Whence is Creativity?
Bipin Indurkhya (bipin@agh.edu.pl)
Department of Computer Science, AGH University of Science and Technology, Cracow, Poland
Cognitive Science Lab, International Institute of Information Technology, Hyderabad, India
Abstract
We start with a critical examination of the traditional
view of creativity in which the creator is the major
player. We analyze many different examples to point
out that the origin of all different creativity scenarios is
rooted in the viewer-artifact interaction. To recognize
this explicitly, we propose an alternative formulation of
creativity by putting the viewer in the driver’s seat. We
examine some implications of this formulation, especially
for the role of computers in creativity, and argue
that it captures the essence of creativity more accurately.
 Introduction: Traditional View of Creativity
In a typical creativity scenario, there is a creator, a product
and the audience. The creator creates the product, and the
audience appreciates it. The creativity is almost always
imputed to the creator. One could talk about a creative
process, but that is again associated with the creator. The
audience plays a role, and has been dubbed the field by
Csikszentmihalyi (1996), but somewhat indirectly, and
even then, one usually makes a distinction between a popular
artist and a creative artist, which are not synonymous.
This framework is questioned in this paper by analyzing
a number of creativity scenarios and tracing the root factor
that allows us to dub them creative. It is not the first time
that these issues are being raised, for they are well known
in the literature. But it may be the first time we are bringing
them all together to suggest that perhaps there is something
fundamentally wrong with this view of creativity. We
then propose an alternative formalism for creativity by
looking at it from the audience’s point of view. Though
this may seem almost heretical at first sight, we argue that
it provides a more accurate framework to address various
issues surrounding creativity, including the role of computers
therein.
Analysis of Some Creativity Scenarios
We present here several creativity scenarios and analyze
them to identify the root cause as to why they are labeled
creative.
Case of Creative Individuals
If we try to think of creative people, who comes to mind?
Perhaps Einstein, Mozart, Michelangelo or Leonardo da
Vinci. In the modern times, we might think of Steve Jobs.
But what do we mean when we say that they are creative?
Perhaps music came naturally to Mozart. In a letter to
his father on Nov. 8, 1777, he wrote: “I cannot write in
verse, for I am no poet. I cannot arrange the parts of speech
with such art as to produce effects of light and shade, for I
am no painter. Even by signs and gestures I cannot express
my thoughts and feelings, for I am no dancer. But I can do
so by means of sounds, for I am a musician.” However,
what makes his work great is because of the way people
have responded to his music over more than two centuries.
(See also Kozbelt 2005; and Painter 2002.)
Steve Jobs has sprouted and nurtured many creative
ideas, but again it is how people responded to the artifacts
based on his ideas, like Mac, iPod, and iPhone, that is the
key factor in his having become an icon of technological
innovation and creativity. And one could easily dispute
whether he was creative when it came to his clothes.
Einstein’s brain was preserved after his death so that
people can study it to get any clues about the biological
basis for creativity. But here also it is the impact of his
theory of relativity, and its eventual acceptance by the scientific
community that was a key factor in him becoming
an icon of scientific creativity of the twentieth century.
Moreover, Einstein was also dogmatic at times, perhaps the
most famous case being his rejection of Alexander Friedmann’s
expanding universe hypothesis (Singh 2004).
Needless to say, we are not trying to argue that these
people were not creative, but merely to point out that they
were creative some of the times, and in some areas; and,
more importantly for the discussion here, that we determine
when they were creative based on how the audience
responded to their ideas or artifacts.
Creativity in Mentally Different Individuals
Take the case of Stephen Wiltshire, discussed in Sacks
(1995). He has an amazing ability to draw a landscape
from memory after seeing it only once. Though he is diagnosed
with autism, his work is highly regarded both by
critics and general population. He was awarded Member of
the Order of the British Empire for services to art in 2006.
So he is no doubt a very creative person, no matter which
criterion one chooses to apply.
But let us think about it a minute. What do we mean by
saying that he is creative? His work has a certain style,
level of details that most people cannot reach, aesthetic
appeal, and all that. As with Mozart, we can go further and
say that perhaps this is the way he expresses himself naturally:
just like you and I might describe what we did on our
last summer vacation, he draws fantastic landscapes. The
International Conference on Computational Creativity 2012 62
landscapes are fantastic to us, his audience, and that is the
crucial factor in his being recognized as a creative genius.
We can now throw in here examples of people with
schizophrenia or brain damage, savants or manicdepressive
people, and so on (Sawyer 2006). When these
people produce work that is considered creative, it is exclusively
the evaluation of the audience that is the key factor
in this judgment. For many of them, this is their mode
of being, and it could not have been otherwise. Often the
intention is missing as well. (See also Abraham et al. 2007;
Glicksohn 2011.)
Cultural Creativity
In many cultures, art is practiced as a group activity. For
example, Maduro (1976) provides a study of Mewari painting
community in a village in the Rajasthan province of
India. It describes a strictly hierarchical group with each
artist belonging to one of the laborers, master craftsmen or
creative artists class. They mostly copy existing forms and
patterns, with rarely an innovation, at least in the way it is
considered in the Western art.
Sawyer (2006) discusses this and many other examples
to argue that for such scenarios, one needs to take into account
cultural context to evaluate creativity, and novelty
may be neither necessary nor sufficient. However, the
audience response is still a key factor. Notice that the culture
itself can be an audience.
Computer and Creativity
We now consider the scenario at the other extreme, when
technically there is no creator and the intent is missing.
Even though the last two or three decades have seen a
steady progress in the development of computer systems
that produce artifacts in the domain of visual art (Cohen
1981; McCorduck 1991), music (Chordia & Rae 2010;
López et al. 2010; Monteith et al. 2010), literature (Kurzweil
2001, Pérez y Pérez et al. 2010); and so on, generally
they have received a negative press as regard to their creativity:
computers cannot have emotions, programs do not
have intents, creativity cannot be algorithmic, etc. etc.
(Boden 2009; Sawyer 2006). In fact, such views blatantly
expose the implicit assumptions underlying creativity:
namely that it crucially needs a creator with emotions, intentions,
and such. However, we have just seen a number
of scenarios above involving humans where the intentions
and emotions are missing and, even when they are there,
what determines whether something is creative or not is the
audience response. So why should we not apply the same
yardstick for computer-generated artifacts?
Creativity in Viewer-Artifact Interaction
In all the scenarios above, we have seen that it is the audience
response that determines whether an action (of the
creator or the group), a process, or a product is creative. So
suppose we drop the pretense, stop being apologetic about
it, and embrace this view formally: We define creativity as
the process by which a cognitive agent acquires a novel
perspective that is useful (or meaningful) to it in some way
by interacting with an object or a situation.
There are two aspects of this definition that need to be
emphasized here. One is that we are taking completely the
audience’s perspective here, so the creator is not even mentioned.
Needless to say, this is not the first time that such a
position is articulated. Barthes’ (1977) concluded: “We
know that to restore to writing its future, we must reverse
its myth: the birth of the reader must be ransomed by the
death of the Author,” and he traced this view to even earlier
scholars. Moreover, even when one does not take this
extreme position, most accounts of creativity do acknowledge
the role of audience (Cropley et al. 2011; Csikszentmihalyi
1996; Horn & Salvendy 2006; Maher 2010). Our
aim here is to explore the implications of this authorless
view of creativity, especially for computational systems.
Secondly, both the novelty and the usefulness are defined
from the agent’s personal point of view. So in this
way, this definition refers to little-c creativity, or Pcreativity
(Boden 1990; Kaufman & Beghetto 2009). However,
as we will see soon, it can be extended to Big-C or Hcreativity.
Implications of the Proposed View
We will now discuss how many intuitive notions associated
with creativity can be rooted in the formulation proposed
above. We start with the four Ps of creativity (Runco
& Kim 2011).
Product: Potential of Artifacts for Creativity
If an agent can interact with an object to get a novel and
useful perspective, we can impute creative potential to that
object. If many people can get a novel and useful perspective,
it should be emphasized that the perspective that each
agent gets need not be the same, or need not be useful in
the same way. So, for instance, different agents may see a
work of modern art in various ways, and find it meaningful
in different ways, and some may not see anything at all.
Though most accounts of creativity incorporate audience
response to some extent, the extreme view we are examining
here would allow Oracle readings of tea leaves and
such as creative interactions. (See also Indurkhya 2007.)
Just to contrast, in the traditional view, such objects lack a
creator so, by association, lack creativity as well. This is a
common argument to deny creativity to computer systems.
Process: Creativity in Generating Artifacts
Now we can take another step back and consider the process
of generating a creative artifact. In other words, we
need to consider the process of generating an artifact with
which a viewer can interact and get a novel and useful perspective.
So the viewer is always hovering in the background,
and has a significant impact on whether the process
is really generating an ordinary artifact as opposed to a
creative artifact.
The main implication for modeling the generative aspect
of creativity is that we cannot pursue it without considering
International Conference on Computational Creativity 2012 63
the audience, and making some assumptions about how
they are likely to interact with the generated artifacts.
Person: Creativity of an Individual/Group
We can step back some more and consider the creativity of
an individual or a group of individuals. Here we are looking
at the ability of the individual to generate artifacts with
which viewers can interact and get novel and useful perspectives.
So the viewer is again in the background and is
playing a critical role. Moreover, even though we speak of
this person or that person as being creative, we are really
focusing on certain artifacts that they have generated in
their career, which have given their audience some novel
and useful perspective.
The implication of this is that though we can certainly
study personality traits of certain individuals who generated
some artifacts during their career that were deemed
creative by the audience, it does not follow that those personality
traits in a different culture, in a different context
and with a different audience will necessarily result in the
generation of artifacts that would also be considered creative.
This point is highlighted in the essay ‘Late Bloomers’
(Gladwell 2009), where early geniuses are contrasted with
late bloomers. The relevant point here is that whether a
work is accepted by the audience or not does not depend
much on whether it was produced early or late in the career,
but on the kind of work and the context and the culture
in which it was produced.
Press: Context, Culture and H-Creativity
Press refers to the environmental factors that have an influence
on the generation of the artifact; but that is taking the
traditional perspective, where the focus is on the creator. If
we are putting the viewer in the drivers’ seat, then an
analogous set of environmental factors can be identified
that determine how a work is received by the viewer, and
whether it is successful or not.
Let us first consider the artifact interaction with an individual
viewer. Clearly, the context in which a viewer interacts
with the artifact can have a major influence in what
perspective is gleaned from it, and whether it is novel or
meaningful. The most classic example of this might be
Marcel Duchamp’s Fountain, which was a urinal turned
around. (See also ‘When is Art?’ in Goodman 1978.) There
is also the effect of the viewer’s background knowledge:
when one views the Parthenon in Athens, one’s knowledge
of the history and culture of ancient Greek certainly effects
one’s perceptions and aesthetic experience.
Moving to larger groups and societies, there are many
instances when a novel and potentially useful idea was not
successful when introduced in one context, but the same
idea was a big hit in another context. We mentioned above
the example of Alexander Friedmann’s expanding universe
hypothesis, which was rejected when introduced because
of Einstein’s influence but was widely heralded later.
Wegener’s (1966) theory of continental drift suffered a
similar fate when it was first introduced in 1915, even
though that was no fault of Einstein.
There are also cases where the theory, although novel
and carefully worked out, never received acceptance: for
instance, Velikovsky’s theory, which hypothesized Earth’s
encounters with a large comet expelled from Jupiter and
provided explanations for many biblical events (Casti
1989, pp. 7–10).
History of marketing and product development also provides
many such examples that are studied in business
schools all over. Gladwell (2009), for instance, recounts
Jim Wigon’s not-so-successful odyssey to develop creative
ketchup flavors, and contrasts this with mustard and spaghetti
sauce, for which similar efforts were more readily
accepted by the consumers.
The implication of all this simply is that one needs to
study all these contextual factors that make an idea or an
artifact novel and meaningful, and thereby eligible for the
‘creativity’ label. But this is essentially what is called Hcreativity:
novelty and usefulness for a culture or society.
We should emphasize here that this novelty and usefulness
with respect to a culture is not the same as popularity. Certain
ideas or artifact can be popular in a society without
being considered novel (by the members of the society
themselves), and vice versa.
Who is Creative?
This is one question that is often asked with regard to artificial
creativity systems. For example, who is being creative
when the computer program Aaron generates a painting
in which many people see some aesthetic value? Is it
the program? Is it the programmer? In many complex computational
systems, the programmer cannot see all the
consequences of what their system can generate, and can
be quite surprised by the artifacts it produces.
We would like to argue that this question is meaningless
in the framework of creativity we are proposing here. Just
to get away from the computational system scenario, consider
a painting by a schizophrenic person that many people
find interesting and insightful. Now obviously, the
schizophrenic person is the creator of the painting. But is
she or he creative? How can we determine this? Perhaps it
is a natural way for them to express themselves; they may
not see what all the fuss is about the artifact they created.
In other words, if we zoom out and look at the larger picture,
they are creating artifacts that many people find insightful,
and so in that sense we can ascribe them creativity
as explained above. But if we zoom in on the processes by
which they generate these artifacts, where is the creativity?
They may not be trying to generate some aesthetically
pleasing object, and may not even be aware of any audience.
The point is that there is nothing distinctive about the
generation process itself that we can label it as creative.
Creativity and Computational Systems
Once we acknowledge that it is meaningless to ask who is
being creative, the stigma surrounding the potential creativity
of computational systems recedes away. The goal
becomes simply to create artifacts that give them some
novel and meaningful perspectives.
International Conference on Computational Creativity 2012 64
This seemingly small shift of focus has far reaching consequences,
and our society is already moving towards it.
To start with, modeling the audience, their cultural tastes
and preferences, their cognitive processes that influence
their response to novel stimuli, and so on, becomes very
crucial. In the last 10-15 years, research in neuroscience
has revealed that at least some of our aesthetic values are
hardwired in the structure of the brain (Ramachandran &
Hirstein 1999; Zeki 2000). Then to add to that, machine
learning techniques can learn about the cultural preferences
of an audience based on the past data. For instance,
Ni et al. (2011) trained their program with the official UK
top-40 singles chart over the past 50 years to learn as to
what makes a song popular. A program like this might successfully
predict, for instance, the winner of the future
Eurovision competitions.
To reiterate a subtle point, creativity is not the same as
popularity. So to be able to predict whether a song, or a
book, or a video will become popular (Szabo & Huberman
2010) is not the same thing as evaluating their creativity.
Nonetheless, we expect that similar techniques, perhaps
with some adaptations, are more likely to yield the key to
creativity.
Going one step further, once audience-based models of
creativity are articulated, we can design, implement and
experiment with computational systems that generate artifacts
that are more likely to appeal to the audience, both
with respect to their novelty and meaningfulness. One
could even argue that computational systems are more ideally
suited than humans to explore this space of creative
possibilities (Harry 1992; Indurkhya, to appear).
Designing Creativity-Support Systems
A related issue is how to stimulate and enhance creativity
in people, if it is possible at all. Indeed, a number of approaches
have been proposed and tried out over the years
(de Bono 1975; Gordon 1961; Holstein 1970; Rodari 1996;
Shapira & Liberman 2009). One key observation that recurs
in many of these studies is that trying to associate unrelated
objects or situations stimulates creativity (Indurkhya
2010). In our past research, we have explored some
approaches to design creativity-support systems based on
this observation (Indurkhya 1997; Indurkhya et al. 2008;
Ishii et al. 1998), but much more remains to be done.
Conclusions
Though most of the research on computational creativity
has implicitly assumed that the creative value is in the artifact,
they have been sort of apologetic about it. For example,
Colton (2008) argues that it is not enough to generate
an interesting or creative artifact, but one must also take
into account the process by which the artifact was generated.
Krzeczkowska et al. (2010) took pains to project
some notion of purpose in their painting tool so that it
might be perceived as creative.
In this paper we have sought to not only drop this veil of
apology, but move to the other extreme by proposing a
formulation of creativity that puts the onus on the viewer
by characterizing it as the process of getting a new and
meaningful insight about an object or situation.
We have argued that this formulation reflects more accurately
what actually goes on in the whole creativity cycle.
Moreover, other situations, like when we deem an artifact
or an individual as being creative, we are really implicitly
relying on the viewer-artifact interaction to make this
judgment. Therefore, creativity of an agent and that of an
artifact are best seen as derived concepts based on our proposed
formulation of creativity.
We hope that this will stimulate further discussion about
the nature of creativity and, more importantly, will generate
new approaches to the design and development of
computational creativity systems.
<references_biblio/>
References
Abraham, A., Windmann S., McKenna P., & Güntürkün,
O. 2007. Creative Thinking in Schizophrenia: The Role of
Executive Dysfunction and Symptom Severity. Cognitive
Neuropsychiatry 12(3), 235–258.
Barthes, R. 1977. The Death of the Author. In R. Barthes,
Image Music Text (trans. S. Heath), London: Fontana.
(Original work published 1967.)
Boden, M. A. 1990. The Creative Mind, London: Weidenfeld
and Nicholson.
Boden, M. A. 2009. Computer Models of Creativity, AAAI
AI Magazine 30(3), 23-33.
Casti, J. L. 1989. Paradigms lost: Images of man in the
mirror of science. New York: William Morrow & Co.
Chordia, P., & Rae, A. 2010. Tabla Gyan: An Artificial
Tabla Improviser, Proceedings of the International Conference
on Computational Creativity: ICCC-X, Lisbon,
Portugal.
Cohen, H., 1981. On Modeling of Creative Behavior. Rand
Corporation Report Series, P-6681.
Colton, S., 2008. Creativity versus the Perception of Creativity
in Computational Systems. Proceedings of the AAAI
Spring Symposium on Creative Systems.
Cropley, D.H., Kaufman, J.C., Cropley, A. J. 2011. Measuring
Creativity for Innovation Management, Journal of
Technology Management & Innovation 6(3), 13-30.
Csikszentmihalyi, M. 1996. Creativity: Flow and the psychology
of discovery and invention. New York: HarperCollins.
de Bono, E. 1975. New Think: The Use of Lateral Thinking
in the Generation of New Ideas. New York: Basic Books.
Gladwell, M. 2009. What the Dog Saw and Other Adventures,
London: Allen Lane.
Glicksohn, J. 2011. Schizophrenia and Psychosis. In M.A.
Runco & S.R. Pritzker (Eds.) Encyclopedia of Creativity,
(2nd ed.), New York: Academic Press.
Goodman, N. 1978. Ways of Worldmaking. Indianapolis,
IN (USA): Hackett Publishing.
International Conference on Computational Creativity 2012 65
Gordon, W.J.J. 1961. Synectics: The Development of Creative
Capacity. New York: Harper & Row.
Harry, H. 1992. On the role of machines and human persons
in the art of the future. Pose 8, 30–35.
Holstein B.I. 1970. Use of Metaphor to Induce Innovative
Thinking in Fourth Grade Children. Ph.D. thesis, School
of Education, Boston University, Boston, Mass.
Horn, D. and Salvendy, G. 2006. Consumer-based assessment
of product creativity: A review and reappraisal. Human
Factors and Ergonomics in Manufacturing & Service
Industries 16(2), 155-175.
Indurkhya, B. 1997. On Modeling Creativity in Legal Reasoning.
Proceedings of the Sixth International Conference
on AI and Law, Melbourne, Australia.
Indurkhya, B. 2007. Creativity in Interpreting Poetic Metaphors.
In T. Kusumi (Ed.) New Directions in Metaphor
Research, Tokyo: Hitsuji Shobo.
Indurkhya, B. 2010. On the Role of Metaphor in Creative
Cognition. Proceedings of the International Conference on
Computational Creativity: ICCC-X, Lisbon, Portugal.
Indurkhya, B. to appear. Computers and Creativity. In C.
Forceville & T. Veale (eds.) Agile Mind, Mouton.
Indurkhya, B., Kattalay, K., Ojha, A., & Tandon, P. 2008.
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,
Amsterdam: IOS Press.
Ishii, Y., Indurkhya, B., Inui, N., Nose, T., Kotani, Y., &
Nishimura, H. 1998. A System Based on Rodari’s ‘Estrangement’
Principle to Promote Creative Writing in
Children. Proceedings of Ed-Media & Ed-Telecom 98,
Freiburg, Germany.
Kaufman, J.C., & Beghetto, R.A. 2009. Beyond Big and
Little: The Four C Model of Creativity. Review of General
Psychology 13(1), 1–12.
Kozbelt, A. 2005. Factors Affecting Aesthetic Success and
Improvement in Creativity: A Case Study of the Musical
Genres of Mozart. Psychology of Music 33(3), 235–255.
Krzeczkowska, A., Colton, S., & Clark S. 2010. Automated
Collage Generation – With Intent. Proceedings of
the International Conference on Computational Creativity:
ICCC-X, Lisbon, Portugal.
Kurzweil, R. 2001. Cybernetic Poet.
http://www.kurzweilcyberart.com/poetry/ Last accessed on
29 January, 2012.
López, A.R., Oliveira, A.P., & Cardoso, A. 2010. RealTime
Emotion-Driven Music Engine. Proceedings of the
International Conference on Computational Creativity:
ICCC-X, Lisbon, Portugal.
Maduro, R. 1976. Artistic creativity in a Brahmin painter
community. Berkeley, CA: Center for South and Southeast
Asian Studies.
Maher, M.L. 2010. Evaluating Creativity in Humans,
Computers, and Collectively Intelligent Systems,
DESIRE’10: Creativity and Innovation in Design, Aarhus,
Denmark.
McCorduck, P. 1991. Aaron’s code: Meta-art, artificial
intelligence, and the work of Harold Cohen. New York: W.
H. Freeman.
Monteith, K., Martinez, T. & Ventura, D. 2010. Automatic
Generation of Music for Inducing Emotive Response. Proceedings
of the International Conference on Computational
Creativity: ICCC-X, Lisbon, Portugal.
Ni, Y., Santos-Rodriguez, R., Mcvicar, M., & De Bie, T.
2011. Hit Song Science Once Again a Science? Presented
at the 4th International Workshop on Machine Learning
and Music: Learning from Musical Structure, Sierra Nevada,
Spain.
Painter, K. 2002. Mozart at Work: Biography and a Musical
Aesthetic for the Emerging German Bourgeoisie. The
Musical Quarterly 86(1), 186–235.
Pérez y Pérez, R., Negrete, S., Peñalosa, E., Ávila, R.,
Castellanos, V., & Lemaitre, C. 2010. MEXICA-Impro: a
Computational Model for Narrative Improvisation. Proceedings
of the International Conference on Computational
Creativity: ICCC-X, Lisbon, Portugal.
Ramachandran, V.S., & Hirstein, W. 1999. The Science of
Art: A neurological theory of aesthetic experience, Journal
of Consciousness Studies 6(6-7), 15–51.
Rodari, G. 1996. The Grammar of Fantasy (J. Zipes,
Trans.) New York: Teachers & Writers Collaborative.
Runco, M.A., & Kim, D. 2011. The Four Ps of Creativity:
Person, Process, Product and Press. In M.A. Runco & S.R.
Pritzker (Eds.) Encyclopedia of Creativity, (2nd ed.), New
York: Academic Press.
Sacks, O. 1995. An Anthropologist on Mars: Seven Paradoxical
Tales. New York: Alfred A. Knopf.
Sawyer, K. 2006. Explaining Creativity. Oxford University
Press.
Shapira, O., & Liberman N. 2009. An Easy Way to Increase
Creativity. Scientific American (Mind Matters), July
21, 2009.
Singh, S. 2004. Big Bang. New York: Harper Collins.
Szabo, G., & Huberman, B.A. 2010. Predicting the Popularity
of Online Content. Communications of the ACM
53(8), 80–88.
Wegener, A. 1966. The origin of continents and oceans
(4th ed.) (trans. J. Biram), London: Dover (Original work
published 1915.)
Zeki, S. 2000. Inner Vision: An Exploration of Art and the
Brain. Oxford (UK): Oxford University Press.
International Conference on Computational Creativity 2012 66