehicle or the source domain (FLAVORS).
Metaphors are regularly employed to aid explanations and even to create knowledge in
scientific contexts (Nuessel, 2006). Scientific language is abundant with novel metaphors in
psychology (Leary, 1994) and cognitive science (Forgács, 2013) as well. Metaphors in science
are partly rhetorical tools, partly attempts at highlighting a key feature of a phenomenon, but
very often they give rise to analogies, which can become obstacles in thinking. Scientific
scrutiny can excavate a hidden or even unintended analogy that leads theoretical and
experimental work or can develop it to such a detail, that it does not hold any longer. Scientific
debates can revolve around the literal or figurative interpretation of a model or a key concept
for decades. Still, a well-struck metaphor can provide a concrete handler for a distant concept.
Metaphors are produced in clusters in lectures, lessons, and sermons when a difficult or
unfamiliar topic is discussed, and a few central or “root” conceptual metaphors are extended
systematically to illuminate various details (Cameron & Stelma, 2004). Similarly, in everyday
conversations an initial metaphorical topic can be expanded into an entire target domain and its
vehicle could be a first allowance from an entire source domain, but developing systematic
mappings seem neither necessary nor compulsory. The reason why most metaphorical
mappings are partial, incomplete, and diverse could be that one metaphor might describe a few
key aspects well and perhaps it could be extended a bit further, but the analogy breaks down
eventually, and then a different source might be a better choice to further elucidate the target.
Sensorimotor experiences, wild images, physical, concrete words might be used as
sources not only because of their expressive power, but because they are the easiest to relate to:
after all they allude to percepts, as if they pointed at something tangible. However, such effects
might be pragmatic in nature, not semantic or neural, as embodiment claims (Gallese & Lakoff,
2005; Lakoff 2014). Concrete words do activate corresponding sensorimotor areas
(Pulvermüller, 2005), but motor activations, for example, happen very early on, around 200 ms
(e.g., Pulvermüller et al., 2005), hence they could be explained by early automatic cascading
activations and priming effects: the causal role of sensorimotor processes in language
12
comprehension has not been established (Mahon & Caramazza, 2008). Importantly, the
irrelevant meaning of ambiguous words is activated initially but suppressed around 200-300 ms
(Gergely & Pléh, 1994; Seidenberg et al., 1982; Swinney, 1979; Thuma & Pléh, 1999, 2000),
before semantic retrieval is completed, as indexed by the N400 response (Kutas & Federmeier,
2011). Therefore, when used figuratively, a concrete word’s concrete, literal meaning might
activate sensorimotor areas initially, but it well could be suppressed by the time the intended
figurative sense is reached. The electrophysiological concreteness effect which indicates
sensorimotor feature processing (Barber et al., 2013) shows that novel metaphors are processed
similarly to abstract not to concrete words (Forgács, 2020; Forgács et al., 2015). Even though
sensorimotor activations do not seem to play a central role in the processing of metaphors as
suggested by Embodiment, figurative language can nevertheless induce strong cognitive
effects, when controlled mental imagery, analogical thinking, and verbal creativity mash
together in the mental representations of speakers and hearers.
Sensualizing through metaphorical sources is an important communicational-pragmatic
tool that provides imageable, meaningful, and easy-to-relate-to raw material for thought. Under
the pressure of expressing ourselves in communicative situations we regularly come up with
unusual wordings and phrasings to create novel meaning. It is particularly true for young
language learners, who do not have a large lexicon, but certainly present in all conversations
where stakes for self-expression are high. Such verbal creativity and meaning making (Bruner,
1990) could be a drive for linguistic change, when innovative terms or senses spread and
stabilize in a language community.
4.2 Speaking Figuratively: Broadening for Creating Ambiguity
Sometimes metaphors are used deliberately to create ambiguity by activating both a
literal and metaphorical layer of meaning to grab attention, for example in newspaper headlines
(Brone & Coulson, 2010). However, speaking metaphorically and ambiguously can also be
analogous with indirect speech in that it can open up a convenient escape route by enabling
deniability and transformation of social negotiations into negotiations of what was said (‘What
do you mean by »heating up our relationship?«’). If the other understood the metaphorical
referents, a covert and safe rejection or acceptance is also on offer. During social bargains and
relationship negotiations meaning is broadened but in a strategic manner: not to create
ambiguity in meaning but to open a space for a choice of an interpretation and ensuing
acceptance or rejection of social offers. For example, in emotionally loaded conversations of
the metaphor LIFE IS A JOURNEY can have both its literal (actual traveling) and figurative
13
meaning (emotional reconciliation) simultaneously available (Cameron, 2007), which could
make the intended meaning negotiable. For example, when an unreliable friend Therefore,
language is broadened in a particular direction, where the speaker strives to leave little
ambiguity regarding the veiled topic, while can keep an escape route of denial open. A
discussion of emotional matters (e.g., Cameron, 2007) or sexual advances (cf. Pinker et al.,
2008) could develop from an offer conveyed by a metaphor and if it is appropriated by the
conversational partner an agreement could be reached before or even without turning the topic
explicit (Cameron & Stelma, 2004).
In fact, entire conversations can be conducted under the veil of an innocent topic if both
parties are aware of the sensitive topic underneath. Adults regularly engage in such
conversations in front of children, when adult topics are discussed under the guise of a
seemingly innocent subject or by using words for which it is difficult to determine the exact
referents. Negotiations could be pushed to an overt level, by either party, the hearer could miss
the hidden layer, or the speaker could speak unintendedly in a manner that could be interpreted
on an emotional level, which could lead to unexpected consequences. When such interactions
are successful, a number of different metaphor vehicles could be employed while keeping the
topic constant. Such a conversational strategy could be the reason for having a multitude of
source domains available (e.g., JOURNEY, MADNESS, MAGIC or WAR) for certain target domains
(e.g., LOVE), while it also allows for a swift change of topic using the same words, which could
account for having the same source domain (e.g., JOURNEY) subserving various target domains
(LOVE, CAREERS or LIFE in general).
A closely related speaker strategy is suggesting that a statement was not meant literally
– after having made it. This strategy is typically a face saving act if and when an expressed
belief or desire did not bring applause (just like indirect speech, by way of politeness, Brown
& Levinson, 1987). Speakers can claim that they did not mean seriously or strictly literally what
they said in order to soften hearers and loosen up the search for possible interpretations.
Sometimes there is not even another sense to be found, because the statement was intended
narrowly, and requesting a broadening is merely backtracking. Such ‘de-literalization’ is
regularly employed from household quarrels to high politics to cover up statements that turned
out to be unacceptable for the hearers or which could have legal consequences. Since what was
said was not meant figuratively initially, it depends on the audience whether such post-hoc
offers of meaning broadening are accepted or rejected.
The pendant of the request for a figurative reinterpretation from a speaker is an offer
one by the hearer. Such a non-literal interpretation is similar is similar to the idea of “imposed
14
metaphoricity” (Porat & Shen, 2015), which suggests that practically any expression can be
interpreted metaphorically, regardless of its original content. Legal language is a prime example
for the opposite attempt: to exclude all alternative interpretations because conflicting partners
regularly try to offer retrospective reinterpretations of what was said and meant. The desire to
get rid of ambiguity makes legal texts not only dry but devoid of metaphors, which makes it a
paradigmatic case of literal language use. Proposing and requesting broadened interpretations,
beyond literal extensions well into the metaphorical, is often employed in conceptual art. Some
kind of meta-meaning is sought for by titles such as “Fountain” by Marcel Duchamp or “Cece
n’est pas une pipe” by René Magritte.
A search for a deeper meta-meaning behind words is the motivation of Fónagy’s (1971)
dual coding conception. He proposed that the arbitrary, symbolic, and propositional primary
code of language is complemented by a natural, archaic, emotion driven, iconic, secondary code
of phonotactics and intonation, which carries “symptomatic” or “symbolic” relationships
between signifiers and signifieds – as a kind of anchoring of metacommunication (Pléh, 2018).
From symptomatic behavior through intonation to choices of words with particular
phonological characteristics and hidden symbolic content and figurative senses, there is
metaphorical value in language use and expression, which reports of deep emotional
dimensions, via an intertwining of form and content, as a sort of verbal metacommunication.
A kind of metaphorical or symbolic interpretation is the approach of psychoanalysis as
well. Even though such a search for meta-meaning is more of a hermeneutic attitude rather than
a linguistic process per se, verbal expressions are of central importance here (Pléh, 2020). Freud
(1901/1989) provides an entertaining and insightful collection of slips of tongue and illusory
mishearings in emotionally charged topics and situations in an almost linguistic-pragmatic
analysis. Psychoanalysts investigate a fixed target domain of hidden emotional content, which
directly manifests in Fónagy’s primary code, metacommunicatively, but for which a semantic
content can be excavated via metaphorical jumps when broadenings are requested (via
association) or offered by the therapist. Szabó (2015) suggests that Lakoffian metaphors can be
understood as constituent of Jungian symbols in therapeutic contexts. Metaphorical language
use by patients could indicate therapeutic progress (Bach, 1994): metaphorical reinterpretations
reintegrate fragmented layers of meaning and personality. Although psychoanalytical insights
are not primarily linguistic in nature, the interpretive attitude together with the search for
mappings of the hidden target domain of the unconscious demonstrate how metaphorical
meaning can be put to work in therapeutic contexts.
15
5 Abstract Conceptual Substitution
How can metaphors fulfill these multifaceted pragmatic functions, and how can their pragmatic-
semantic aspects be bridged? In order to function as an efficient communicational tool, there
might be no need to assume systematic mappings (Lakoff & Johnson, 1980), complete
analogies (Bowdle & Gentner, 2005), blends (Fauconnier & Turner, 1998), ad hoc categories
(Glucksberg, 2003) or ad hoc concepts (Carston, 2010). Instead of structure mapping,
categorizing, filtering, or broadening some part of their literal meaning, the basis of metaphor
comprehension could be the abstract-concrete dimension. The theory of Abstract Conceptual
Substitution (Forgács, 2020, 2014) proposes that the meaning of metaphorical vehicles
(‘shark’) might be computed by the suppression of concrete, sensorimotor features (e.g., can
swim, marine predator, etc.) and the enhancement of abstract properties (tenacious, aggressive,
cruel). Consequently, the contextually most relevant abstract property can be simply substituted
as a figurative meaning (‘My lawyer is a shark’, becomes ‘My lawyer is tenacious’). The
filtering mechanism very similar to the one for polysemous words but it is carried out along the
abstract-concrete dimension, which yields an outcome superficially similar to a literal-
metaphorical distinction (Gernsbacher et al., 2001), yet it is not based on literal meaning in any
way. Sensorimotor information is neither carried over via mappings, nor rejected as literal
meaning, and no ad hoc categories or concepts need to be generated for each metaphorical sense
in each context. If the vehicle is abstract (‘The fakir’s bed is an oxymoron’) the taxonomic
information (TROPE) is suppressed but it is a still more abstract property that is enhanced and
substituted (CONTRADICTION). Therefore, literal meaning simply stands for concrete,
sensorimotor and/or taxonomic information. The model addresses the tension between the right
amount of simultaneous meaning narrowing and broadening for metaphors, as suggested by
Relevance Theory (Sperber & Wilson, 2008) and the lexical pragmatists (Wilson & Carston,
2007), if narrowing is understood as a procedure to exclude concrete features and broadening
to enhance or activate abstract properties. Thereby it also gets rid of the rather elusive notion
of literal meaning as a reference point to metaphoricity and replaces it by concrete, experiential,
physical meaning. Such a procedure is a unique combination of well described semantic
operations that also has the explanatory power to account for all metaphor-related phenomena
and neurocognitive data (Forgács, 2020; Forgács et al., 2015).
6 Conclusion
Metaphors stand out from other forms of loose language use by involving multiple layers of
meaning and combining distinct conceptual domains. They might not be the building blocks of
16
the conceptual system, but they seem to be unique communicational tools. They allow us to
point to concepts we have no words for, highlight similarities, initiate analogies, but also to
speak covertly. Offering a figurative reinterpretation as a speaker or a hearer, or
metaphorization creates a negotiation space either in a veiled, safe manner, or in a direct attempt
to address topics that are difficult to verbalize. Taken together, metaphor might reveal some
fundamental aspects of human cognition, from a search for meaning through meaning making
(Bruner, 1990) to the symbolic and abstract nature of thought: the process of abstracting away
from the concrete to the imaginary.
References
Aristotle (1952). Rhetoric. In Ross, W. D. (Ed.) & Roberts, W. R. (Trans.), The works of
Aristotle (Vol. 11): Theoretica, de rhetorica ad alexandrum, poetica. Oxford, England:
Clarendon Press. (Original work published ca. 322)
Aristotle (1952). Poetics. In Ross, W. D. (Ed.) & Bywater, I. (Trans.), The works of Aristotle
(Vol. 11): Theoretica, de rhetorica ad alexandrum, poetica. Oxford, England:
Clarendon Press. (Original work published ca. 335)
Bach, S. (1994). The language of perversion and the language of love. Lanham, Maryland:
Jason Aronson.
Barber, H. A., Otten, L. J., Kousta, S. T., & Vigliocco, G. (2013). Concreteness in word
processing: ERP and behavioral effects in a lexical decision task. Brain and
Language, 125(1), 47-53.
Black, M. (1962). Models and Metaphors. Ithaca, NY: Cornell University Press.
Blasko, D., & Connine, C.M. (1993). Effects of familiarity and aptness on metaphor
processing. Journal of Experimental Psychology: Learning, Memory, and Cognition,
19(2), 295–308.
Bowdle, B. F., & Gentner, D. (2005). The career of metaphor. Psychological Review, 112(1),
193–216.
Broadbent, D. E. (1958). Perception and communication. NY, US: Elmsford
Brône, G., & Coulson, S. (2010). Processing deliberate ambiguity in newspaper headlines:
Double grounding. Discourse Processes, 47(3), 212-236.
Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. New
York, NY: Cambridge University Press.
Bruner, J. S. (1990). Acts of Meaning. Boston: Harvard University Press.
17
Cameron, L. J. (2007). Patterns of metaphor use in reconciliation talk. Discourse & Society,
18(2), 197-222.
Cameron, L., & Deignan, A. (2006). The emergence of metaphor in discourse. Applied
linguistics, 27(4), 671-690.
Cameron, L. J., & Stelma, J. H. (2004). Metaphor clusters in discourse. Journal of Applied
Linguistics and Professional Practice, 1(2), 107-136.
Carston, R. (2010). XIII—Metaphor: Ad hoc concepts, literal meaning and mental images.
Proceedings of the Aristotelian Society 110(3), part 3, 295-321.
Evans, V., Bergen, B. K., & Zinken, J. (2007). The cognitive linguistics enterprise: An
overview. In V. Evans, & J. Zinken (Eds.), The cognitive linguistics reader (pp. 2–36).
London: Equinox.
Fauconnier, G., & Turner, M. (1998). Conceptual integration networks. Cognitive Science,
22(2), 133–187.
Fonagy, I. (1971). Double coding is speech. Semiotica, 3, 189–222.
Forgács, B. (2009). Verbal metacommunication – Why a metaphorical mapping can be
relevant? (In Hungarian) Hungarian Psychological Review, 64(3), 593-605.
DOI:10.1556/MPSzle.64.2009.3.8
Forgács, B. (2013). The right hemisphere of cognitive science. In Cs. Pléh, L. Gurova, & L.
Ropolyi (Eds.), New Perspectives on the History of Cognitive Science (pp. 129–141).
Budapest: Akadémiai Kiadó.
Forgács, B. (2014). Figures of Language in Cognitive Science in the Light of Figurative
Language Processing in the Brain. (Unpublished doctoral dissertation). Budapest
University of Technology and Economics, Budapest.
Forgács, B. (2020). An Electrophysiological Abstractness Effect for Metaphorical Meaning
Making. eNeuro 7(5), 1-12.
Forgács, B., Bardolph, M., Amsel, B. D., DeLong, K. A., & Kutas, M. (2015). Metaphors are
physical and abstract: ERPs to metaphorically modified nouns resemble ERPs to
abstract language. Frontiers in Human Neuroscience, 9(28).
Forgács, B., Bohrn, I., Baudewig, J., Hofmann, M. J., Pléh, Cs., & Jacobs, A. M. (2012). Neural
correlates of combinatorial semantic processing of literal and figurative noun noun
compound words. Neuroimage, 63(3), 1432–1442.
Forgács, B., Lukács, Á., & Pléh, Cs. (2014). Lateralized processing of novel metaphors:
disentangling figurativeness and novelty. Neuropsychologia, 56, 101–109.
18
Forgács, B., & Pléh, Cs. (2019). What Are You Thinking About Where? Syntactic Ambiguity
Between Abstract Arguments and Concrete Adjuncts in Hungarian, Modulated by
Concreteness. Psychology in Russia, 12(1), 67-78.
Freud, S. (1901/1989). The Psychopathology of Everyday Life. WW Norton & Company.
Gallese, V., & Lakoff, G. (2005). The brain's concepts: The role of the sensory–motor system
in conceptual knowledge. Cognitive Neuropsychology, 22(3–4), 455–479.
Gergely, Gy., & Pléh, Cs. (1994). Lexical processing in an agglutinative language and the
organization of the lexicon. Folia Linguistica, 28, 175–204.
Gernsbacher, M. A., Keysar, B., Robertson, R. R., & Werner, N. K. (2001). The role of
suppression and enhancement in understanding metaphors. Journal of Memory and
Language, 45(3), 433–450.
Glucksberg, S. (2003). The psycholinguistics of metaphor. Trends in Cognitive Sciences, 7(2),
92–96.
Glucksberg, S., & Haught, C. (2006). On the relation between metaphor and simile: When
comparison fails. Mind & Language, 21(3), 360-378.
Grady, J. (2005). Primary metaphors as inputs to conceptual integration. Journal of pragmatics,
37(10), 1595–1614.
Grice, H. P. (1975). Logic and conversation. In P. Cole, & J. L. Morgan (Eds.), Syntax and
Semantics, Vol. 3: Speech Acts (pp. 41–58). New York: Academic Press.
Kövecses, Z. (2002). Metaphor: A Practical Introduction. Oxford: Oxford University Press.
Lakoff, G. (1987). Women, fire, and dangerous things: What categories reveal about the mind.
Chicago: University of Chicago Press.
Lakoff, G. (2014). Mapping the brain's metaphor circuitry: Metaphorical thought in everyday
reason. Frontiers in human neuroscience, 8, 958.
Lakoff, G., & Johnson, M. (1980). Metaphors We Live By. Chicago: University of Chicago
Press. (Second edition 2002)
Lakoff, G. & Johnson, M. (1999). Philosophy in the Flesh: The Embodied Mind and its
Challenge to Western Thought. New York: Basic Books.
Leary, D. E. (Ed.). (1994). Metaphors in the History of Psychology. Cambridge: Cambridge
University Press.
Lee, J. J., & Pinker, S. (2010). Rationales for indirect speech: the theory of the strategic speaker.
Psychological Review, 117(3), 785–807.
19
Kutas, M., & Federmeier, K. D. (2011). Thirty years and counting: finding meaning in the N400
component of the event-related brain potential (ERP). Annual Review of Psychology,
62, 621-647.
Mahon, B. Z., & Caramazza, A. (2008). A critical look at the embodied cognition hypothesis
and a new proposal for grounding conceptual content. Journal of Physiology–Paris,
102(1), 59–70.
McGlone, M. S. (2007). What is the explanatory value of a conceptual metaphor? Language &
Communication, 27(2), 109–126.
Mithen, S. J. (1996). The prehistory of the mind: a search for the origins of art, religion, and
science. London: Thames and Hudson.
Nuessel, F. (2006). Figurative language: semiotics. In A. Barber & R. J. Stainton (Eds.),
Concise Encyclopedia of Language & Linguistics (pp. 446-459). Oxford: Elsevier.
Ortony, A., Schallert, D. L., Reynolds, R. E., & Antos, S. T. (1978). Interpreting metaphors and
idioms: Some effects of context on comprehension. Journal of Verbal Learning and
Verbal Behavior, 17(4), 465–477.
Pinker, S. (2010). The cognitive niche: Coevolution of intelligence, sociality, and language.
Proceedings of the National Academy of Sciences, 107(Supplement 2), 8993–8999.
Pinker, S., Nowak, M. A., & Lee, J. J. (2008). The logic of indirect speech. Proceedings of the
National Academy of sciences, 105(3), 833–838.
Pléh, Cs. (2000). Modularity and pragmatics: Some simple and some complicated ways.
Pragmatics, 10(4), 415–438.
Pléh, Cs. (2018). A kettős kódolás és a nem tudatos mozzanatok a nyelvben: Fónagy Iván
koncepciója [Dual Coding and unconscious moments in language: The conception of
Iván Fónagy]. IMÁGÓ Budapest, 7(3), 19-29.
Pléh, Cs. (2020). A jelentés jelentősége a pszichoanalízis örökségében [The significance of
meaning in the psychoanalytic tradition]. Lélekelemzés 1, 21-26.
Porat, R., & Shen, Y. (2015). Imposed metaphoricity. Metaphor and Symbol, 30(2), 77-94.
Pulvermüller, F., Shtyrov, Y., & Ilmoniemi, R. (2005). Brain signatures of meaning access in
action word recognition. Journal of Cognitive Neuroscience, 17(6), 884-892.
Pulvermüller, F. (2005). Brain mechanisms linking language and action. Nature Reviews
Neuroscience, 6(7), 576–582.
Seidenberg, M. S., Tanenhaus, M. K., Leiman, J. M., & Bienkowski, M. (1982). Automatic
access of the meanings of ambiguous words in context: Some limitations of knowledge–
based processing. Cognitive Psychology, 14(4), 489–537.
20
Sopory, P., & Dillard, J. P. (2002). The persuasive effects of metaphor: A meta-analysis.
Human communication research, 28(3), 382-419.
Sperber, D., & Wilson, D. (1986). Relevance: Communication and cognition. Oxford:
Blackwell. (Second edition 1995)
Sperber, D., & Wilson, D. (1998). The mapping between the mental and the public lexicon. In
P. Carruthers & J. Boucher (Eds.), Thought and language (pp. 184–200). Cambridge:
Cambridge University Press.
Sperber, D., & Wilson, D. (2008). A deflationary account of metaphors. In R. W. Gibbs (Ed.),
The Cambridge handbook of metaphor and thought (pp. 84–105) Cambridge:
Cambridge University Press
Swinney, D. (1979). Lexical access during sentence comprehension: (Re)consideration of
context effects. Journal of Verbal Learning and Verbal Behavior, 18(6), 645–659.
Szabó, R. (2015). Metaforák és szimbólumok: CG Jung szimbólumértelemzésének és a fogalmi
metaforák elméletének összevetése [Metaphors and Symbols: The Comparison of the
Symbol Interpretation of C. G. Jung and Conceptual Metaphor Theory]. Budapest:
Oriold és Társai Kiadó.
Tendahl, M., & Gibbs Jr, R. W. (2008). Complementary perspectives on metaphor: Cognitive
linguistics and relevance theory. Journal of Pragmatics, 40(11), 1823–1864.
Thuma O., & Pléh Cs. (1999). Ambiguous data on Hungarian ambiguity resolution. In M.
Prinzhorn, & J. Rennison (Eds.), Dressler Festschrift
Thuma, O., & Pléh Cs. (2000). Ambiguity and morphological decomposition in Hungarian. In
Ch. Schaner–Wolles, J. Rennison, & F. Neubarth (Eds.), Naturally! Linguistic essays in
honour of Wolfgang Ulrich Dressler on the occasion of his 60th birthday (pp. 128–142).
Torino: Rosenberg & Sellier.
Van Lancker Sidtis, D. (2006). Does functional neuroimaging solve the questions of
neurolinguistics? Brain and Language, 98(3), 276–290.
Wilson, D., & Carston, R. (2007). A unitary approach to lexical pragmatics: relevance,
inference, and ad hoc concepts. In Noel Burton-Roberts (Ed.), Pragmatics (pp. 3).
Palgrave McMillan.
Wilson, D., & Sperber, D. (2004). Relevance theory. In L. Horn, & G. Ward (Eds.), Handbook
of Pragmatics (pp. 607–632). Oxford: Blackwell
Wilson, D., & Sperber, D. (2002). Truthfulness and relevance. Mind, 111(443), 583–632.
