On the Role of Metaphor in Creative Cognition
Bipin Indurkhya
Cognitive Science Laboratory
International Institute of Information Technology
Hyderabad 500032, India
bipin@iiit.ac.in
Abstract. We consider some examples of creativity in a number of diverse
cognitive domains like art, science, mathematics, product development, legal
reasoning, etc. to articulate an operational account of creative cognition. We
present a model of cognition that explains how metaphor creates new insights
into an object or a situation. The model is based on assuming that cognition
invariably leads to a loss of information and that metaphor can recover some of
this lost information. In this model we also contrast the role of traditional
analogy (mapping based on existing conceptualization) with the role of
metaphor (destroying existing conceptualizations in order to create new
conceptualizations).
1 Introduction
Though there have been many approaches to characterize creativity [17] [25] [31], we
start with a simple approach which sees creativity as a process of generating a new
perspective on a problem or a situation. We are limiting ourselves to individual
creativity here, so the information resulting from this process need only be novel to
the cognitive agent, and we do not yet concern ourselves with creativity in a society.
Secondly, we do not consider usefulness of the generated information: it is sufficient
for us here that the information be novel to the agent. In fact, if the model presented
here is correct, it implies that there cannot be some domain-independent principle or
heuristic that would generate only (or largely) useful perspectives.
With these assumptions in place, the task we are undertaking is to propose a model
that articulates the role of metaphor in the creative process and also explains why
metaphor is so effective in generating new perspectives. In this model, we will also
compare the role of analogy with the role of metaphor, and argue that the two play
complementary roles in creative cognition.
The paper is organized as follows. In the next section we will present a few
examples to illustrate how creative insights are obtained in a few diverse domains.
Following this, in Section 3, we will present an account in which cognition is seen to
necessarily involve loss of some information, and in which metaphor becomes one of
the tools that makes it possible to recover some of this lost information. At the end of
this section we will also compare the role of analogy with the role of metaphor.
Finally, in Section 4, we will summarize the main points of this paper, and mention
future research directions.
51
2 Creativity in Cognition: Some Examples
We start by considering some concrete instances where a new insight or a new
perspective was generated. The examples are taken from a number of diverse domains
including art, legal interpretation, mathematics, and product development. At the end
of this section we will present a brief overview of the cognitive mechanisms
underlying creativity that have been proposed in the past research.
2.1 Creativity in Art
In a recent study, Okada et al [20] consider the evolution of artistic style and
creativity in the works of a Japanese artist Shinji Ogawa over several years. One
interesting point in this study is how the artist hit upon an idea that led to a series of
work: “[Shinji Ogawa] was a part-time teacher at a vocational-technical school of
media art. When he was preparing for a class, he accidentally erased part of a picture
on a computer screen by mistakenly pushing a keyboard button. At that moment, he
came up with the idea that if something very important and valuable suddenly
disappears, a new value may be generated and a new world could be created. With
this idea, he tried to create a new movie poster for Roman Holiday by erasing the
main actress, Audrey Hepburn, from the original poster. This was the beginning of the
artwork series, ‘Without You’.” [p. 194].
Though the authors chose to interpret this example in terms of analogical
modification, it resonates strongly with Piaget’s account of how new schemas emerge
through sensorimotor interactions with the environment. The example presented
above bears a strong resemblance to Piaget’s account of how a child brings a toy to
her mouth in order to suck, accidentally notices the bright color of the toy and starts
bringing toys near her face to look at them, eventually generalizing into a schema of
‘bringing objects to the face in order to look at them’ [21][22]. In Mr. Ogawa’s case,
he accidentally discovered the operation of ‘delete figure from a picture’, realized
artistic potential of it, and a new style of artwork was born. That the discovery was
made accidentally is not so relevant for our argument here, but what we would like to
emphasize is that the discovery resulted from the application of a familiar operation
(‘delete’) to a familiar object but in a novel way.
Interestingly, similar episodes occurred later as well in Mr. Ogawa’s career. Okada
et al note: “Mr. Ogawa happened to pick up a postcard at hand with old Western
scenery and drew a duplicate building next to an original one. Then he mailed it, as a
postcard, to a gallery owner. When he heard from the gallery owner telling him that
staff members of the gallery talked highly about his postcard, Mr. Ogawa decided to
start a new artwork series, ‘Perfect World’, in which he duplicates a person or a thing
in postcards or photographs of scenery.” [p. 195]
The operations of ‘delete’ and ‘duplicate’ are quite similar. In the framework of
Hofstadter [7], one could say that one operation slipped into a neighboring operation
to lead to another creative insight. Or one could see it in terms of a Piagetian schema
of related operations that are applied to a different class of objects. It is important to
underscore the ‘different’ part here. When handling photographs of famous
landmarks, people, etc., we could change the contrast, brightness level, perhaps apply
red-eye reduction tool, but we do not normally delete or duplicate objects, and much
less so if the object is the main theme of the photograph. In other words, we could say
that the creative insights resulted from applying a set of familiar operations to a set of
[also familiar] objects that are not usually associated with the operations. This is the
key point that makes metaphor an invaluable tool for generating creative insights,
something that we will keep reiterating in the rest of the paper.
52
2.2 Creativity in Legal Interpretation
Even though law is a domain that is normally not associated with creativity — for one
expects a straightforward application of legal principles and many judicial scholars
frown on any deviation from the literal interpretation of the legal text — in our
previous research [8] [10] we have found a number of situations where a new
perspective or insight was a key factor in a legal discourse. We briefly present two
such examples here.
The first example is taken from [8]. In Australia and England, when a married
couple divorces, the division of property was determined in large part by the old case
law of ‘Husband and Wife’ and by various Acts. These generally provided for
division according to economic value added into the marital assets. This was plainly
unjust where the husband had worked, while the wife cared for children and
maintained the household. In such situations, the standard decision was, until recently,
that the husband would get the lion’s share of the property. However in an example of
productive thinking, in Baumgartner v Baumgartner [(1987) 164 CLR 137] the High
Court of Australia introduced a principle from a completely different area of law and
held that the wife’s work placed into the house meant she had an equitable interest in
it. The husband, though legally the owner of the house, actually held part of it in a
‘constructive trust’ for his wife. This decision was soon followed by a number of
other similar decisions by other courts, and is now the standard approach.
This illustrates a novel application of the legal concept of constructive trusts to a
set of situations for which it was not originally intended, which resulted in a new way
of rendering judgment on them. Similarly, the decision of Lord Denning in the High
Trees case [Central London Property Trust Ltd v High Trees House Ltd [1947] KB
130], which modified contract law by introducing another equitable principle,
‘promissory estoppel’, is another example where a legal concept from a different area
was applied to deal with a problem in a domain for which it was not originally
intended. (See also [27].)
The second example [10] concerns the case of a hot-dog stand operator, who
claimed tax-deduction for the kitchen at home where hot-dogs were prepared [Baie,
74 T.C. 105 (1980)]. One argument made by B. was that her kitchen was a
manufacturing facility of the business. The judges remarked: “We find this argument
ingenious and appealing, but, unfortunately, insufficient to overcome the
unambiguous mandate of the statute.” [74 T.C. 110 (1980)]. The point to emphasize
here is that the category ‘manufacturing facility’, which is not normally associated
with this situation, is applied to the kitchen where hot dogs are prepared resulting in a
novel perspective.
To summarize, we see that the application of a legal concept to a domain or a
situation for which it was not originally intended, can sometimes result in a new way
of looking at the situation thereby leading to a novel judgment.
2.3 Creativity in Mathematical Reasoning
Consider George Cantor’s theory of transfinite numbers, in particular, his arguments
concerning the levels of infinity [1]. Two of his key proofs, namely that 1) rational
numbers have the same cardinality as natural numbers, and that 2) real numbers are
more numerous than natural numbers, can now be understood by a high-school
student. However, when originally proposed, they were considered very radical.
Many leading mathematicians at that time refused to accept his formalization of set
theory and its implications for infinite sets. Yet, Cantor’s insights were derived from
applying the operation of making one-to-one correspondence, which was already well
known for finite sets for hundreds of years, to infinite sets. In addition, he used a
53
particular way of arranging infinite numbers in an array and counting them in such a
way that two (or more) infinite dimensions can be mapped onto a single dimension of
infinity. A somewhat different operation applied to a similarly arranged twodimensional
layout of numbers led to his famous diagonal argument, where he
showed that certain sets cannot be put in a one-to-one correspondence with natural
numbers.
To emphasize, we see again that the application of familiar operations to a different
sets of objects resulted in a novel perspective. For indeed, the theorems and proofs
discovered by Cantor revealed a whole new aspect of numbers and opened a fresh
chapter in mathematical research.
2.4 Creativity in Product Development
Consider a case study described in Schön [24] where a product development team was
faced with the problem of figuring out why synthetic-fiber paintbrushes were not
performing as well as natural-fiber paintbrushes, and to improve their performance.
The members of the team tried many ideas — for instance, they noticed that the
natural fibers had frayed ends, and they tried to have synthetic fibers with frayed ends
too — but without success. The breakthrough came when one member of the team
suggested that the paintbrush might work as a pump. This idea was initially
considered quite shocking, for a paintbrush and a pump were thought to be very
dissimilar. Yet, in trying to make sense of the analogy, a new ontology and structure
for the paintbrush was created. In this new representation, the paint was sucked in the
space between the fibers through capillary action, and when the fibers were pressed
against the surface to be painted, the curvature of the fibers caused a difference in
pressure that pumped out the paint from the space between the fibers onto the surface
to be painted. From this new ontology, when the synthetic-fiber and natural-fiber
paintbrushes were compared, it was found that the synthetic fibers bent at a sharp
angle against the surface, whereas the natural fibers formed a gradual curve. Thus,
juxtaposition with pumping caused a new perspective to be created on the process of
painting and paintbrush.
There are many other such examples [Gordon 1961] where seeing one familiar
object as another familiar object, but one that is not normally associated with the first
object, led to a new perspective and eventually to solving a difficult problem.
2.5 Cognitive Mechanisms of Creativity
So far we have seen a number of examples where a set of operations or concepts are
applied to an object or a situation with which they are not normally associated,
resulting in a novel perspective. Perhaps not surprisingly, such mechanisms have been
noted and studied in the past by various researchers, and they have been known under
different labels. Here we summarize a few major veins of this research.
Making the Familiar Strange: Gordon and his colleagues [6] studied creative
problem solving in real-life situations for many years, and found that one way to get a
new perspective on the target problem is to look at in a strange way. The mechanism
is they proposed is to juxtapose the target problem or object with a completely
unrelated object or situation.
Displacement of Concepts: Schön [24] emphasized that in order to get a new insight
about a concept, it needs to be displaced, that is, put in the context of other unrelated
concepts. He emphasized that the most important step in problem solving is problem
54
setting, that is how the problem is stated and viewed, and metaphors play a key role in
this step.
Bisociation: Koestler [18] coined this term to emphasize that the pattern underlying a
creative act is the perception of a situation or an idea in two self-consistent but
habitually incompatible frames of reference.
Lateral Thinking: Edward de Bono [4] contrasted vertical thinking with lateral
thinking. In the former, one starts with some assumptions and explores their
implications deeper and deeper. But in lateral thinking, the goal is to look at the
problem in different ways so that the familiar assumptions one makes about it can be
questioned and perhaps a new set of assumptions can be brought in.
Estrangement. Rodari [23] focused on creativity in inventing stories, and proposed
many practical methods that stimulate imagination and creativity in children (and in
adults). Many of his methods rely on random juxtaposition of concepts. One
mechanism he emphasizes as the first step in creating riddles is estrangement, where
you are asked to see the object as if for the first time. In other words, instead of seeing
the object in terms of the familiar categories it naturally evokes, you are asked to
consciously block this evocation and try to view the object as if it is a strange object
you are seeing for the first time.
Conceptual Blending. Fauconnier and Turner [5] analyzed how people combine
perceptual, experiential and conceptual aspects of different concepts subconsciously
to generate new insights.
Though each of these approaches has its own peculiarities, they all emphasize that in
order to get a new insight about an object or situation, we need to get away from, or
break, its existing conceptualization. In this task, viewing the object in terms of (or
juxtaposing it with) another unrelated object can be a key step.
3 An Account of Creativity in Cognition
We saw numerous examples in the last section that show that to get a new perspective
on an object or a situation, an approach that often works is to apply operations that are
not normally associated with that object, or to see that object as another unrelated
object. Here we will propose a model to explain why this process works as it does.
3.1 Cognition and Loss of Information
Here we argue that every act of conceptualization (or cognition) invariably involves
some loss of information. When we choose to label an object as a ‘chair’ numerous
specific details of the object, like its color, the material it is made of, shape, etc. are
all lost. Of course, we could make our conceptualization of the object more specific
— it is a red chair, made of teak, with a high back, and so on — but no matter how
detailed the conceptual representation is made, there is always some aspects of the
object that are excluded, and it is these excluded aspects that constitute the
information lost in the conceptualization. (This precisely is the theme of a short story
Del Rigor en la Ciencia (On Exactitude in Science), by Jorge Luis Borges and Adolfo
Casares.)
55
Whenever this lost information becomes crucial to solving the problem, then the
existing representation becomes hopelessly inadequate. In the paintbrush example
presented above, the information about the spaces between the brush fibers etc. was
discarded in the then existing model of painting, so no matter how much and how
hard the product development team tried, the problem could not be solved. It was
necessary to change the representation or the conceptualization of the object.
3.2 Interaction with the Environment Through Actions and Gestalt Projection
If the hypothesis presented above is correct, namely that some information is
invariably lost in conceptualization, the next question is how can we recover, at least
partially, this information. Here we assume that we do not have the God’s eye view of
the world, meaning that we do not have another way to access the object except
through the cognitive agent. This may seem a technicality, but it is a very crucial
point, so let us elaborate a bit. In a computer simulation or a model, one can posit a
very rich and detailed representation of the object, and then show how a
conceptualization picks out some aspects of this rich representation, while ignoring
others. For example, the rich representation of a chair may include its material, shape,
color, weight, and so on, but the conceptualization can only include legs, seat and
back. However, for us here, the rich representation is not available, for if it were, it
would be just another conceptualization, and there will still be some lost information.
In other words, that our conceptual representation of an object does not include all the
information about the object is like an existence proof: we can argue about its
existence but we cannot say what it is. So the key question is how can we become
aware of this lost information, and how can we recover at least some of it.
Piaget’s action-oriented approach provides a possible way to addressing this
question. Piaget argued that an object is relevant or meaningful to a cognitive agent in
only as far as how the agent may act on it. Thus, a ball is something that a baby might
roll, kick, squeeze, and so on. In this approach, novel aspects of an object may be
revealed when the agent carries out new actions on it. Moreover, the actions can be
internalized actions, which are called operations in Piaget’s framework. In our earlier
work [13], we have used the term gestalt projection to emphasize that it is not just
individual operations, but a network of operations, namely a schema or a gestalt, that
are projected onto the internalized object or situation. In other words, a cognitive
agent can get more information about an object or a situation by projecting a different
gestalt, or a different set of operations onto it.
3.3 Metaphor: A Tool for Generating New Information about the Environment
So far we have argued that all conceptualization involves some loss of information,
and that some of this lost information may be recovered by projecting a different
gestalt or a different set of operations onto it. But this is essentially what a metaphor
does! By inviting us to see one object as another, we are forced to project the
conceptual organization of the second object (usually referred to as the source) onto
the experiences, images etc. of the first object (usually referred to as the target). Thus,
metaphor can be a useful and powerful tool to get new information about the
environment.
This is essentially the crux of the arguments made by Turbayne [28]. He argued
that though we can understand the world only through some metaphor or other, we
enrich our understanding by viewing the world through two different metaphors.
What a metaphor does is essentially give us an alternate conceptualization of the
target. While this alternative conceptualization also loses some information (as all
56
conceptualizations do), the point is that this loses a different kind of information than
what was lost in the original conceptualization, and taking them together we recover
some of the lost information. (See also [14] and [24].) In this way, metaphor becomes
a potent cognitive tool for generating creative insights.
An interesting consequence of this view is that there cannot be any a priori
criterion for determining which metaphors will be useful for a particular problem or to
achieve a particular goal. If it is the missing information that is the key to solving the
problem, then the existing conceptualization is hopelessly inadequate in pointing the
way to recovering this information. The metaphor approach presented here is
essentially a trial-and-error method that makes no promise to deliver even in a
probabilistic sense. We elaborate on this further below.
3.4 To Analogize or Not To Analogize
If we follow the arguments presented above, analogies, in their traditional sense at
least, turn out to be an anathema to creativity. The reason is that analogies are based
on mapping the structure or attributes of the source, to the structure and attributes of
the target. So an analogy, which is based on the existing conceptualization of the
source, will retrieve sources that are similar to the structure, thereby further
strengthening the existing conceptualization of the target. But if the problem could not
be solved because of the missing information, then an analogy-based approach will
not be very useful.
Yet, analogy has also been recognized as a key mechanism of creativity [2] [6] [7]
[18] [19] [20]. One must distinguish between two modes of analogy here though. On
one hand, analogy refers to “seeing one thing as another”, which is essentially the
same as how we have characterized metaphor above. The other use of the term
analogy refers to the process whereby the structure and the attributes of the source are
mapped to the target. It is this latter mechanism that seems contrary to creativity
according to the view presented here, and so it needs some further elaboration.
The cognitive structures (categories and conceptualizations) that naturally evolve
through a cognitive agent’s interaction with the environment reflect the priorities of
the agent. The information that is retained in the conventional conceptualization is the
one that has been useful to the agent (or to its ancestors) in the past, and the lost
information may not be very relevant. So as long as one stays in the familiar domain
(in which the conventional conceptualizations are very useful), and the problem does
not require the lost information, reasoning from conventional operations and
conceptualizations may be very efficient. Indeed, many of the case studies that show
effectiveness of analogy in creative problem solving either stay within the same
domain, or they use a source that is already similar to the target in a way that leads to
a successful solution to the problem. However, as soon as the problem becomes
different requiring new information, analogy becomes a hindrance, and the metaphor
approach is called for. (See also [6] and [9].)
To put this in another way, metaphor in the making-the-familiar-strange mode is a
cognitively expensive operation, with no a priori guarantee if it will succeed, or when
it will succeed. Therefore, this is used sparingly, and only when other avenues (like
reasoning from analogy) have been tried out and were not successful.
3.5 Implications for Computational Modeling
The account of creativity and cognition articulated here has a number of implications
for computational modeling, and we will briefly highlight a few major ones. First of
all, traditional approaches based on mapping existing symbolic representations clearly
57
have limitations [3] as far as creativity is concerned. They do capture a certain aspect
of creativity in noticing new connections between existing knowledge, and in
importing novel hypotheses from the source to the target, but they do not produce a
paradigm shift of Kuhnian kind. In this regard, models based on corpus-based
analyses and distributed representations seem more promising [26] [29] [30], but so
far they are limited to linguistic metaphors.
Another approach is to model the representation building process itself so that new
representations can emerge through an interaction of concept networks and low-level
object details that are available through sensory system or through imagination [7]
[13]. This comes closest in spirit to the cognitive mechanisms underlying metaphor
that we mentioned above in Sec. 2.5, for the creative insights emerge from applying a
concept to an object (or a low-level representation of it) that is not habitually
associated with it. In our earlier work, we have formalized this process [9] [12], and
have applied it to model creativity in legal reasoning [10], but clearly much more
work remains to be done. Moreover, in real-life, a number of different cognitive
processes may act in consort to generate a creative insight, modeling of which may
require hybrid architectures [19].
4 Conclusions and Future Research
We have articulated an account of cognition here in which cognition necessarily
involves loss of some information. Creativity essentially lies in recovering some of
this lost information, and metaphor play a fundamental role in this process. This,
however, is a cognitively expensive operation. In many situations, such a novel
perspective is not needed, so other problem-solving methods, including analogy, may
be more efficient.
Following the ideas outlined in our earlier research [11], it is possible to build a
number of computer-based creativity-support systems to reduce the cognitive load on
the agent in generating novel ideas and perspectives, or to stimulate their imagination
in coming up with more creative ideas. We have demonstrated this point in a storytelling
system that was designed and implemented earlier [16]. Currently we are
working on designing and implementing another system to retrieve and display pairs
of pictures that are based on perceptual similarities but are conceptually very
different, in order to stimulate the user’s creativity [15].
<references_biblio/>
References
1. Bell, E.T.: Men of Mathematics. Simon & Schuster, New York (1937).
2. Bonnardel, N.: Towards Understanding and Supporting Creativity in Design: Analogies in a
Constrained Cognitive Environment. Knowledge-Based Systems 13, 505–513, (2000).
3. Chalmers, D.J., French, R.M., Hofstadter, D.R.: High-Level Perception, Representation, and
Analogy:A Critique of Artificial Intelligence Methodology. Journal of Experimental and
Theoretical Artificial Intelligence 4 (3), 185–211, (1992).
4. de Bono, E.: New Think: The Use of Lateral Thinking in the Generation of New Ideas.
Basic Books, New York (1975).
5. Fauconnier, G., Turner, M.: The Way We Think: Conceptual Blending and the Mind’s
Hidden Complexities. Basic Books, New York (2002).
6. Gordon, W.J.J.: Synectics: The Development of Creative Capacity. Harper & Row, New
York (1961).
7. Hofstadter, D., The Fluid Analogies Research Group: Fluid Concepts and Creative
Analogies. Basic Books, New York (1995).
58
8. Hunter, D., Indurkhya, B.: ‘Don’t Think, but Look’: A Gestalt Interactionist Approach to
Legal Thinking. Proceedings of the Workshop on Advances in Analogy Research, pp. 345–
353. Sofia, Bulgaria (1998).
9. Indurkhya, B.: Metaphor and Cognition. Kluwer Academic Publishers, Dordrecht, The
Netherlands (1992).
10. Indurkhya, B.: On Modeling Creativity in Legal Reasoning. In Proceedings of the Sixth
International Conference on AI and Law, pp. 180-189. Melbourne, Australia (1997).
11. Indurkhya, B. Computers and Creativity. Expanded version of the keynote speech ‘On
Modeling Mechanisms of Creativity’ delivered at Mind II: Computational Models of
Creative Cognition, Dublin, Ireland (1997): http://www.iiit.ac.in/~bipin/
12. Indurkhya, B.: An Algebraic Approach to Modeling Creativity in Metaphor. In C.L.
Nehaniv (ed.) Computation for Metaphors, Analogy and Agents. LNAI, Vol. 1562, pp. 292–
306. Springer-Verlag, Berlin (1999).
13. Indurkhya, B.: Emergent representations, interaction theory, and the cognitive force of
metaphor. New Ideas in Psychology 24(2), 133–162 (2006).
14. Indurkhya, B.: Rationality and reasoning with metaphors. New Ideas in Psychology 25(1),
16–36 (2007).
15. Indurkhya, B., Kattalay, K., Ojha, A., Tandon, P.: Experiments with a Creativity-Support
System based on Perceptual Similarity. In H. Fujita and I. Zualkernan (eds.) New Trends in
Software Methodologies, Tools and Techniques, pp. 316–327. IOS Press: Amsterdam
(2008).
16. Ishii, Y., Indurkhya, B., Inui, N., Nose, T., Kotani, Y., Nishimura, H.: A System Based on
Rodari’s ‘Estrangement’ Principle to Promote Creative Writing in Children. Proceedings of
Ed-Media & Ed-Telecom 98, pp. 1685–1686, Freiburg, Germany (1998).
17. Kaufman, J.C., Beghetto, R.A.: Beyond Big and Little: The Four C Model of Creativity.
Review of General Psychology 13(1), 1–12 (2009).
18. Koestler, A.: The Act of Creation. Hutchinsons, London (1964).
19. Nersessian, N.J., Chandrasekharan, S.: Hybrid Analogies in Conceptual Innovation in
Science. Cognitive Systems Research 10 (3), 178–188, (2009).
20. Okada, T., Yokochi, S., Ishibashi, K., Ueda, K.: Analogical Modification in the Creation of
Contemporary Art. Cognitive Systems Research 10 (3), 189–203, (2009).
21. Piaget, J.: The Origins of Intelligence in the Child (M. Cook, Trans.) Penguine, New York
(1977). (Original work published 1936).
22. Piaget, J.: Play, Dreams and Imitation in Childhood (C. Gattegno & F.M. Hodgson, Trans.)
W.W. Norton, New York (1962). (Original work published 1945).
23. Rodari, G.: The Grammar of Fantasy (J. Zipes, Trans.) Teachers & Writers Collaborative,
New York (1996).
24. Schön, D.A.: Displacement of Concepts. Humanities Press, New York (1963).
25. Sternberg, R.J.: Creating a Vision of Creativity: The First 25 Years, Psychology of
Aesthetics, Creativity, and the Arts, Vol. S (1), 2–12 (2006).
26. Sun, R.: A microfeature based approach towards metaphor interpretation. Proceedings of the
14th IJCAI, pp. 424–429. Morgan Kaufman, San Francisco (1995).
27. Teeven, K.M.: Origins of Promissory Estoppel: Justifiable Reliance and Commercial
Uncertainty Before Williston's Restatement. University of Memphis Law Review, Spring
2004.
28. Turbayne, C. M.: The Myth of Metaphor. University of South Carolina Press, Columbia
(SC) (1970).
29. Veale, T.: Re-Representation and Creative Analogy: A Lexico-Semantic Perspective. New
Generation Computing, 24 (3), 223-240 (2006).
30. Veale, T., Hao, Y.: A Fluid Knowledge Representation for Understanding and Generating
Creative Metaphors. Proceedings of COLING 2008, pp. 1–5. Manchester (2008).
31. Ward, T.B.: Creative Cognition as a Window on Creativity. Methods 42, 28–37 (2007).