Вы находитесь на странице: 1из 499

Aesthetics and Innovation

Edited by

Leonid Dorfman
Perm State Institute of Art and Culture

Colin Martindale
University of Maine

Vladimir Petrov
State Institute for Art Studies, Moscow

Cambridge Scholars Press

Table of Contents

Preface..................................................................................................... vii


A Psychological Theory of the Aesthetic Experience
Bjarne Sode Funch................................................................................. 3

Objective and Subjective Components of Aesthetic Experience
Natalja V. Nekrylova............................................................................... 21

What is Beauty? On the Dichotomy between Subject and Object
around 1900
Volker A. Munz......................................................................................... 32

Complementary Relations between Quantitative and Qualitative
Approaches to Research in Aesthetics
Gerald C. Cupchik and Michelle C. Hilscher......................................... 46

When the Real van Gogh is Real! Cognitive Top-Down Effects in Art
Helmut Leder and M. Dorothee Augustin.............................................. 66
Quantitative Estimations of Creativity: Social Determination and Free
Lidia A. Mazhul and Vladimir M. Petrov............................................... 82

Neuropsychological Support to the Novelty Generation Process
T. Sophie Schweizer............................................................................... 101

Plural Self, Plural Achievement Motives, and Creative Thinking
Leonid Dorfman and Anastasia Ogorodnikova..................................... 126

Perversion and Creativity in the Language of War
Robert Hogenraad................................................................................. 162

Art Constructs as Generators of the Meaning of the Work of Art
Victor F. Petrenko and Olga N. Sapsoleva........................................... 181

Portrayal of Women and Jungian Anima Figures in Literature: Quant-
itative: Content Analytic Studies
Anne E. Martindale and Colin Martindale.......................................... 205
Translation of Values through Art: Non-Classical Value Approach
Dmitry Leontiev..................................................................................... 227

Individual and Professional Differences in the Perception of Dramatic
Dmitry Leontiev and Larissa Lagoutina............................................... 242

The Role of Attachment Patterns in Emotional Processing of Literary
János László and Éva Fülöp................................................................. 258

Pictures in the Mind: Symmetry and Projections in Drawings
Diane Humphrey and Dorothy Washburn............................................ 274

Automatic Affective Evaluation of Pictures and Words
Stefano Mastandrea............................................................................... 288

Computer Sound Analysis in Musicology: Its Goals, Methods, and
Alexander V. Kharuto............................................................................ 306
Sociocultural Oscillations and their Analogies with Physical Waves
Colin Martindale................................................................................... 327

“Evolutionary Genius” and the Intensity of Artistic Life: Who Make
Musical History?
Peter Kulichkin...................................................................................... 366

The Expanding Universe of Literature: Principal Long-range Trends in
the Light of an Informational Approach
Vladimir M. Petrov................................................................................ 402

A Biological Approach to a Model of Aesthetic Experience
Oshin Vartanian and Marcos Nadal..................................................... 434

Gender Differences in Creativity: A Psychophysiological Study
Olga M. Razumnikova, Nina V. Volf, and Irina V. Tarasova............... 450

Subject Index......................................................................................... ???


In this book we have attempted to gather together a set of chapters that de-
scribe new ways of approaching questions about aesthetics and innovation.
Rather than going over old ground, the chapters describe attempts to break
new ground. A number of chapters are by Russian scholars. A valuable aspect
of Russian scholarship is that many topics, such as art history, are studied
with quantitative methods rather than being left to imprecise qualitative hu-
manistic approaches. As well as describing new methods and results, they
will be novel to most western readers, because the Russian perspective on
aesthetics and innovation is rather different than the traditional western per-
spective. Looking at phenomena from a new viewpoint never hurts and very
often helps in science.
The chapters in Part I deal with method, theory, and the history of aesthet-
ics. Much of the time, we merely glance at art works, and they have rather
little impact upon us. However, art can have a very striking and long lasting
effect upon us. The book begins with Funch's analysis of this phenomenon.
Munz describes von Ehrenfel's development of his Gestalt theory of beauty in
the context of intellectual life in Vienna around 1900. He makes this period
come to life in way that leads the reader to wish that he or she were living in
Vienna then rather than wherever he or she may be living now. Nekrylova
and Cupchik and Hilscher describe new and profitable ways in which art can
be studied by a combination of both qualitative and quantitative methods.
The two methods together give us a better picture of the perception of art
than either used in isolation. Helmut Leder and Dorothee Augustin describe a
useful framework for describing the perception of art and outline work deal-
ing with top-down influences on how we view a work of art. Art does not
merely elicit a response. Rather, the response is heavily influenced by what
we have been led to believe about a work of art.
The second part of the book deals with some new approaches to the creat-
ive process. Mazhul and Petrov examine the degree to which creators are free
to work as they wish as opposed to being influenced by external factors. So-
cial support for creative work has been little studied. Schweizer looks at in
detail and most interestingly does so from a neuropsychological perspective.
Based on the socioindividual world theory, Dorfman and Ogorodnikova dis-
criminate the plural self into subselves and the plural achievement motives
into submotives either of motives to achieve success or to avoid failure. They
examine the link between subselves, achievement submotives and creative

thinking. We seldom think of war and creativity as being closely allied, but
Hogenraad describes research showing that the rhetoric of politicians seems
to be more creative when war is immanent as opposed to when it is drawing
to a close.
The chapters in Part III deal in one way or another with the perception
and understanding of art and literature. Petrenko and Martindale and Mar-
tindale deal with the "personality" of characters in film and literature. Pet-
renko describes a novel method of studying these personalities based upon
Kelly's personal construct theory. Martindale and Martindale use content ana-
lytic methods to study Jungian hypotheses about the portrayal of women in
literature. Though Jung is usually accused of formulating theories that cannot
be tested, they show that this is not at all the case. Leontiev and Lagoutina
also use a method derived from Kelly's work to study the influence of values
upon both the creation and the perception of art. The chapters by Leontiev
and Laszlo deal with individual differences in the perception of art. Leontiev
shows that those with training in theater view theatrical performances in a far
different way than those without such training. Laszlo shows that reading the
same work of literature will arouse quite different emotions in readers with
different personality types.
In Part IV we have grouped three chapters that deal with innovative meth-
ods of studying the arts. Humphrey and Washburn compare the drawings of
artists and non-artists on variables such as the seven band symmetries, affine
projections, and so on that have seldom if ever been used in the study of art
works. Mastandrea describes studies showing the intimate relationship of im-
agery and emotion that promise to shed light on how visual art is perceived.
Finally, Kharuto describes a fascinating computer program that analyzes mu-
sic in ways that had before to be done only by ear in an imprecise way.
The chapters in Part V deal with the evolution of art and literature. Mar-
tindale describes the oscillations that are set in motion by innovation not only
in the arts but also in science and the economy. He raises the question of the
degree to which these oscillations resemble the physical oscillations found
for example in light waves and sound waves. The oscillations found in art
arise from the periodic introduction of new styles in the various genres of art
and literature. Based on his study of 6453 composers of the 13th—20th centur-
ies. Kulichkin describes in detail how such styles are introduced by creative
geniuses. The arts did not always consist of different genres. Petrov uses in-
formation theory to explain how such differentiation occurred.
The book ends with two chapters on the new and fascinating work being
done on the brain and the creation and perception of art. Vartanian and Nadal
review studies of brain scans done when people are observing visual art. The
studies have yielded consistent results and, if nothing else, lead us to question
the notion of the disinterested nature of aesthetic perception. Razumnikova,
Volf, and Tarasova describe a large scale study of EEG patterns and creative
thought. Very interesting sex differences suggest that men and women may
use quite different strategies in thinking of new ideas.
We thank Yury Maximov who put the book into camera-ready form and
offered invaluable help along the way.

Leonid Dorfman,
Colin Martindale,
Vladimir Petrov



A Psychological Theory of the Aesthetic


Bjarne Sode Funch

A psychological theory of art’s influence on the human psyche and existence

is presented in this chapter. With a point of departure in a phenomenological
description of the aesthetic experience as a transcendent phenomenon it is
suggested that an aesthetic experience provides a distinct form to an existen-
tial theme in the life of the perceiver. A phenomenological model of con-
sciousness serves as a basis for an understanding of how a momentary experi-
ence of a specific work of art on rare occasions exerts an extraordinary influ-
ence on the psyche and existence by constituting an emotional quality and
giving options for reflecting an existential theme in the life of the perceiver.

Several years ago I went to the Tate Gallery in London. I remember walk-
ing around and enjoying the fine collection of works of art. At some point I
came to an opening with a curtain in front of it and at first I was not sure if it
was open to the public, but decided to enter nonetheless and found myself in
a room with dimmed light surrounded by a group of large paintings in black
and maroon. When I left the room I learned that it was the Seagram Series
(1959) by the American painter Mark Rothko. I was totally absorbed in these
paintings. They created a vibrating atmosphere that deeply touched me, and I
don’t know how long I stood there in front of each painting letting myself
disappear in their infinite color fields. It was an unusually powerful emotion-
al sensation; it is difficult to say exactly what kind of emotion it was, apart
from it being extremely pleasurable.

4 / Aesthetics and Innovation
The Aesthetic Experience
It is well-known that people on exceptional occasions have experiences
with works of art that transcend the ordinary stream of consciousness. Among
other accounts of such experiences is one by the Roman theologian and
philosopher Augustine who describes in his Confessions how the church mu-
sic has affected him. He (1961, p. 190) writes, “The tears flowed from me
when I heard your hymns and canticles, for the sweet singing of your Church
moved me deeply. The music surged in my ears, truth seeped into my heart,
and my feelings of devotion overflowed, so that the tears streamed down. But
they were tears of gladness.”
Throughout the history of the fine arts there are similar accounts and sug-
gestions of a type of art appreciation that is so extraordinary that it presents
an issue that cannot be ignored. Philosophers and psychologists such as Mon-
roe Beardsley (1981), Martin Lindauer (1981), Robert Panzarella (1980), Mi-
haly Csikszentmihalyi and Rick E. Robinson (1990) have already embarked
on a description of this phenomenon of art appreciation and in a continuation
of their work, I (Funch 1997) have established on the basis of accounts from
their studies, as well as my own experiences and the interviews that I have
carried out, a number of phenomenological characteristics of what I suggest
to call the aesthetic experience.
The phenomenology of the aesthetic experience with a work of visual art
includes cognitive, visual, emotional as well as intentional aspects. Especially
the visual and the emotional aspects are conspicuous in comparison to the
same aspects during ordinary perception, whereas the cognitive aspect of the
aesthetic experience is relatively ordinary by promoting spontaneous recogni-
tion of the subject matter. The intentional aspect seems by first sight quite or-
dinary by focusing on the work of art with a passive and disinterested atti-
tude, but further phenomenological studies reveal an extraordinary relation-
ship between spectator and work of art. In aesthetic experiences within other
disciplines such as music and literature, the visual aspect will be replaced by
an audible or imaginary aspect and these will be conspicuous just like the
visual aspect in encounters with works of visual art.
In the question of the phenomenology of the aesthetic experience it is im-
portant to be aware that an aesthetic experience is elicited by a specific work
of art and therefore, the aspects of the experience are determined by the sub-
ject matter. Consequently, an aesthetic experience of Guernica (1937) by
Pablo Picasso, for example, is different from an aesthetic experience of
Dance (1910) by Henri Matisse. Not only will visual perception and recogni-
tion vary, but the emotions will vary as well according to the painting’s sub-
Aesthetic Experience / 5
ject matter. At the same time there are some general characteristics of the
visual and emotional dimensions of the aesthetic experience which are of ut-
most importance for the understanding of the psychological effects of the ex-

Visual aspect of the aesthetic experience

The visual aspect of an aesthetic experience with a painting is emphasized

by complete unity, lucidity, eminence and originality. The picture appears in
experience as a unified whole in which every detail is an inevitable compon-
ent of the composition. No visual element is more prominent than the others;
rather they all contribute to a common image. The experience is limited to the
work of art in question and the picture’s surroundings are excluded from ex-
perience. Monroe Beardsley as well as Mihaly Csikszentmihalyi point to this
aspect of visual unity. Beardsley (1981, p. 527) writes, “An aesthetic experi-
ence is one in which attention is firmly fixed upon heterogeneous but interre-
lated components of a phenomenally objective field—visual or auditory pat-
terns, or the characters and events in literature.”
Lucidity or luminosity is another characteristic of the visual aspect. Paul
Tillich, the German-American philosopher and theologian, once described
this aspect from an experience he had with a painting by Sandro Botticelli at
the Kaiser Friederich Museum in Berlin. He (1955, p. 235) writes, “Gazing
up at it, I felt a state approaching ecstasy. In the beauty of the painting there
was Beauty itself. It shone through the colors of the paint as the light of day
shines through the stained-glass windows of a medieval church.”
This aspect of luminosity as if the painting is imbued with its own inner
light is occasionally referred to in phenomenological accounts, but neither
Beardsley (1981), Csikszentmihalyi and Robinson (1990), Martin Lindauer
(1981), or Robert Panzarella (1980) mention this aspect as a characteristic of
an aesthetic experience.
Sublimity or eminence is a third characteristic of the visual aspect. The
picture in question appears during the aesthetic experience with such a pro-
found prominence and vital power that it transcends what is depicted. The
subject matter appears with a visual singularity and autonomy that makes its
references subordinate to its own presence.
This aspect is hinted at when Panzarella (1980) speaks of “an altered per-
ception” during the aesthetic experience and possibly also when Csikszentmi-
halyi and Robinson (1990, 68) refer to a state of “heightened awareness”, but
only in Erich Neumann’s description is it given full attention. He (1989, p.
188) describes the sublime aspect, or the “eternal” aspect as he calls it, as
6 / Aesthetics and Innovation
something that becomes visible to us as “the authentic reality”. He (p. 189)
writes, “Yet in every case, we find ourselves confronted with the ineffable
magnificence of something which can never be grasped, but which animates
everything that is real as a crowning super-real essence.” This sublime qual-
ity of the visual aspect is difficult, if not impossible, to isolate from other as-
pects of the aesthetic experience, but it is important to notice that the visual
presence of the subject matter is imbued with a distinct form or appearance.
This leads to a fourth characteristic of the aesthetic experience, namely, ori-
ginality, which is often associated with works of art.
Originality in regards to visual art is associated with the visual appearance
of the subject matter. The spontaneous experience of originality may be diffi-
cult to differentiate from the aftermath of comprehension. The American art
historian H. W. Janson states originality as a defining quality of a work of art.
He (1995, p. 21) writes, “Whether we are aware of it or not, tradition is the
framework within which we inevitably form our opinion of works of art and
assess their degree of originality.” This attempt to define originality within
the context of tradition is very different from the spontaneous experience of
something which has never been experienced before. The subject matter may
actually not be foreign to the spectator, but it appears spontaneously in a form
that is novel and unique without any evaluating references at all.

Emotional aspect of the aesthetic experience

Art appreciation and emotions are closely related to each other. Whenever
we speak about an experience with art, the emotional response is recognized
as fundamental. This emotional response in ordinary encounters with art is
expressed from two different levels of consciousness. One being the spontan-
eous response of horror, joy, or any other specific emotional quality elicited
by the work in question; the other being the reflective response evaluating the
work as being a good, exciting, bad or any other evaluation of the work in
question. The emotional aspect of the aesthetic experience is always a spon-
taneous response and it usually transcends ordinary emotionals by being ex-
ceptionally intense. The Dutch literary scholar Ernst van Alphen once de-
scribed such an emotional response to a painting by Francis Bacon. He (1992,
p. 9) writes, “Seeing a work by Francis Bacon hurts. It causes pain. The first
time I saw a painting by Bacon, I was literally left speechless. I was touched
so profoundly because the experience was one of total engagement, of being
dragged along by the work. I was perplexed about the level on which these
paintings touched me: I could not even formulate what the paintings were
about, still less what aspect of them hurt me so deeply.” Similar descriptions
Aesthetic Experience / 7
of emotional responses to works of art have been brought forward by James
Elkins (2001) and Robert Panzarella (1980) and these accounts demonstrate a
broad spectrum of different emotional qualities as well as an emotional in-
tensity that is overwhelming to the spectator.
These emotional responses to works of art differ from ordinary emotions
mainly because they are elicited by fiction. Ordinary emotions are caused by
incidents in real life originating, for example, from a relationship where an-
other person hurts, irritates, or pleases you, whereas the emotional response
to art is a response to an occurrence that has no influence on life apart from
the relationship to the work of art. Nevertheless, the emotional impact seems
to be so significant that it is ascribe great importance and remembered as an
exceptional experience. A major reason for this distinguishing quality is an
experience of terrific pleasure which arises together with the specific emo-
tional quality associated with the work of art in question. Even in cases a
work of art elicits an emotion associated with a feeling displeasure as for ex-
ample disgust, anxiety, or terror, the aesthetic experience is characterized by a
feeling of transcending pleasure.

Focus of the aesthetic experience

The intentional aspect or the focus of the aesthetic experience is similar to

the unusual character of the visual and emotional experience. One’s focus be-
comes completely concentrated on the work of art to the exclusion of
everything else. The surrounding context drops away from consciousness and
there are no reflections, associations, or comparisons to disturb the focus on
the work of art. The visual appearance of the work of art defined by spontan-
eous identification and emotionality completely occupies consciousness. Fur-
thermore, the focus transcends the usual dichotomy between subject and ob-
ject which means that the work of art is experienced with a significant quality
of existential density. It is as if the spectator is transparently present in what
is being perceived which makes the experience rooted in the present moment.
Perceiving without the usual division between subject and object invigor-
ates the viewer, causing him or her to feel present without necessarily being
self-aware. There are no intentions that point to specific intrinsic or extrinsic
interests, as if the experience contains the intentions themselves. It is a mo-
mentary incident as if time is abolished and without any practical, relational,
or other kinds of perspectives. This type of intentionality is described by
Mikel Dufrenne (1973, p. 406) as participation with rather than an aim to-
8 / Aesthetics and Innovation
The special character of the intentional aspect has invited many different
and sometimes contradictory descriptions. “Being absorbed” or “losing self-
awareness” are typical descriptions of the focus of the aesthetic experience.
Csikszentmihalyi and Robinson (1990, p. 122) put forward that “attention is
so completely focused, so completely enmeshed in the interaction with the
artwork, that the viewer gives up, at least momentarily, his most human at-
tribute: self-consciousness.” This description of absorption or loss of ego is
from time to time contradicted by descriptions such as “detached affects”,
“disinterested enjoyment” and “psychical distance”. Absorption and distance
seem to contradict each other in conception of the perceiver’s ego, but actu-
ally they emphasize or rather overemphasize two features of the intentional
aspect. On the one side, the work of art lays claim to the entire experience
without any specific attention given to the viewer. On the other side, the ego
or self is intensively present without being focused on. This mixture of ab-
sorption and personal presence can be recognized as a characteristic quality
of the aesthetic experience.

Spiritual reflection

The phenomenology of the intentional focus indicates a distinctive state

of being. Building on Søren Kierkegaard’s idea, I suggest identifying this as-
pect as an example of self-actualization. He (1980, p.13) defines the self as “a
relation that relates itself to itself.” Applying his idea to the aesthetic experi-
ence reveals a double nature of it. First, the aesthetic experience is a relation-
ship between the perceiver and the work of art in a direct and momentary en-
counter. At the same time, it is an example of a relationship between an exist-
ential theme and the personality of the perceiver in the sense that personal
life-experiences are reflected in the theme in question. The sublime quality of
the visual aspect indicates that not only the specific work of art defines the
experience, but it is also permeated by ideals that can only originate from the
perceiver’s life-experiences. The first relationship reflected in the second
refers to what Kierkegaard calls “the self” or “spirit”. He (p. 13) states that “a
human being is a synthesis of the infinite and the finite, of the temporal and
the eternal, of freedom and necessity, in short a synthesis. A synthesis is a re-
lation between two.”
Defining the self as a relation relating to itself, Kierkegaard establishes
the self as a process of becoming oneself. The self is constituted in the mo-
ment or the process of relating one relation to the other rather than as a trait
or skill that characterizes the person in question. Kierkegaard (1980, p. 14)
concludes that “the self rests transparently in the power that establishes it.” In
Aesthetic Experience / 9
other words, an experience of existential density originates from a ruling art
encounter being reflected in life-experiences. The self is a process of being
constituted by the process itself.
Kierkegaard (1980) mentions the will not to be oneself and the will to be
oneself as major obstacles for the constituting process of the self. These are
also major obstacles for the aesthetic experience when an art audience either
rejects to engage in a direct encounter or uses the work of art for self-con-
formation or self-orchestration.
The aesthetic experience is a sudden and momentary incident that occurs
without any proceeding intentions. It is a case of a person’s personality being
reflected in his or her current art encounter. The question is how this process
of constituting the self can be understood from a psychological point of view.

Emotional Constitution
With a point of departure in the proceeding phenomenological description
of the aesthetic experience I postulate that an aesthetic experience is an in-
stance in which an emotion is constituted in its existential actuality. My reas-
ons for this thesis are first of all based on the transcending character of the
aesthetic experience in which the sensual and emotional aspects are conspicu-
ous. This phenomenological emphasis on the two aspects makes it evident to
relate them to each other.
In psychology, emotions have always been closely related to the senses.
Visual, auditory, tactual, olfactory, and gustatory impressions are primary
sources for eliciting emotions. Here I take the liberty to include imagination
among the primary senses because it is clear that dreams, memories, and
fantasies caused by thinking bring forth emotions just as sense impressions
do. Cognitive reflections on primary sense impressions on the other hand do
not bring forth emotional responses by themselves and have even an imped-
ing influence on emotional states of being. Sense impressions in ordinary per-
ception are linked to specific emotional qualities so that each time a person
encounters a vicious dog for example he or she will feel fear. With this ex-
ample in mind it should be taken into consideration that emotional responses
vary, at least to some degree, from person to person. In each case people re-
cognize the specific quality of such emotional responses because they are fa-
miliar with them from previous encounters. The exceptional thing about the
aesthetic experience is that the person does not recognize the specific quality
of the emotion elicited by the work of art. The emotion is not only ineffable;
it has never before been associated with a sense impression. It may seem fa-
miliar although it doesn’t bring forth any memories. This leads to the as-
10 / Aesthetics and Innovation
sumption that the emotion during an aesthetic experience is linked to a sense
impression for the first time.
A precondition for the thesis that the appearance of a work of art elicits an
unconstituted emotional quality is that the repertoire of emotions increases
during the course of life and it is not limited to a fixed number of emotions. It
is clear from watching a newborn that we respond with appreciation or rejec-
tion right from the beginning, even before we have any idea of the world. At
first the child still lives in a kind of symbiosis with the surrounding world,
and within this symbiotic state of being the infant sometimes expresses a
feeling of harmony with the world as for example when the infant nurses or
simply cuddled up in her arms. Being undressed or hearing a loud crash will
cause a response of rejection and despair, and the baby will demonstrate feel-
ings of displeasure. The emotional responses of appreciation and rejection are
little by little associated with objects in the child’s surroundings. The emo-
tions are no longer just states of being but are also related to specific sense
impressions. The baby will soon be able to recognize and even anticipate ex-
terior circumstances and respond with appropriate emotions.
The child’s emotional repertoire develops little by little as new emotions
arise and as the child learns how to differentiate between the already existing
emotions. Appreciation is associated with new objects and situations and new
qualities are constituted. As an adult the repertoire of emotions continues to
expand as new existential encounters require adequate emotional responses.
In this way the repertoire of emotions, as with knowledge, is increasing dur-
ing the entire course of life.
Most people will throughout life encounter new existential themes and
consequently new emotional qualities need to be constituted. The repertoire
will be different from person to person, although we all have a good number
of emotions in common. The interesting thing is that the emotional life varies
from generation to generation. People today encounter life situations that are
very different from those encountered in former times. Their living and work
situations are different, they relate differently to their partners, children, and
friends, and their cultural and spiritual engagements are also different. Emo-
tions experienced by people today may have been totally unknown to former
generations. From a phenomenological point of view, human existence holds
an uncountable number of different emotional qualities, and in future there
will appear emotional qualities that we do not know of yet.
The origin of emotions experienced during an aesthetic experience is dif-
ferent from emotions experienced in daily life. A specific feeling of repulsion
or attraction in front of a painting or a piece of music is no different from the
same feeling arising in ordinary situations, but since emotions during aesthet-
Aesthetic Experience / 11
ic experiences are elicited by fiction, they are not associated with intentions
or requirements for action. The emotional intensity may be just as powerful
as emotions in daily life, but they cause no action.
This peculiar lack of initiation in the aesthetic experience leads to the as-
sumption that an emotion during an aesthetic experience is a reawakening of
an emotion rather than a proper emotional response. When an adult, just like
a child, encounters a specific situation he or she experiences a specific emo-
tional quality. If this is the first time this emotional quality occurs it may not
be constituted right away by the concurrent sense impression. A reason may
be that the sense impression has not been sufficiently distinct to serve as con-
stituting factor. Such emotions which are not constituted by sense impres-
sions can only be actualized in new encounters that are similar to the first;
they cannot be remembered and recalled. When a work of art reawakens such
a latent emotion it is because the work elicits an experience that is equal to
the original experience. The work displays at the same time a distinct form
which is appropriate for the emotion in question. In other words, the work of
art elicits an existential theme in compliance with similar experiences from
past life encounters and provides an adequate sense impression for the emo-
tion to be constituted.

Phenomenological Model of Consciousness

A phenomenological conception of human consciousness serves as the
basis for understanding the aesthetic experience.
Ordinary consciousness is according to phenomenological studies an in-
teraction between two different approaches to life-world. One is a spontan-
eous approach characterized by an uninterrupted stream of consciousness, an-
other is a reflective approach basically characterized by reflecting the spon-
taneous stream. The two approaches alternate with each other in such a short
interval that they seem to form one kind of consciousness.
During the stream of consciousness sense impressions, imagery, emo-
tions, and thoughts randomly follow each other without any specific focus.
There is no subject-object relationship in play, and therefore, no point of view
or an ego in command. Under exceptional circumstances such as in dream,
meditation, contemplation, and activities with an unusually concentrated fo-
cus, the period of spontaneous consciousness is extended without interrup-
tion. Csikszentmihalyi (1990) talks about “flow” as a characteristic of fo-
cused activity. Extreme states of spontaneous consciousness consist of sense
impressions alone without any kind of conceptualization except from the
form appearing before one. Less extreme states include emotions, imagery,
12 / Aesthetics and Innovation
and even thinking in its non-reflective form. These aspects of consciousness
are determined not only by sense impressions elicited by the current sur-
roundings but also by what randomly arises from memory and mental im-
agery. The number of impressions during the spontaneous approach is stag-
gering and it is unthinkable that each and every one of them is stored in
memory. Nevertheless, it is well-known that some of these impressions if
they are loaded with strong emotions are stored in memory and become part
of a person’s life-experiences. If such impressions are not followed up by re-
flection they are stored in a concealed memory and are usually impossible to
recall at a later time. This happens with traumatic experiences for example.
Impressions that are stored in mind without access create a dynamic factor by
randomly intruding on the stream of consciousness. Such intrusions make a
claim on consciousness and influence a general state of being. In case of trau-
matic memories they contribute with unrest and anxiety and the person in
question is not able to do anything or control the situation. In this way a per-
son’s general state of being is influenced by concealed memories whenever
they are brought into action and the person is inable to interrupt and bring an
end to such powerful memories.
Concealed memories are important to the understanding of the aesthetic
experience by providing the emotional aspect of the experience. When a per-
son encounters a work of art, it is not the work as such that stimulates an
emotional response, but it brings forth an existential theme with a corres-
ponding emotional quality. This emotional quality originates from past exper-
iences which are stored as concealed memories. When such an emotional
quality is actualized through an art encounter it is linked to the appearance of
the work of art, and through these means, constituted as an emotion that can
be approached through reflection.
Ordinary consciousness alternates between the spontaneous stream of
consciousness and the reflective approach. The process of reflection basically
mirrors the spontaneous stream, but during this process different cognitive
capacities such as identification, discrimination, comparison, and so forth, are
also brought into action. For example, if a person is contemplating a land-
scape and suddenly something happens, he or she will switch to reflective
consciousness by asking what it is. This attentive attitude brings cognitive
schemata into action in an attempt to identify the subject. When the subject is
identified this subject may be the cause of a new sequence of spontaneous
stream of consciousness and so forth.
Consciousness in the reflective mode is different from the spontaneous
stream by constituting a subjective perspective. In trying to identify
something moving in the landscape, the spectator is constituted in experience
Aesthetic Experience / 13
as an identity that relates to the surrounding landscape. This identity may just
be defined as me without further qualifications than being the onlooker of the
landscape. Thus, the subject-object relationship is constituted through the re-
lationship between the spontaneous stream and the reflective process and this
relationship means that the cognitive identifications are accentuated in exper-
ience. Whereas the spontaneous stream has no priorities, the reflective mode
is focused and determined by personal cognitive capacities. When a person
from a city for example looks at a field and recognizes the crop as grain, the
farmer will be able to discern subtle differences and identify the same crop as
wheat, oats, rye, or barley.
The reflective focus lays the ground for ordinary memory—a sort of
archive in which specific entities can be reactivated to bring up previous ex-
periences and knowledge. This depository of sense impressions, concepts,
and other kinds of cognitive entities, serve as grounds for cognitive schemata.
These schemata provide the basis for recognition and identification.
Ordinary consciousness continuously goes back and forth between spon-
taneous and reflective modes and they interact in such a way that they com-
plement each other. The spontaneous approach provides new material for the
reflective process and the reflective approach initiates spontaneous aware-
ness. Their interactions are so subtle that they usually are recognized as one
kind of consciousness.
The reflective mode in the ordinary sense is excluded from the aesthetic
experience, although spontaneous identification necessary for constituting the
existential theme in question is an aspect of the spontaneous stream of con-
sciousness. The reflective process during the aesthetic experience is ex-
traordinary because it reflects appropriate parts of the concealed memory and
elicits emotions from previous life experiences. I suggest calling this process
spiritual reflection. During spiritual reflection life experiences are recalled in
their living state without being subject for recognition and cognitive pro-
cessing. This reawakening of familiar emotional qualities that cannot be iden-
tified creates a feeling of existential density which is a major characteristic of
the aesthetic experience.
During the spiritual reflective process the spontaneous stream of con-
sciousness constituted by the art encounter is reflected in life-experience. The
same existential theme as the one elicited by the work of art is reawakened
from the concealed memory. It is not the memories as such that are recalled,
but the emotional reminiscences that are elicited as if they are part of the
present experience. When these emotional reminiscences are linked with the
visual appearance of the work of art they are brought to a new stage of con-
sciousness. They are brought out of the concealed memory by being objecti-
14 / Aesthetics and Innovation
fied in the form of the work of art, and as a result, they will no longer have a
disturbing influence on the steam of consciousness. The spiritual reflective
process of the aesthetic experience brings in this way harmony to the stream
of consciousness and provides a platform for better contact to present life.
This theory of the aesthetic experience as a constituting factor in a psy-
chological process of increasing personal integrity cannot be approved
through systematic empirical research because the aesthetic experience is un-
predictable and because the complexity of the interaction between person and
work of art prevents controlled investigations. Therefore, the theory’s truth
value can only be tested on its ability to account for art’s psychological and
sociological status in daily life. The following account is an attempt to put
personal, societal, and artistic circumstances into perspective.

Personal Perspectives
For many people art is a leisure activity. They are attracted and fascinated
by works of art as products of artistic craft and creative imagination. Even so,
art appreciation is often regarded as unproductive, and without fulfilling any
practical purpose. Thus, art is generally relegated to the role of entertainment.
The same people, however, who recognize art as a leisure activity have
also experienced on rare occasions works of art that have made deep impres-
sions on them. They will remember encounters with a specific book of fic-
tion, a piece of music, or a painting, in which they were touched at an exist-
ential level far beyond amusement. Experiences that set the ground for vital-
ized reflection and are imprinted in the mind forever. Such experiences are
often counted among the most important life-experiences, right alongside of
the birth of a child, the death of a friend or relative, and other occasions of
dramatic life changes. People may not talk much about such experiences with
art because they are imbued with mysticism and are difficult to describe.
Authors such as Orhan Pamuk (1997), Rainer Maria Rilke (1989) and
Nathalie Sarraute (1958) have suggested that art experiences of existential
character makes the person wish for a change of life. After experiencing the
archaic torso of Apollo, Rilke claims explicitly, “You must change your life.”
Such poetic descriptions may at first seem a bit exaggerated, but observations
reveal that an aesthetic experience is followed by a state of enlightenment.
Beardsley (1981, p. 560) writes, “There is often a very special refreshing
feeling that comes after aesthetic experience, a sense of being unusual free
from inner disturbance or unbalance.” He maintains that an aesthetic experi-
ence creates a feeling of personal integrity or harmony. It may also resolve
lesser conflicts, develop imagination, refine perceptual discrimination, and
Aesthetic Experience / 15
improve the ability to put oneself in place of others. He (p. 576) concludes
that an aesthetic experience is “an aid to mental health” and offers “an ideal
for human life.” Similar observation have been put forward by Mihaly
Csikszentmihalyi and Rich E. Robinson (1990), John Dewey (1934), Margh-
anita Laski (1961), Abraham H. Maslow (1968), Rollo May (1975; 1991),
and I. A. Richards (1925).
After all, art appreciation on some rare occasions gives cause for far-
reaching existential changes. The aftermath of an aesthetic experience indic-
ates a radical change of existence that can be understood in the light of the
thesis that an aesthetic experience provides a latent emotional quality with an
appropriate and distinct form.
Artistic engagement is another personal perspective of the psychological
effects of art. Professional artists have to quite often renounce economical se-
curity. In some instances they may have to risk their lives because of political
and religious restrictions. Furthermore, working as an artist is often an isol-
ated activity because the profession is independent of the democratic organiz-
ation of the society which means that there is no position as artist that would
provide working conditions, economic security, social relations, and so forth.
The artist is independent or even excluded from any type of societal support.
Moreover, the creative work is usually based on a personal engagement that
only adds to the isolation often experienced by artists. Nevertheless, some are
ready to devote their lives to art in spite of all privations and difficulties.
The question is, why do people choose to be artists when it isn’t directly
supported by the society? Artists often say that they cannot stop themselves
from engaging in creative activity. The work is so alluring that it is im-
possible to push it aside.
Although the prospects of fame and wealth, however slight, may be reas-
ons for embarking on an artistic career, but the intensity of creative engage-
ment reveals more essential reasons. Artists are often absorbed in their work
and pursue their ideas with persistent courage. From a psychological point of
view, creative engagement provides the artist with an existential anchoring
that is fundamental to his or her commitment. This anchoring of creativity in
existential dimensions of life makes it plausible that art appreciation may
have an existential impact on viewers as well.

Social Perspectives
Art is given great importance within the society. Institutions such as art
academies, art museums, libraries, theatres, and concert halls are established
with the only purpose of serving the art world. Government investments in
16 / Aesthetics and Innovation
art are sometimes considerable and buildings erected for the fine arts are of-
ten pretentious and displaying outstanding architectural designs. Moreover,
top artists are greatly admired among the public and art and artists are likely
to be the major focal points whenever culture, past or present, is studied or
celebrated. From this perspective, there is no doubt about the great import-
ance attached to art.
On the other hand, art also gives rise to great furor. Under totalitarian
governments certain expressions of art are commonly banned and artists pro-
secuted. When totalitarian states are brought down, art is once more target of
attack. This time people run riots against art previously adopted by the au-
thoritarian leaders. Throughout history iconoclasm has demonstrated how
works of art are closely associated with ruling powers as if art were a tool for
influencing people’s values and political engagement.
Art assumes a central position within the religious domain just as it does
within the secular society. Stories, pictures, and music are essential for the
practice of any religion. In Christianity for example the Bible contains par-
ables, poetry, and narratives that could be compared to secular literature. Pic-
tures, sometime highly estimated works of art, are displayed and used for de-
votion within the Roman-Catholic Church. A Lutheran service without music
is almost unthinkable. Art of every kind is a natural part of Christian rituals
and the same is true for other religions.
At the same time, there are restrictions concerning what may be depicted.
Within Judaism as well as Christianity it is forbidden according to the Holy
Scriptures to make pictures of God. This prohibition is administrated with
varying degrees of stringency, but in all cases it is based on a conviction that
pictures may exert such an influence on people that they will confuse the pic-
tures with God.
Censorship of art is also common within the educational domain. It is a
well-known phenomenon even in western democracies today that books are
banned from libraries and schools. There are very little resources given to art
education in schools with the focus placed predominantly on subjects such as
language, mathematics, and history. Art has to great extent been excluded
from the curriculum and reckoned as an amusement without much education-
al value.
The prevailing ambivalent attitude towards art within the society spans
from great respect to hostile rejection. The passion behind these opposing at-
titudes reveals that art has the potential to exert a great influence on people
and their existence making it something far more essential than a momentary
amusement. If art does constitute emotional qualities then it likely that art can
lay the ground for existential values.
Aesthetic Experience / 17
Artistic Perspectives
Artistic media in all the disciplines such as painting, music, and writing
can be manipulated and combined in an unlimited number of ways. The his-
tory of fine art demonstrates how painting for example has varied in style and
subject matter for hundreds or even thousands of years. Even if artistic media
are comprised of such basic elements as colors, tones, and words, still they
provide the options for variations that surpass the appearance of the sur-
rounding reality. A landscape depicted in a painting can be in comparison to
the same landscape in reality manipulated in a great many ways. Further-
more, the artistic media provides the means for creating artistic expressions
that have never before been introduced. New expressions are continuously
coming up and demonstrate the unending process of innovation associated
with art. From an existential-phenomenological point of view it is reasonable
to relate diversity and innovation within the fine arts to the reality of emo-
tional life. Just as in art, one’s emotional life varies in an unlimited number of
ways and new emotional qualities emerge as living conditions change. As-
suming that art constitutes emotions, it is plausible to say that art echoes
emotional life. As living conditions change and people encounter new exist-
ential themes in life, art has to produce new artistic expressions to provide the
necessary means for their constitution.
Works of art are ideally suited to play a role constituting emotions, not
only because of art’s diversity and its capacity for innovation, but also be-
cause works of art display a compositional unity which relates primarily to
only one of the human senses. The phrase “unity in variety” has often been
used to describe a work of art, meaning that every detail contributes to the
work as a whole. This makes works of art well suited for providing a distinct
form for the constitution of emotional qualities. The surrounding world may
also provide constituting forms, but in comparison to art, reality is rarely
sharply outlined nor do the details necessarily have harmony. Reality stimu-
lates more senses at once and creates a more multifarious sense impression
than works of art, and therefore, it is less suited for providing a limited and
integrated sense impression to reflect an existential theme.
Art has often been claimed to be a mirror of its contemporary culture.
Form and content have changed over time and it is only reasonable to believe
that these changes are caused by new existential challenges in the society. It
is easy to detect a shifting focus on spirituality and religion, individual iden-
tity, powers of the unconscious, technology, and so forth. An increase of tech-
nological inventions in our world has accounted for an increase in existential
challenges during the last century and this has also been followed by an in-
18 / Aesthetics and Innovation
crease of changes in artistic expressions. Assuming that new existential chal-
lenges inspire artistic innovations it might be said that art provides the art
audience with an aid for retaining the existential theme in question. When
current existential themes are objectified through works of art these objecti-
fications make it possible for the audience to constitute and retain new emo-
tional qualities.
I could put forward further details of the reality of art and art appreciation
in an attempt to evaluate my thesis, but let me instead return to my aesthetic
experience at the Tate Gallery.

My experience with Rothko’s works has had a profound impact on me. In
retrospect, I see that it gave me a new insight that I was not yet able to grasp
or determine, though I was aware of the elevated state of being that it left me
with. It was not until many years later that I realized how I was actually influ-
enced by the experience. I wrote an article with the title “Eternity” (1990)
after a visit to the Grand Teton National Park in America, and I realize now
many years later that this article reflects an interest in the infinite as a spiritu-
al dimension. When I wrote the article I did not relate it to my experience
with Rothko’s paintings, but today I am convinced that his paintings provided
a distinct form for an existential theme in my life. A few years ago I visited
the Rothko Chapel in Houston, Texas, and the Matisse Chapel in Vence,
France, in order to carry out studies on the relationship between contempor-
ary art and spirituality—a research project that I am still working on today.
What seemed to be an accidental incident at the Tate Gallery became a driv-
ing force in my life as a scholar.

Alphen, E. van (1992). Francis Bacon and the loss of self. Cambridge: Har-
vard University Press.
Augustine (1961). Confessions. London: Penguin Books. Original work pub-
lished ca. 400.
Beardsley, M. C. (1981). Aesthetics: Problems in the philosophy of criticism.
2nd ed. Indianapolis: Hackett.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience.
New York: Harper and Row.
Aesthetic Experience / 19
Csikszentmihalyi, M., & Robinson, R. E. (1990). The art of seeing: An inter-
pretation of the aesthetic encounter. Malibu: J. Paul Getty Museum and
Getty Center for Education in the Arts.
Dewey, J. (1934). Art and experience. New York: Minton, Balch.
Dufrenne, M. (1973). The phenomenology of aesthetic experience. Evanston:
Northwestern University Press. Original work published 1953.
Elkins, J. (2001). Pictures & tears. New York: Routledge.
Funch, B. S. (1990). Eternity. Psyke & Logos, 11, 312–328.
Funch, B. S. (1997). The psychology of art appreciation. Copenhagen: Mu-
seum Tusculanum Press.
Janson, H. W. (1995). History of art. Fifth ed. Revised and expanded by An-
thony F. Janson. New York: Harry N. Abrams.
Kierkegaard, S. (1980). The sickness unto death: A Christian psychological
exposition for upbuilding and awakening. Princeton: Princeton University
Press. Original work published 1847.
Laski, M. (1961). Ecstasy in secular and religious experiences. Los Angeles:
Jeremy P. Tarcher, 1990.
Lindauer, M. S. (1981). Aesthetic experience: A neglected topic in the psy-
chology of the arts. In D. O’Hare (Ed.), Psychology and the arts (pp. 29–
75). Brighton: Harvester Press.
Maslow, A. H. (1968). Toward a psychology of being. 2nd ed. New York: Van
Nostrand Reinhold.
May, R. (1975). The courage to create. New York: W. W. Norton.
May, R. (1991). The cry for myth. New York: W. W. Norton.
Neumann, E. (1989). Creative man and the “great experience.” In E. Neu-
mann (Ed.), The place of creation: Six essays (pp. 131–202). Princeton:
Princeton University Press. Original work published 1959.
Pamuk, O. (1997). The new life. New York: Farrar, Straus and Giroux.
Panzarella, R. (1980). The phenomenology of aesthetic peak experiences.
Journal of Humanistic Psychology 20, 69–85.
Richards, I. A. (1925). Principles of literary criticism. London: Routledge
and Kegan Paul.
Rilke, R. M. (1989). The best of Rilke. Hanover, NH: University Press of
New England. Original work published 1906.
Sarraute, N. (1958). Portrait of a man unknown. New York: George Braziller.
Original work published 1948.
Tillich, P. (1955). One moment of beauty. In J. Dillenberger (Ed.), On art and
architecture by Paul Tillich (pp. 234–235). New York: Crossroad Publis-
hing Company, 1987.

Objective and Subjective Components

of Aesthetic Experience

Natalja V. Nekrylova

Art perception and aesthetic experience are the central notions of the present
research. We consider art perception as a complicated multilevel process, the
highest level of which is a meaningful process accompanied by a subjectively
significant experience. This chapter describes the results of art perception re-
search applying the combination of an objective psycho-physiological meth-
od, registration of galvanic skin responses (GSR) and a qualitative research
method, a phenomenological interview. With such a combination we man-
aged to carry out a systematic study of art perception at the bodily, behavi-
oural, and personal levels, to describe three levels of art perception and to
define some factors of aesthetic experience.

Art perception is a complicated multilevel process of personality and art-

work interaction, which can vary by the degree of intensity and personal in-
volvement. According to humanistic, existential, and phenomenological psy-
chology, we consider the highest level of art perception as a meaningful pro-
cess accompanied by a subjectively significant experience.
The central notion of our research is aesthetic experience. In the phe-
nomenological tradition, experience is considered a sense unit characterized
by the phenomenon according to which the observed objectivity is both a no-
tion and a part of life experience. The concept of experience is highly ranked
in aesthetics. For instance, according to Gadamer, “Artwork can be defined
as its ability to become an aesthetic experience” (1988, p. 114). Many re-
searchers in the field of psychology of art also note the role of sense and per-
sonal involvement in processes of interaction with art (May, 1975; D. Leon-
tiev, 2000; Znakov, 2000).

Aesthetic Experience / 21
Thus, aesthetic experience is defined as a process of sense actualization,
in which the experience itself is both a prerequisite for sense occurrence and
a reply to it, while sense is a way for the actualization of experience.
We should distinguish comprehension achieved in the aesthetic experi-
ence process and comprehension achieved during reflection delimited from
personality. Even in the case of professional valuation of an advanced expert
without individual involvement and beyond the dialogue between personality
and artwork, the work of art itself is just an object under observation and in-
tellectual evaluation. On the other hand, as Leontiev (2000) noticed an emo-
tional reaction without any efforts to understand it also does not lead to art-
work comprehension, as far as personality is just a passive object of artwork
influence. It is obvious that aesthetic experience does not belong only to a
cognitive or emotional process; rather, it represents a synthesis of both, res-
ulting in a new quality of consciousness and existence.
Thus, at the level of aesthetic experience, person and artwork relations ac-
quire special characteristics. First, the artwork perception process accompan-
ied by an aesthetic experience transforms it from being evaluative to being
valued. Second, in the process of aesthetic experience, special “disinterested”
or relations free from pragmatic interest between person and work of art ap-
pear. The present peculiarity of aesthetic experience was mentioned in Bul-
lough's conception of Psychical Distance (1912). Third, in the course of aes-
thetic experience, the subject-object dichotomy is shifted. Person and artwork
relations are developing as those of a dialogue (Bakhtin, 2000) with each
party being independent and self-valued. Such an attitude is an encounter pre-
condition (May 1975), in the course of which personality opens its own exist-
ential themes via the work of art itself.
In order to study the subjective and objective components of aesthetic ex-
perience, we conducted empirical research, and its process and results will be
described later. The research was based on Vincent Van Gogh’s art.
The method is an original combination of an objective psycho-physiolo-
gical method, registration of galvanic skin response (GSR), and a qualitative
research method, a phenomenological interview. With such a combination we
managed to carry out a systematic study of aesthetic experience at the bodily,
behavioural, and personal levels of development, taking into account the ex-
amined process’ complexity.
The empirical research itself was preceded by a study of Vincent Van
Gogh’s letters. Based on biographical research, several paintings were selec-
ted, which, according to the artist himself, imparted those things which were
given to him by nature or model and which evoke a response in the viewer.
Seven paintings were selected.
22 / Aesthetics and Innovation
At the first stage, copies of Van Gogh paintings were given to the parti-
cipants, one after another. GSR was registered during the perception of each
painting. After that the participant was asked to evaluate each painting ac-
cording to two criteria: 1) whether he/she liked or disliked it; and 2) whether
it impressed him/her or not.
At the second stage, after a 2-3 day break, each participant was asked to
describe the paintings which he/she ranked the highest and the lowest, and
the painting which caused the greatest change of GSR. The discussion was
organized according to the structure of a phenomenological interview without
any external disturbances such as outside people and time limits. The re-
searcher made efforts to create a reliable atmosphere, so that each participant
could feel safe and share his experience of art appreciation sincerely. The de-
scriptions were recorded by a voice recorder.
In total there were 11 subjects participating in this study: men and women
aged 21-28 years old. 51 reports were made in the course of the phenomeno-
logical interview.

Results and Discussion

Levels of Paintings Perception

At first, six types of descriptions were singled out in the course of the re-
ports analysis. Type of description means a complete unit of expression re-
flecting this or that level of viewer and artwork relations:
1. Evaluation of the painting techniques (composition, colors, and mas-
2. Description of the painted objects, their locations, the painting’s theme;
3. Description of the image, the spirits of the image, or the painting as a
whole, and the interpretation of the artist’s idea;
4. Emotional reaction and an appreciation of the painting’s attractiveness;
5. Individual meanings: remembrances, and associations appearing while
perceiving a painting;
6. Aesthetic experience and the expression of meanings arising in the pro-
Each description type can more or less be found at any level of art percep-
tion. The level itself is defined by domination of this or that type of descrip-
tion. We distinguished and described three levels:
Aesthetic Experience / 23
The first level is mainly an intellectual way of art perception, character-
ized by the detached “expert” position of the viewer toward the painting.
Here, descriptions of the first, second, and third types prevail. Art perception
proceeds as a process of external observation or expert evaluation. The view-
er interprets the author’s meanings, evaluates an artistic composition or tech-
nique, etc., not being personally involved. Art perception at this level does
not cause any significant changes of GSR.
Description by one of the participants of the painting “Sorrow” can illus-
trate this level. In this case, attention is paid mainly to the evaluation of the
painting technique: “A fine color spectrum and nice graphics. The painting
looks as if it is a just a sketch. The graphics and the painting itself are warm
—due to it being a pencil drawing, you have the feeling of something alive,
warm, embodied. Accents are made in a very interesting way: an empty level
corner and a slightly overloaded upper side. The painting is soft, smooth,
thereat it is drawn with one contour. I do not know anything regarding the
meaning of the painting—I do not like posing into another person’s world”.
Let us compare this description with the artist’s opinion. In this quote,
Van Gogh compares “Sorrow” with the painting “Roots”: “Now I tried to put
the same sentiment into the landscape as I put into the figure: the convulsive,
passionate clinging to the earth, and yet being half torn up by the storm. I
wanted to express something of the struggle for life in that pale, slender wo-
man's figure, as well as in the black, gnarled and knotty roots” (2000, p.
The second level is mainly emotional way of art appreciation. This level
is characterized by the fourth and fifth types of description prevailing, and by
some kind of egocentrism regarding the viewer, as far as individual content
becomes relevant in the process of painting perception, thus as if covering the
content of the painting itself. Evaluation of a painting’s attractiveness, emo-
tional reactions without efforts to understand them, and the reproduction of
the viewer’s past experience associating with the painting are the most typical
for this level. Connection with GSR is ambiguous.
A description of the “Night Cafe” painting can be used as an illustration
of this level; in this case the participant is in contact with his own associ-
ations and does not grasp the intent of the painting: “…it is cheerful, there is
billiards in the center, and its atmosphere reminds me of a bar in Prague. It’s
quiet, cosy, and late at night. It reminds me of Prague, meetings with friends,
and billiards. It all has a relation to rest. I’m not in this painting, I am off-
screen. But we will come in and fill this place with noise and fun. The painter
showed his favourite place or vice versa”.
24 / Aesthetics and Innovation
“Night cafe” is one of the hardest and strongest paintings of Van Gogh;
however, it is a challenge for the viewer, as far as it contains nothing to be at-
tractive. This painting is a revolt against deception, against pretended solidar-
ity, which in reality means desperate loneliness and aloofness. “In my paint-
ing of the All-Night Café, I've tried to express the idea that the café is a place
where one can ruin oneself, become crazy and criminal. Through the contrast
of delicate pink, blood red, and dark red, of mild Louis XV and Veronese
green against the yellow-green and stark blue-green tones—all this in an at-
mosphere like the devil’s inferno and pale sulphurous yellow… I’ve tried to
convey the sinister power of such a place” (2000, p. 543).
The sixth type of description corresponds to the third level occurring
much rarer than the first two levels and significantly differing from them. Art
perception here is a process of aesthetic experience, and is accompanied by
high personal involvement and subjective significance for the viewer. At this
level, change of GSR was recorded almost without exception.
In the process of aesthetic experience, behavior of the participants is char-
acterized by some features: the tempo of speech slowing down; pausing
while searching for a word or an image which expresses the experience in the
most precise way. The participants face such a problem - common words and
definitions are insufficient for the description of feelings and meanings nu-
ances. That is why they start using metaphors and images to express the con-
tent of their experience.
The viewer is absorbed in the process, truly forgetting about time. Such a
phenomenon is mentioned in the phenomenological description of the flow
experience (Csikszentmihalyi, 1990) and is called the “loss of ego” phe-
nomenon. It is worth noting that the process is accompanied with a feeling of
pleasure as a part of aesthetic experience.
Description of the painting “Road with Cypress and Star” is an example
of an aesthetic experience of the viewer being in fine resonance with the
painting: “For a moment it seemed to me, that the painting turned something
over inside me; some kind of space sensation. For a fraction of a second I
saw a sort of mirror corridor. It seems, as if it is a reflection in water, a feel-
ing of irreality. Something on the verge of reality and unreality. (Pause). The
painting can be perceived as a plane, but you can also see the depth.
(Pause). A sensation of constant motion, change. (Pause). Something related
to the world’s birth. The world is changeable, constantly appearing anew.
The lines are fluctuating, moving as if they can draw you inside. (Pause). The
painting provokes a feeling of the world’s renewal. The painting itself is mov-
ing; it gives an opportunity to see the world in this way. Via this painting, you
find new depths inside you and an opportunity to see the world this way.
Aesthetic Experience / 25
Thereat it does not matter what in particular is painted, this is just a part of
the world and an edge in it”.
Landscapes with cypresses were one of the main topics of Van Gogh’s
paintings when he was in San-Remy. At that time, during a long-lasting
crisis, life and death issues were reflected in his works. Cypresses are tradi-
tional trees in Southern cemeteries. But his paintings with cypresses are not
so unambiguously related to death. Van Gogh had an exalted relation to these
trees; he tried to express their strength, tenacious aspirations high into the
sky, exertions in contrast with their seeming immobility and the quiet sky.
This contradiction of tension and peace expresses the essence of these trees
and life’s mystery.

Correlation of Aesthetic Experience Levels and GSR

The bar chart (Fig. 1) shows how often this or that type of painting de-
scription occurs depending on the presence (GSR) or absence (no GSR) of
GSR changes.
It turned out that in the group “no GSR”, intellectual types of descriptions
prevail, and the amount of descriptions of an emotional level is lower, and the
level of aesthetic experience is actually absent.
The main change in the “GSR” group relates to the increase of aesthetic
experience level share, and we consider this level in particular connected
with GSR changes.
Comparison of the descriptions of “Road with Cypress and Star” (Fig. 2)
and “Starry Night” (Fig. 3) is one of the most demonstrative examples show-
ing a general tendency.
It turned out that in the contemplation of the painting “Road with Cypress
and Star”, all participants had changes of GSR and all of them have appreci-
ated influence of this painting. The largest number of aesthetic experience de-
scriptions related to this painting.
On the contrary, in the contemplation of “Starry Night” received the
largest number of “attractiveness” appreciations, but none of the participants
had GSR changes, and actually there were no descriptions which could be re-
lated at the aesthetic experience level.
On the whole, the chart to the painting, “Road with Cypress and Star”
(Fig. 2) reflects a general tendency typical for the level of aesthetic experi-
Comparing two diagrams we can see that at the level of aesthetic experi-
ence, the number of intellectual level descriptions reduces. This again proves
the fact that GSR change relates to the aesthetic experience level.
26 / Aesthetics and Innovation
Regarding "Starry Night", in spite of the fact that the majority of parti-
cipants appreciated it as “attractive”, almost two thirds of the descriptions be-
long to the intellectual level.

Figure 2-1. Ratio of description types and GSR

Note: TE—technique evaluation; OD—description of the objects pictured; ID

—image description; EA—emotional appreciation; IM—individual mean-
ings; AE—aesthetic experience.

It is interesting that both paintings having similar theme are antipodes not
only for the viewers, but also for the artist. “Starry Night” was painted under
the direct influence of Gauguin's technique and style, and after Van Gogh fin-
ished it, being disappointed with the result, he said, that this painting does not
tell him anything. On the contrary, “Road with Cypress and Star” was highly
appreciated by the artist, as far as he managed to show the “spirit of these
mysterious trees”. We can suppose that our participants were sensitive to the
processes accompanying the paintings’ creation.

Relation of Artwork’s Attractiveness and Impact Factors

Regarding the factors of the artwork’s attractiveness and impact particip-

ating in the artwork’s perception, we had two questions: first, to what extent
these factors contribute to or detract from the development of aesthetic exper-
ience, and, second, how the evaluation of a painting’s impact comprehended
by the viewer correlates with the real influence shown by the GSR.
We have identified some cases of a lack of comprehension of the paint-
ing’s influence. There were situations when the viewer attributed the influ-
Aesthetic Experience / 27
ence to the painting, but during its perception there was no change of GSR.
And there were cases when the viewer denied the painting’s impact, but
changes of GSR were found.

Figure 2-2. Ratio of description types of “Road with Cypress and Star”

Note: TE—technique evaluation; OD—description of the objects pictured; ID

—image description; EA—emotional appreciation; IM—individual mean-
ings; AE—aesthetic experience.

We have come to the conclusion that if the viewer does not comprehend
the influence of the painting aesthetic experience not follow and art percep-
tion occurs at the intellectual or emotional levels. Art perception is empty and
meaningless if neither the viewer realizes a painting’s impact nor the GSR
has any changes.
Evaluation of aesthetic attractiveness has no relation with GSR dynamics
and does not influence the aesthetic experience actualization. Moreover, in
some cases, positive assessment of aesthetic attractiveness of an artwork pre-
vents understanding of its sense. For example, Van Gogh intentionally viol-
ates the principles of aesthetic attractiveness in his painting “Night Café”.
The artist called it “ugly” and simultaneously “owing the same strength as
Dostoevsky’s works”. In our research, those who perceived it as aesthetically
attractive did not understand its sense. On the other hand, a negative evalu-
ation of this painting does not prevent the appearance of aesthetic experience.
But if the viewer assesses the painting’s impact as very low even if there
were significant changes in GSR aesthetic experience does not appear.
28 / Aesthetics and Innovation

Figure 2-3. Ratio of description types of “Starry Night”

Note: TE—technique evaluation; OD—description of the objects pictured; ID

—image description; EA—emotional appreciation; IM—individual mean-
ings; AE—aesthetic experience.

During the research we have identified three levels of art perception with
descriptions of each of them. We have shown that art perception remains
fragmentary if it occurs only on the intellectual and emotional levels.
We have found out that if there is no change in GSR, the intellectual level
of art perception dominates, so GSR changes are related to art perception at
the level of aesthetic experience.
Besides, our research has shown that the participants successfully differ-
entiated such factors as attractiveness and impact of the painting. The results
confirm the view that the effect of an artwork cannot be reduced to aesthetic
pleasure. In our opinion, these factors are independent from each other and
they organize the act of art perception in various ways. Aesthetic experience
is directly connected with the painting’s impact factor and appears as a multi-
level response, simultaneously shown at bodily, behavioral, and personal
levels. Evaluation of aesthetic attractiveness has no relation with the GSR dy-
namics and does not influence the occurrence of art experience.
Aesthetic Experience / 29
The main conclusion is that the necessary condition of aesthetic experi-
ence is a combination of two factors: the first factor is the presence of corpor-
al reactions expressed by GSR change during artwork perception (objective
factor) and the comprehension of a painting/s impact by the viewer (subject-
ive factor).

Bakhtin, М. М. (2000) Avtor i geroj: K philosofskim osnovam gumanitarnyh
nauk. Saint Petersburg: Alphavit. (Author and hero: To philosophical
bases of the humanities).
Berezanskaja, N. B., & Nekrylova, N. V. (2003). Analiz vzaimosvjazi prots-
cessov sozdanija i vosprijatija hudozestvennogo proizvedenija. In G. V.
Akopov (Ed.), Psikhologija iskusstva (Vol. 3, pp. 80–86). Samara: Samara
State Teacher Training University. (Analysis of interconnection between
processes of creation and perception of art work).
Bullough, E. (1912). "Psychical distance" as a factor in art and an aesthetic
principle. British Journal of Psychology, 5, 87–118.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience.
New York: Harper and Row.
Funch, B. S. (1997). The psychology of art appreciation. Copenhagen: Mu-
seum Tusculanum Press.
Gadamer, H. G. (1988). Istina i metod. Moscow: Progress (Truth and meth-
Gadamer, H. G. (1991). Aktualnost prekrasnogo. Moscow: Iskusstvo. (The
relevance of the beaitiful).
Lekhtsier, V. L. (2002). Vvedenie v fenomenologiju hudozestvennogo opyta.
Samara: XXI vek. (Introduction to phenomenology of art experience).
Leontiev, D. A. (1991). Making sense: The psychological function of art. In
L. Y. Dorfman, D. A. Leontiev, V. M. Petrov, & V. A. Sozinov (Eds.), Art
and emotions (pp. 51–60). Perm: Perm State Institute of Culture.
Leontiev, D. A. (1992). What's Hecuba to us? Basic propositions for a psy-
chological theory of art. In L. Y. Dorfman, D. A. Leontiev, V. M. Petrov,
& V. A. Sozinov (Eds.), Emotions and art: Problems, approaches, explor-
ations (pp. 45–64). Perm: Perm State Institute for Arts and Culture.
May, R. (1975). The courage to create. New York: W. W. Norton.
Van Gogh, V. (1935). Pisma. Moscow: Academia. (Letters).
Van Gogh, V. (2000). Pisma. Saint-Petersburg: Azbuka. (Letters).
Vasiljuk, F. E. (1988). The psychology of experience. Moscow: Progress Pub-
30 / Aesthetics and Innovation
Znakov, V. (2000). Ponimanie huduzestvennoj pravdy. In L.Y. Dorfman, C.
Martindale, V. M. Petrov, P. Machotka, D. A. Leontiev, & G. C. Cupchik
(Eds.), Tvorchestvo v iskusstve—Iskusstvo tvorchestva (pp. 274–289).
Moscow: Smysl. (Understanding of the art truth).

What is Beauty? On the Dichotomy between

Subject and Object around 1900

Volker A. Munz

The subject-object dichotomy has undoubtedly been one of the central philo-
sophical issues in modern philosophy. Whereas rationalists such as René
Descartes or Gottfried Wilhelm Leibniz were defending a subjective concep-
tion of the “I”, representatives of empiricism like David Hume and later Ernst
Mach rejected any substantial approaches both for external objects as well as
for the subject. In his works on aesthetics, Christian Ehrenfels introduces a
concept of beauty that is internally related to the classical dualism between
the mental and the physical. Although his central aesthetical doctrines are
based on Ernst Mach’s bundle theory of elements, which denies any substan-
tial claims and tries to solve the mind-body dualism by a neutral monism,
Ehrenfels established his famous concept of Gestalt qualities, which reintro-
duces subjective components into his understanding of aesthetic judgements.
This paper will discuss some of Ehrenfels’ approaches in connexion with
Mach’s analysis of human sensations with respect to the role of both the
judging subject and the aesthetic objects in question. To put this discussion
into a discursive context, I shall start with a few general remarks on the so-
cio-cultural and scientific background of both thinkers in Vienna and Central
Europe around 1900.

To understand the genesis and development of science and art production,

it seems unavoidable not only to reflect such processes within their immanent
progressions but also with respect to the socio-economic, political, and cul-
tural background. For our purposes this means that we should particularly
look at the historical context within which the cultural and philosophical con-
troversies took place.

32 / Aesthetics and Innovation
First of all, massive economic and social transformations of the Industrial
Revolution led to a radically new situation in Europe. Those changes also had
a massive impact on everyday life, particularly in the urban milieus. Besides
positive impulses such as technical innovations, improvements of infrastruc-
ture, or a very high surplus of market goods, we can also observe quite op-
posite effects that were basically caused by the drastic enlargement of inhab-
itants in the cities. In Vienna, for instance, the population explosion led to a
number of more than 1.9 million people in 1910 as opposed to 840.000 in
1870 (Le Rider, 1990, p. 25). Furthermore, since the effects of the Industrial
Revolution in Central Europe also took place during this period, which was
rather late as compared to the general development in Europe, these parallel
processes led to radical deteriorations, such as high rates of unemployment
and extensive poverty. Social dissents were only logical consequences that
conveyed a serious status of uncertainty and disorientation. Now, the pro-
cesses of highly accelerated social diversification count as one of the domin-
ating phenomena of European modernity in general. Vienna and Central
Europe were, however, also determined by a phenomenon which was hardly
observable in other European regions and that is its peculiar ethnic-cultural
and linguistic heterogeneity. This multiethnicity and multi-linguality basic-
ally determined a peculiar development of horizontal differentiations and it is
important to notice that these horizontal ethnic-cultural differentiations also
exponentiated vertical asymmetries such as social diversifications as effects
of modernism (Csáky, Feichtinger, Karoshi, Munz, 2004, p. 17). The variety
of different languages and religions as well as the multicultural pluralities
thereby caused quite positive and creative impulses with respect to the out-
comes of scientific and artistic production. Karl Acham quite rightly argues
that it was certainly not just an accident that psychoanalysis, phenomenology,
logical empiricism, and Austromarxism or sociology emerged in such a so-
cially and ethnically heterogeneous region (Acham, 1996, p. 42).
These processes did, however, also evoke many conflicts of all different
kinds both on an individual as well as on a collective level. Interestingly
enough, we can observe a rather peculiar way, this pessimistic scenario was
reflected upon in various intellectual circles in Vienna around 1900. One of
the consequences of those different forms of crises was a quite radical change
in various ways of perceiving both oneself and the outer world. This new
awareness and reflection upon the ‘I’ and the surrounding life-world manifes-
ted itself in various different ways. In processes of art production, literature,
and science those new perceptions became particularly obvious.
The philosophical discourse after the second part of the nineteenth cen-
tury was, generally speaking, dominated by a rejection of Kantian metaphys-
What is beauty? / 33
ics and by converging against a new paradigm in the natural sciences. Espe-
cially Gustav Fechner’s introduction of psychophysics (Fechner, 1860/1998)
had a massive impact on new theories of perception and it was particularly
Mach, who was deeply influenced by Fechner’s approaches (Mach, 1919).
The new developments in natural sciences not only effected a new under-
standing of the relation between appearance and reality, but also revitalised
the classical subject-object dichotomy. In his paper on the philosophy of
modernity, Rudolf Haller argues that if it counts as symptoms of disorienta-
tion to particularly focus on questions of self-identity, than those symptoms
became especially obvious at the turn of the twentieth century. Within philo-
sophical discourse, Haller speaks of the two following poles that were repres-
ented: Ernst Mach’s rejection of any real entity of the ‘I’, and his opponents,
who regarded the existence of the first person as a necessary condition for the
pure possibility of knowledge. Those philosophers include Franz Brentano,
Anton Marty, Kasimierz Twardowski, Robert Reininger, and Alois Riehl. Be-
sides those two contraries, we can also diagnose mixed forms with respect to
the relation of subject and object, just to mention Hans Hahn, Otto Neurath,
or Christain Ehrenfels (Haller, 1996, pp. 132–133).
This paper will, however, only concentrate on Ehrenfels’ introduction of
his famous Gestalt qualities as part of his aesthetic theory. In order to under-
stand their genesis and contents, it will be necessary to take a brief look at
Ernst Mach’s conception of elements, because they build the foundation of
Ehrenfels’ aesthetic approaches.

The Unsaveable I
It is basically two of Mach’s central theses, which he elaborates in his in-
troduction to the Analysis of Sensations (Mach, 1996), that are of relevance
for our context (Munz, 2004). The first argument concerns the classic sub-
ject-object dichotomy: In following David Hume’s so called ‘bundle theory
of perceptions’ (Hume, 1985) Mach strictly rejected Kantian things in them-
selves (“Dinge an sich”) both for bodies outside of us as well as for the first
person. According to Mach, all that is given to us by immediate experience is
just qualities such as colors, sounds, smells, spaces, times, etc., what he calls
elements (“Elemente”) or sensations (“Empfindungen”), and those elements
appear to us in all different kinds of combinations. This applies to both our
inner experiences and to the experiences of our body and the objects outside
of us. What is relevant for scientific research is just their functional connex-
ion to each other. Concerning their ontological status, Mach claims them to
be neutral. This is a somewhat new approach, since it does not involve any
34 / Aesthetics and Innovation
existential commitments. Solely depending upon the particular context of
single experiments, we can regard the elements of our sensations as psychical
or as physical. Their ontological identification does not affect scientific re-
search, since we can analyse their mutual relations in purely functional terms.
This ontological monism entails the consequence that expressions such as
“object”, “body” or “I” exclusively refer to different bundles of perception,
which are more stable than others and therefore make us only think, that they
refer to some Cartesian substances, mental or physical. Indeed the Ego is as
little absolutely permanent as are all objects, including our body:
As soon as we have perceived that the supposed unities “body” and “ego”
are only makeshifts, designed for provisional orientation and for definite
practical ends […], we find ourselves obliged, in many more advanced sci-
entific investigations, to abandon them as insufficient and inappropriate. The
antithesis between ego and world, between sensation (appearance) and thing,
then vanishes, and we have simply to deal with the connexion of the elements
α β γ ..., A B C ..., K L M ..., of which this antithesis was only a partially ap-
propriate and imperfect expression. This connexion is nothing more or less
than the combination of the above-mentioned elements with other similar ele-
ments (time and space). Science has simply to accept this connexion, and to
get its bearings in it, without at once wanting to explain its existence. (Mach,
1996, p. 14)
Here, we can clearly see in what way Mach tries to dissolve the dicho-
tomy between appearance and reality. Ontologically speaking, all bundles are
of the same nature. Whether we regard a particular combination of elements
as psychical or physical is therefore no more a question about their form of
givenness but solely about the particular perspective that is involved in differ-
ent experiments. For instance, if we are interested in the dependency of a cer-
tain color from a luminous source, we take it as a physical object. If we,
however, put our focus on its relation to the retina, we regard it as psycholo-
As a consequence of Mach’s rejection of Kantian noumena, the analysis
of objects in terms of elements also holds for his conception of the self. Here,
too, it is only due to a certain way of ordinary conversations that we apply
terms such as ‘I’, when referring to ourselves. Scientifically speaking the ex-
pression ‘I’ does not refer to some mental substance as its meaning, since, as
is the case for our inner experiences, we can never come across anything like
a Cartesian Ego. Introspection only shows us a collection of different ele-
mentary complexes. Hence, Mach’s notion of the ‘I’ is nothing but a logical
outcome of his analysis of sensations within a sensualistic-positivistic tradi-
tion of empiricism:
What is beauty? / 35
If we regard the ego as a real unity, we become involved in the following
dilemma: either we must set over against the ego a world of unknowable en-
tities […], or we must regard the whole world, the egos of other people in-
cluded, as comprised in our own ego […]. But if we take the ego simply as a
practical unity, put together for purposes of provisional survey, or as a more
strongly cohering group of elements, less strongly connected with other
groups of this kind, questions like those above discussed will not arise, and
research will have an unobstructed future. (Mach, 1996, p. 28)
But now, Mach was confronted with the question of how one is able to re-
cognise complex structures such as spatial structures, melodies, etc. based on
an elementary conception of our external world, since those Gestalt experi-
ences are certainly different from an aggregate of sense data. In other words,
what guaranties the unity of such complexes? This problem does of course
also effect the question about the unity of a particular object we actually per-
ceive. To put it differently, how are we, from an elemetaristic point of view,
able to grasp a particular complex formation within a present sense percep-
tion? And second, how are we able to recognise a certain figure, the structure
of which we have already experienced, in a different instantiation of it? This,
too, affects the continuity of what we usually call ‘object’ in space and time.
Mach argued that the recognition of sameness can not be grounded on the
qualities of our ideas, for those qualities are always different. This is of
course due to his theory of elements that only allows immediate present-tense
experiences. On the other hand, according to the laws of psychology, Mach
argues, recognition is only possible as a recognition of the same qualities.
Therefore, we can only imagine sequences of different qualities by assuming
that they are necessarily connected with sequences of the same qualities
(Mach, 1865).
Now, when confronted with the concept of Gestalt, Mach wanted to avoid
categories that do not belong to the elements of our experiences and that
would question his neutral monism. Therefore he tried to retranslate structur-
al experiences in terms of sense data experiences. But when he realised that
Gestalt experiences are not reducible to atomic perceptions, he introduced a
new kind of sensation that would guarantee the necessary connexion and
which he defined in terms of particular muscular sensations (Mach, 1865).
Whenever we recognise the same structure in different bundles of perception,
this is only possible because those perceptions are internally connected to a
particular muscular sensation that always appears when we perceive a com-
plex of sense data, which is structurally identical with another complex that
we experienced in the past. And this explains why we can recognise the same
complex, even though the elementary sense data are different. It is only due
36 / Aesthetics and Innovation
to our human mind and the principle of ‘thought-economy’ that we assume,
according to Mach, that we are actually perceiving a complex sensual forma-
tion. And the muscular feelings might also be correlated with other physiolo-
gical movements, particular reactions of the eye, etc. (Smith, 1996, p. 244)
The weakness of this explanation is, however, quite obvious, for a muscu-
lar sensation is itself simple and perishable in time and therefore we cannot
experience the same muscular sensation within a particular period of time.
Now, the experience of a melody is, however, extended in time. And what
would be the criterion for judging whether we do indeed always perceive the
same muscular sensation in time? Moreover, the muscular sensation would
have to be internally related to every part of a perceived complex in order to
identify it as such (Mulligan, Smith, 1986). Furthermore, we often perceive
certain spatial or visual complexes without thereby experiencing an addition-
al muscular feeling, when we, listen to a particular kind of music. And we
can also experience the same complexes on the basis of quite different empir-
ical data, when we look for instance at a photograph or model of a building
and the building itself. Additionally, we often recognise particular sensual ex-
periences. That is, we identify them as experiences we had in the past. And
this fact makes it possible to recall certain perceptions to our mind without
actually experiencing them. We are, so to speak, able to form ideas of past
impressions to use David Hume’s terminology (Hume, 1985, pp. 49–77). And
these are only a few remarks to illustrate the weakness of Mach’s introduc-
tion of muscular sensations.

Ehrenfels on Gestalt Qualities

Besides his outstanding contributions to value theory, Christian von
Ehrenfels, a pupil of Franz Brentano, was also highly influential in psycho-
logy, economics, evolution theory, and aesthetics. In this paper, I shall con-
centrate on Ehrenfels’ introduction of Gestalt qualities and their role with re-
spect to aesthetic judgments.
Ehrenfels like Mach argues that we do have to assume different qualities
connected with qualities of the same kind. In contrast to Mach, however,
identical ideas (Vorstellungen) are perceptional complexes, which means that
our sense data are not reducible to simple objects of immediate experience.
Ehrenfels allows Mach’s conception of elements or sensations as the basis of
our experience but subsequently reintroduces the subjective-objective dicho-
tomy by proposing a new type of qualities, which he calls Gestalt qualities
What is beauty? / 37
In his paper “Über Gestaltqualitäten” (Ehrenfels, 1988) first published in
1890, Ehrenfels initially argues, that the aspect of duration, as in the case of a
melody, clearly shows that the experience of such a phenomenon cannot be
purely sensual. He thereby refers to Mach, who claims that we can experi-
ence spatial or audial structures such as melodies immediately. This assump-
tion is obviously due to the fact that Mach, in contrast to Kant, introduced
space and time as being relative and as ontologically equal elements com-
pared to colors, sounds, shapes, etc. This is a quite radically new proposal to
the established maxim that space and time are absolute and a priori as condi-
tions for any possible sense experience. Mach further stresses the immediacy
of our perceptions without any further mental activity, which does, according
to Ehrenfels, raise the question whether we receive Gestalten from outside or
whether we produce them by collecting particular elements or complexes of
elements. If the latter were the case we could additionally ask whether
Gestalten, such as shapes or melodies are mere collections of elements or
something new compared to them, which is given together with the element-
ary bundles but is nonetheless distinguishable from them. Since Mach’s ele-
ments are simple and he considers Gestalten as elementary, it obviously fol-
lows that they are something new and distinct. Ehrenfels, however, argues
that the similarities of melodies, figures, etc. by a constant difference in their
spatio-temporal determination prove the existence of Gestalt qualities (Ehren-
fels, 1988, p. 134). He therefore also consequently rejects Mach’s assumption
of Gestalten as being simple. To grasp a melody as a melody presupposes the
mental act of remembering a complex idea. And the similarity of shapes, fig-
ures, and musical themes clearly shows the existence of an additional kind of
qualities, namely Gestalt qualities, because the various instantiations of an
identical form or structure differ according to the particular time and place.
This implies that the assumption of Gestalten as mere sums of spatial or audi-
al determinats is not acceptable. And as we have seen, Mach’s proposal to re-
cognise similarities via additional physical sensations that accompany our
immediate experiences is obviously false, which is shown by the simple fact
that in cases of identical melodies with different tunes, we would not be able
to identify the melodies as identical and similarly different melodies com-
posed with exactly the same tunes would prevent us from identifying the
complexes as different. Rather, our perception of sensory elements and their
spatial identification allow us to grasp a Gestalt as an additional quality, since
we can perceive the same Gestalt in different combinations of elements such
as paintings, photographs, miniaturisations, models, drawings, and so on. Or
we can experience a particular melody on different audial levels, different
transpositions, different kinds of performative acts, media. Those complexes
38 / Aesthetics and Innovation
of elements all have the same structure, or as Ehrenfels calls it, “Tongestalt”.
Those “Tongestalten” allow us to reproduce simple past ideas in analogy to
already experienced complexes. Therefore, as opposed to Mach, Gestalt qual-
ities are not just combinations of elements of our experience but further
unique objects. This means, that they are ontologically relevant.
Ehrenfels defines his concept of Gestalt qualities as kinds of positive
ideas (positive Vorstellungsinhalte) that are bound to the existence of actual
complexes of ideas (Vorstellungskomplexe) in our consciousness. These
complexes themselves consist of distinguishable elements, in the sense that
they can be imagined separately and Ehrenfels regards them as necessary
foundations of Gestalt qualities (Ehrenfels, 1988, p. 136). They differ,
however, from our fundamental sense impressions with respect not only to
their reproducibility, but also to the capacity of our human fantasy to freely
create new tokens of hitherto unexperienced Gestalt qualities.
This leads us to a further distinction between Ehrenfels and Mach. Since
Mach regards Gestalten as sensations, no particular further inner activity is
engaged in when we experience a present structure or form. For Ehrenfels
this assumption is wrong. Rather, we are often not aware of such mental pro-
cesses. To argue that whenever we experience a complex in our mind, which
can form the basis of a Gestalt quality, this quality is also provided without
our influence is proved wrong by simple experience, for we can always per-
ceive tunes without a melody, trees without a forest or colors without pictures
(Ehrenfels, 1988, p. 150). On the contrary, the grasping of particular colors as
a painting or the realisation of a sequence of tones as a melody demands a
very difficult and highly complicated mental activity. Large paintings have to
be grasped in time, by letting our view wander over the canvas. Particular
parts, that we are not actually perceiving, have to be added by our imagina-
tion in order to experience the complex as a complex, as a whole. Although
we do not consciously experience further mental activity in all cases when
grasping a particular Gestalt quality, it does not follow that we do not possess
such ideas, only, we are not able to handle them. Furthermore, in order to
work out the foundation of a “Tongestalt”, we do not only have to simply
listen to the various tunes, but also recall a series of past tunes we have just
experienced. But if we connect the basis of a melody this way, we do not
need a further mental act in order to identify it as a melody. In other cases,
such Gestalt providing activities are complements to particular sensual
foundations that help us to identify the simple sensual elements as the basic
complexes of particular Gestalt qualities. Only then are we able to form aes-
thetic judgments, because those complexes are the necessary foundation of
Gestalt qualities. And even though the same Gestalt qualities are always con-
What is beauty? / 39
nected to the same foundations, the individual human fantasy is able to
change those foundations and thereby also the accordant Gestalt qualities on
the basis of similar external sensual stimuli (Ehrenfels, 1988, p. 153).
Interestingly enough, according to Ehrenfels, those mental processes,
when perceiving a particular object, also explain the differences of aesthetic
judgments: First, we are all confronted with the same aesthetic object, but
this very object can produce different ideas in different minds. We experience
different “Phantasiebilder”, as Ehrenfels calls them. That is, different pictures
in our imagination or fantasy, or we receive different complexes of ideas
from the particular object in question. Second, we have to assume different
individual imaginatory mental activities. In order to experience the beauty of
a piece of art, we have to grasp it as a whole within only one view (“mit
einem Blick umspannen”). And different mental capacities create different
aesthetic attitudes due to the complexity and hardness of such an intentional
act (Ehrenfels, 1986).
Now, if we have elaborated such a capacity as the foundation or presup-
position of all Gestalt qualities, then they are not produced by an additional
mental act. In this case, they are rather added to the complexes by themselves
(“gleichsam von selbst mitgegeben”). So, sometimes, as already mentioned,
we do not experience an additional mental activity, sometimes Gestalt qualit-
ies are amendments to their perceptual preconditions. The same qualities are
thereby always connected to the same foundational complexes of ideas, and
the structure of our experience is based on these Gestalt qualities, which ac-
company the experiences of simple sense data. Everything else depends upon
the individual capacities and efforts to run through such processes.

What is Beauty?
From what we know so far, we can conclude that Ehrenfels obviously re-
gards his Gestalt qualities as both independent and at the same time as ob-
jects of our own imagination. They are not elements of the external world but
rather ‘complexes of our ideas’ (“Vorstellungskomplexe”). In his paper “Was
ist Schönheit” (‘What is Beauty’) from 1906, he particularly tries to support
his subjective-objective approach by introducing his conception of beauty.
The text is written in the form of a dialogue between the author and a fict-
icious sceptic. Ehrenfels starts the dialogue by maintaining that there is such
a thing as absolute beauty. The question of how different objects can have
something (i.e., beauty) in common, leads Ehrenfels to the assumption that
beauty cannot be located within the objects. This shows that he is not defend-
ing a Platonic view by arguing, that differing objects share something univer-
40 / Aesthetics and Innovation
sal, or are, so to speak, instantiations of a universal idea or concept named
beauty’ Ehrenfels remarks: “Absolute beauty is, of course, not a tangible ob-
ject but an object of our imagination (Vorstellung). It consists of similar ma-
terial as does ‘2 + 2 = 4’” (Ehrenfels, 1986, p. 155). He than continues to ar-
gue that absolute beauty is always given, when we grasp a natural law by ap-
plying our feelings, not our mind. This idea is connected with the assumption
that every ‘true and real’ (wirklich und wahrhaftig) object contains a secret
natural law that could be expressed in a mathematical formula. So for in-
stance, music would be nothing but experienced, so to speak felt mathematics
(Ehrenfels, 1986, p. 161).
Now, since we only grasp the laws by applying our feelings, it follows,
that no immediate further mental act is involved. Therefore we can only gain
a vague idea of the inherent structure. (Ehrenfels, 1986, p. 163). Only after-
wards our minds can grasp what art has created. Therefore we cannot formu-
late any presupposed rules for the arts. According to Ehrenfels, absolute
beauty is always given when we realise a particular law by means of our feel-
ings, a law that cannot be understood by our minds, while perceiving a spe-
cific object. But because every law presents something unique in various di-
versities and we can only experience this unity by applying our sensations or
feelings, beauty is only a divination or premonition. Consequently, Ehrenfels
defines beauty as a “supposed unity in the manifold” (“geahnte Einheit im
Mannigfaltigen”). And absolute beauty is accomplished in all those minds
that scent the unity in question (Ehrenfels, 1986, p. 164). Whenever a human
being grasps the inner law that arises from the supposed unity without,
however understanding it, we can speak of an object of absolute beauty. We
have to keep in mind, though, that beauty is nonetheless not part of the beau-
tiful object, but of the realm of our imaginery powers. And depending upon
the capacities of our fantasy, we experience different degrees of beauty. The
sceptical question, how we can prove the unity in the manifold, when we can-
not clearly perceive it, but only assume or anticipate it, the author unfortu-
nately leaves unanswered. Furthermore, the concept of unity in the manifold,
Ehrenfels uses in order to defend the existence of absolute beauty is itself a
notoriously unsatisfactory expression (Haller, 1986, p. 179).
Now with respect to beauty, Ehrenfels continues, all perceivers share the
same taste. This assumption does of course not imply that beauty is reduced
to pure taste but is rather due to the absolute character of beauty. But how is
it possible if we take into account the various subjective emotional or sensual
reactions? Ehrenfels argues that those different experiences are not due to dif-
ferences in taste, but rather to our perceptual capacities, that is our individual
powers of comprehension, fantasy, and memory, as well as the different
What is beauty? / 41
formations of our senses and our appetites. Consequently, possible emotional
or sensual differences can be explained by the fact that one and the same
piece of art produces different complexes of ideas in our minds. And this im-
plies that we not only listen to a piece of music, for example, but that there is
also some inner activity involved. This does of course also allow cases where
absolutely no beauty is experienced by a particular perceiver, according to
the individual’s mental and sensual capacities. Therefore, there are no such
things as absolutely beautiful external objects. Absolute beauty only exists as
a complex of ideas. So, Beethoven’s fifth synphony is not absolutely beauti-
ful, but only in so far, as it can produce absolute beauty in the mind of many
or most people who listen to it. Basically we all possess the same aesthetic
object and if someone experiences it as beautiful, someone else as ugly, this
only shows that both observers receive different complexes of ideas, or to put
it from the perspective of the perceived object, that it produces different com-
plexes in each individual mind. Again, it is important to notice that not a par-
ticular piece of paper contains beauty, but the complex of ideas, it produces
in our minds. This means that absolute beauty only exists in forms of com-
plexes in our imagination or fantasy.
As we have seen, the real bearers of beauty are not external objects but
complexes of ideas, which means objects that are constituted by an individual
recipient while perceiving a particular object. What is demanded is our ima-
gination or fantasy. The real piece of art will only be created in the perceiver-
’s mind (Ehrenfels, 1986: 203–204). The question of what Ehrenfels precisely
means by his concept of “Vorstellungsgebilde” (complex of ideas) is left un-
answered. They seem, however, ontologically relevant and necessary in order
to guaranty the truth or falseness of aesthetic judgments. But, if we take a
closer look at Ehrenfels’ Gestalt qualities, it is quite unclear whether they are
individuals, temporally and spatially determined and subject dependent, or
rather abstract universals, exemplified in the experiences of particular per-
ceivers. This problem is due to the fact that Gestalt qualities are both object-
ive and part of our consciousness. Now, it seems quite obvious that they can-
not exist in a particular human mind, since this would reduce them to a mere
question of subjective taste, which will lead us to a simple emotivism, where
aesthetic judgments are not truth bearers. Questions about beauty would then
just turn into questions about subjective preferences. And interestingly
enough in his value theory, Ehrenfels argues diametrically the opposite by re-
jecting any values as autonomous and adhering to particular valuable objects
(Ehrenfels, 1982). A subjective definition of Gestalt qualities does, however,
contradict Ehrenfels’ assumption of absoluteness and their property as truth-
makers for aesthetic propositions. Now, on the other hand, if we locate those
42 / Aesthetics and Innovation
qualities within the beautiful objects, we are back again to our initial Platonic
problem, about how even categorically different objects such as flowers,
paintings, pieces of music, or faces can nonetheless be ascribed as being
beautiful. Ehrenfels tries to solve those questions by determining Gestalt
qualities as “objects of our imagination” (Vorstellungsgebilde). That is, ob-
jects that are constituted by a particular recipient while actually perceiving a
particular object and by applying one’s fantasy. This implies that the true
beautiful object is only created by the perceiver’s mind. This approach does,
however, at least it seems to me, offer another variant to overcome the sub-
jective-objective dichotomy as in the case of Mach’s neutral monism. In other
words, whereas Mach tries to dissolve the ontological difference by arguing
that the relation between psychical and physical elements can be analysed in
purely functional terms and only depends upon the particular perspective of
the scientist who carries out an experiment, Ehrenfels tries to overcome the
dualism by introducing a concept that contains both subjective and objective
elements as necessary constituents.
One possible way to escape the dilemma in Ehrenfels’ conception of
Gestalt qualities, when trying to unite subjective constituting processes with
an absolute conception of beauty, is proposed by Stephan Witasek, a pupil of
Alexius Meinong. He, too, had to struggle with the problem of objectivity
with respect to aesthetic value judgments. To guarantee the possible truth or
falseness of those judgements, which Witasek defended, he also needed an
additional kind of object besides the external objects of our immediate exper-
ience. And like Ehrenfels, he starts with analysing aesthetic experiences and
quite similarly argues that the bearer of aesthetic qualities cannot be identical
with the real external object (Witasek, 1904, pp. 15–19). But Witasek is much
clearer on the ontological status of such qualities than Ehrenfels. He contin-
ues to argue that they are indeed purely immanent objects. That is, something
that depends upon a mental act. Whereas Ehrenfels sometimes speaks in
terms of ‘grasping’, sometimes in terms of ‘constituting’ with respect to the
possible truth makers, Witasek clearly points out, that the aesthetic qualities
are strictly determined by a particular perceiver. From this it quite correctly
follows that these objects cannot guarantee the objectivity of aesthetic value
judgments. Hence, we cannot speak of beauty in terms of absoluteness.
Therefore, Witasek supports the assumption that aesthetic objectivity is redu-
cible to aesthetic normativity (Witasek, 1903, pp. 353–383). In this sense, an
aesthetic judgment can only be false, in so far as it is not founded on a sensit-
ive reaction according to a particular aesthetic norm. In other words, if a sub-
jective verbal reaction is not like it ought to be with respect to the aesthetic
qualities involved when perceiving a particular object, it has to be judged as
What is beauty? / 43
false. Although this demands a further explication of truth and falsehood in
this particular context, it does at least not raise the question about how
something absolute can at the same time depend upon purely psychological

Acham, K. (1996), Die ‚kulturelle’ Krise der Gesellschaft um 1900 und die
Genese der Sozialwissenschaften. In V. Drehsen & W. Sparn (Eds.), Vom
Weltbildwandel zur Weltanschauungsanalyse. Krisenwahrnehmung und
Krisenbewältigung um 1900, (pp. 39–67). Berlin: Akademie-Verlag.
Csáky, M., Feichtinger, J., Karoshi, P., & Munz, V. A. (2004). Pluralitäten,
Heterogenitäten, Differenzen. Zentraleuropas Paradigmen für die
Moderne. In M. Csaky, A. Kury, U. Tragatschnig (Eds.). Kultur, Identität,
Differenz. Wien und Zentraleuropa in der Moderne, (pp. 13–43).
Innsbruck: Studienverlag.
Ehrenfels, C. von (1890). Über Gestaltqualitäten. In Ehrenfels, C.. von,
Ästhetik. Philosophische Schriften, Vol. 2 (pp. 128–167). Munich:
Philosophia, 1986.
Ehrenfels, C. von (1906). Was ist Schönheit? In Ehrenfels, C. von, Ästhetik.
Philosophische Schriften, Vol. 2 (pp. 201–260). Munich: Philosophia,
Ehrenfels, C. von (1982). Werttheorie. In Ehrenfels, C. von, Philosophische
Schriften, Vol. 1. Munich: Philosophia.
Ehrenfels, C. von (1986). Ästhetik. In Ehrenfels, C. von, Philosophische
Schriften, Vol. 2. Munich: Philosophia.
Ehrenfels, C. von (1986). Über das ästhetische Urteil. (Aus dem Nachlass). In
Ehrenfels, C. von (Ed.), Ästhetik. Philosophische Schriften, Vol. 2 (pp.
155–171). Munich: Philosophia.
Ehrenfels, C. von (1988). Psychologie, Ethik, Erkenntistheorie.
Philosophische Schriften, Vol. 3. Munich: Philosophia.
Fabian, R. (Ed.) (1986). Christian von Ehrenfels. Leben und Werk.
Amsterdam: Rodopi.
Fechner, G. Th. (1860/1998). Elemente der Psychophysik. In R. A. Wozniak
(Ed.), Classics in Psychology, 1855—1914. Bristol: Thoemmes Press.
Haller, R. (1986). Zu Ehrenfels’ Ästhetik. In R. Fabian (Ed.), Christian von
Ehrenfels. Leben und Werk (pp. 172–181). Amsterdam: Rodopi.
Haller, R. (1996). Zur Philosophie der Moderne. In R. Haller (Ed.). Nach
Kakanien. Annäherungen an die Moderne (pp. 103–155). Vienna: Böhlau.
Hume, D. (1985). A treatise of human nature. London: Penguin.
44 / Aesthetics and Innovation
Le Rider, J. (1990). Das Ende der Illusion. Die Wiener Moderne und die
Krisen der Identität. Vienna: Österreichischer Bundesverlag.
Mach, E. (1865). Bemerkungen zur Lehre vom räumlichen Sehen. Zeitschrift
für Philosophie und philosophische Kritik, 46, 1–5.
Mach, E. (1919). Die Leitgedanken meiner naturwissenschaftlichen
Erkenntnislehre und ihre Aufnahme durch die Zeitgenossen. In Sinnliche
Elemente und naturwissenschaftliche Begriffe. 2nd ed. Leipzig: Barth.
Mach, E. (1996). The analysis of sensations and the relation of the physical
to the psychical, London: Routledge.
Mulligan, K., & Smith, B. (1986). Mach und Ehrenfels über Gestaltqualitäten
und das Problem der Abhängigkeit. In R. Fabian (Ed.), Christian von
Ehrenfels. Leben und Werk (pp. 85–111). Amsterdam: Rodopi.
Munz, V. A. (2004), Reception of a philosophical text. A case study.
Newsletter Moderne, 7, 17–23.
Smith, B. (Ed.) (1988). Foundations of Gestalt theory. Munich: Philosophia.
Smith, B. (1996). Austrian philosophy: The legacy of Franz Brentano. Chica-
go: Open Court.
Witasek, St. (1904). Grundzüge der allgemeinen Ästhetik. Leipzig: Barth.

Complementary Relations between Quantitative

and Qualitative Approaches to Research in


Gerald C. Cupchik and Michelle C. Hilsche

The qualitative and quantitative approaches to research have generally been

viewed as incompatible. Qualitative research, following in the tradition of
natural history, attempts to describe and understand multilayered phenomena,
whereas quantitative research adopts the perspective of positivism and looks
for functional relationships between operationalized variables. We propose
that these two approaches are in fact complementary and can in fact help
each other depending on the kind of research questions being posed. The ob-
servational and descriptive talents associated with qualitative research can
help define variables which are then examined within a cause and effect
framework. Three projects are reviewed here which explored (1) the creation
and performance of poetry, (2) the experience of “being moved” by a design
product and (3) the reception of multilayered artworks. The goal is to demon-
strate how accounts of critical episodes and experiences can be structurally
analyzed to infer underlying categories. The frequency of these categories
within each protocol can be determined and interrelationships among them
across the sample can be derived using factor analysis. The relative sensitiv-
ity of each subject or respondent to each factor can then be statistically re-
lated to other behavioural indices. In this manner, the categories derived from
a focused qualitative analysis can be linked with quantitative variables and
the two approaches work together to elucidate phenomena of interest to re-
searchers in aesthetics and other sociocultural disciplines.

46 / Aesthetics and Innovation
Imagine, if you will, an artist and a psychologist walking down a street
one early summer’s evening on their way to a café. To the detached observer
it would appear that they are moving through the same space and experien-
cing more or less comparable realities. This is true at a superficial level but
on closer examination their subjective worlds are endowed with different
meanings. Talking with these two friends quickly revealed their differing sub-
jective experiences. As it turns out, the psychologist was thinking about an
earlier conversation which concerned his friend’s new art project. The artist
was very much in the moment, noticing the dimming sunlight playing off
signs, the mannequins, and 1950s furniture displayed in shop windows. One
person was dominated by reflections on ideas while the other engaged in
visual thinking. This episode reminds us how much we take for granted about
the universal nature of experience, that is, until people tell us about their
lived worlds.
The same argument applies to different ways that people experience art-
works. At a superficial level, a painting or sculpture in a gallery is an entity
with some monetary value attached to it. But the artist will see past its static
presence into the layering of stylistic decisions and gestures which produced
it, the psychologist might consider how its theme or style will affect viewers,
and visitors who happen upon the show might relate the works to their per-
sonal lives or ideas about what art means (Cupchik & Shereck, 1998). Art-
works provide suggestions that stimulate different connections depending on
training, purpose, affective state, and so on. The kind of research strategy that
we adopt when approaching problems in the psychology of art will similarly
depend on the questions that are being asked. We might look for a functional
relationship between general properties of artworks, such as relative com-
plexity, and judgments of liking or arousal. This search for functional rela-
tionships requires the specification of variables and is compatible with a
quantitative strategy. Alternatively, we might want to examine how people
experience their worlds and so adopt a qualitative approach that is more phe-
nomenologically oriented. In other words, we should select our strategies in
accordance with our intellectual goals.
Quantitative researchers focus on functional relationships between opera-
tionalized variables, whereas qualitative researchers use thick description to
achieve a coherent understanding of multilayered phenomena. At first glance,
these two research strategies might seem incompatible. The distance between
them is such “that the claim of compatibility, let alone one of synthesis, can-
not be sustained” (Smith & Heshusius, 1986, p. 4) and the underlying meth-
ods are deemed incommensurable (Lincoln and Guba, 2000). A fundamental
difference between the two approaches lies in their treatment of the keywords
Qualitative Research / 47
“operational measures” and “multilayered phenomena.” A behaviourist who
studies preferences needs to link measurable stimulus properties with verbal
ratings of liking and so on. A constructivist, on the other hand, must specify
the context of an experience such as critical episodes of composing or per-
forming poetry (Hilscher & Cupchik, 2003) and “being moved” by design
objects (Cupchik & Hilscher, in press).
The goal of this chapter is to show that quantitative and qualitative ap-
proaches to research in psychological aesthetics are complementary
(Cupchik, 2001) rather than mutually exclusive. One reason for this is that
the “social order....exists independently of social scientific inquiry (Beach,
1990, p. 217) and the underlying phenomena that researchers explore do not
therefore rely on them for their existence. Artists walk down streets planning
their next projects without a care for the opinions or thoughts of research psy-
chologists or philosophers of art (Avital, 2003). Researchers, either as indi-
viduals or in communities, interpret events in accordance with their intellec-
tual commitments, values, and interests. Accordingly, a behaviourist with in-
formation theory leanings will analyze paintings as stimuli which vary in
complexity (Berlyne, 1971, 1974), while a Gestaltist (Arnheim, 1971) will be
concerned with how compositional forces shape their expressive effects. Pro-
gress in the social sciences as a whole, and in psychological aesthetics in par-
ticular, can benefit from a reconciliation between these two research ap-
proaches. Each has something to contribute and so the question becomes,
how best can they be integrated?
It is also important to note that both communities are concerned with eco-
logical validity, the extent to which a finding meaningfully reflects an event
or process in the world. However, they can each be adversely influenced by
epistemological commitments because of ways they selectively operate on
phenomena. Quantitatively oriented positivists stress operational definitions
which can so deplete a phenomenon of its richness that it all but disappears in
the rush to actuarial precision. Qualitatively oriented constructionists, on the
other hand, can so link a phenomenon with a particular interpretive context
that it runs the risk of being meaningful to a narrow community. The two
kinds of researchers therefore face different problems. The positivist focus on
measurement can transform meaning into nothingness, whereas a constructiv-
ist emphasis on interpretation can restrict a phenomenon to doctrinal commit-
We therefore need a framework for reconciling the two approaches and
extracting the best of what each has to offer. Much is to be gained by recall-
ing that natural history, with its descriptive approach that generates concepts,
precedes experimental science and its emphasis on explanation or prediction.
48 / Aesthetics and Innovation
For our purposes, situating art and literary works back into the contexts of
creation, performance, and reception is part of a natural history of aesthetic
activities. The experimental study of how specific properties of art or literary
works shape dependent measures like preference or memory (Medved,
Cupchik, & Oatley, 2004) can benefit immensely from this preliminary
phase. Natural history and experimental science are complementary in that a
careful descriptive exploration of the natural world is essential to the formu-
lation of meaningful questions and relevant concepts.
The philosopher John Dewey (1934/1969) showed how complementarity
works when it comes to describing artistic creation with his emphasis on the
importance of harmonization between the artist or viewer and the artwork. He
described the active and complementary relations between artistic (i.e., do-
ing) and esthetic (i.e., undergoing) processes that are grounded in experience.
On the one hand, the urge to paint completes an experience for the artist:
“Without external embodiment, an experience remains incomplete” (p. 365).
On the other hand, in order “...to perceive, a beholder must create his own ex-
perience. And his creation must include relations comparable to those which
the original producer underwent.... Without an act of recreation the object is
not perceived as work of art” (p. 367). An important step forward in our un-
derstanding of acts of creation and reception therefore involves reinserting
aesthetic objects in these contexts in order to appreciate processes surround-
ing the emergence of meaning.
A balance between functional and descriptive approaches was also evid-
ent in the earliest work of Daniel Berlyne (1924-1976) who became a thor-
oughgoing behaviourist as a mature researcher (Cupchik, 1988). In his first
publication “Interest as a psychological concept” (Berlyne, 1949), he argued
that "The ability to see external phenomena endowed with 'meaning' is one of
the main functions of both perceptual processes and the pursuit of know-
ledge...It is part of the process of 'assimilation' of new stimuli described by
French-speaking psychologists" (p. 192). The phrase, "endowed with mean-
ing," reflects the influence of Bartlett, his undergraduate and M.A. thesis ad-
visor at Cambridge University. Indeed, the processes associated with
Berlyne's seminal concept of collative variables appear to have been derived
from Bartlett's "'effort after meaning'...the urge to perceive something in
terms of a wider background of past experience and present setting, to com-
pare it with other, more familiar entities" (Berlyne, 1949, p. 192). Judgments
of complexity, orderliness, surprisingness, clarity, etc., are a product of this
comparative process.
One theme which appeared in his first publication but was never pursued
concerned "phenomena" and the "everyday discussion of human behaviour."
Qualitative Research / 49
Berlyne argued, "However psychologists decide to use 'interest', if at all, as a
technical term, they cannot escape the obligation of investigating the various
phenomena with which the word is commonly associated" (p. 189). He
provided examples of disciplines where 'interest' is studied as a phenomenon,
ranging from educational to vocational psychology, psychotherapy and soci-
ology, and of 'interest' behaviour such as "building model boats or solving
cross-word puzzles" (p. 189). This emphasis on phenomena and the descrip-
tion of everyday behaviour is conspicuous by its absence in Berlyne's later
work. In the process of theory building, it was a short hop from Bartlett to
Hull. He saw Bartlett's account of 'schemas' as "strongly reminiscent of Hull's
'anticipatory response', in which, under the influence of a motivating state,
the organism brings forward the most strongly reinforced response..." (p.
186). Rather than studying particular kinds of interest-related behaviour,
Berlyne sought functional relationships between collative properties and the
drive to explore stimuli. We can see in the early work of Berlyne the possibil-
ity of accommodating both viewpoints; a rich description of phenomena com-
plemented by a precise search for functional relationships between variables.

In this next section of our chapter we try to show how qualitative and
quantitative methods can be used in a complementary manner to advance
scholarship in psychology of the arts and creativity. The primary focus will
be on describing a method for deriving concepts or categories which underlie
the structure of experiences. We begin by specifying the context of the exper-
ience, in other words, what the experience is about. The more precisely we
can specify the context, the richer will be the discourse that is collected relat-
ing to the phenomenon in question. Once this discourse is collected, it be-
comes a potential source of data to understand the structure underlying a per-
son’s experience in a particular context. The set of categories which are de-
rived to describe this structure can then be transformed into variables which
are quantified and relations among these variables are determined through
factor analysis. These relations provide a way to describe the underlying pro-
cesses and a statistical component is provided to give a sense for the relative
importance. The factor analysis provides some sense for the relative import-
ance of individual and these factors can then be used to predict behavioural
variables. In this way, qualitatively derived factors can be used to predict
quantitatively oriented measures.
50 / Aesthetics and Innovation
The Collection of Qualitative Data

So long as the experience of a phenomenon is recounted in the form of a

participant-directed anecdotal narrative, a phenomenological qualitative ana-
lysis can be properly applied. It is essential that the telling of the story be
generated and guided entirely by the participant because it is this act of con-
textualizing and describing that will be most enlightening. The research pro-
cess should therefore begin with clear questions such as: “What is it like to be
absorbed in a literary text?”, “What is it like to write (or perform) poetry?” or
“What is it like to ‘be moved’ by a design object?”. Each of these questions
requires the selection of appropriate respondents who have had relevant crit-
ical experiences. An examination of these questions will be rewarding and
productive to the extent that appropriate respondents focus on specific epis-
odes or experiences which provide a focus for their thoughts and feelings. In
this way, participants feel free to express emotions and thoughts that are per-
sonally significant rather than commenting on the researcher’s perspective or
line of questioning. Following the open-ended sharing of life experiences, a
researcher can pose follow-up questions that clarify details left uncertain in
the narrative. However this dialogue should not be used to add details to the
narrative. Aside from checking the facts, this dialogue must be kept separate
from the monologue-style narrative. Themes, ideas and perspectives that are
introduced and discussed during the follow-up questioning may be intriguing,
yet they exist outside the primary narrative context and must be treated separ-
ately during data analysis.

Analysis of Qualitative Data

Random Sampling of Protocols

After having collected a set of paragraphs, interview transcripts or other

participant-generated texts, one can begin the process of qualitative analysis.
The first step in a qualitative analysis is to draw a sample from the overall
data set. This sample must have a balanced number of protocols to represent
any independent variables of interest during the study but at the same time it
is important to randomly sample from the protocols. If, for example, one is
interested in the effects of gender and profession on a certain phenomenon,
then it is important to randomly select the same number of protocols repres-
enting the different groups. Assembling a sample is beneficial because it al-
lows for a preliminary assessment of patterns in the text that isn’t over-
whelmed by the sheer number of protocols that will have been collected. It is
Qualitative Research / 51
best to draw out 15 to 20 protocols from a data set that exceeds 50 total pro-

Seeking Textual Patterns in the Sample

Once a random sample of protocols has been drawn from the entire data
set, one must read through these paragraphs to identify similarities and differ-
ences related to their content, structure, and emotional tone. Content refers to
the subject matter, plot, characters, actions, and setting details that were
provided by the participants in their recounting of life events. Depending on
the episode and individual, more attention may be given to one aspect of con-
tent over another. The structure of the protocols concerns their length, style,
and use of language. Finally, emotional tone of the paragraphs pertains to the
emotions and feelings described and expressed by participants during their
recall of the experience. Based on the characteristics of the sampled para-
graphs, a set of preliminary categories can be identified and should be
defined clearly and illustrated by way of examples drawn from the narratives.
This combination of definition and illustration is very important because it
makes it clear what types of words and expressions qualify as belonging to a
given category. This also facilitates an unbiased and rigorous application of
the categories when one progresses to conducting frequency counts.

The Use of Frequency Counts

A preliminary frequency count must be employed to determine which of

the preliminary categories are likely to be statistically relevant in their dis-
crimination between levels of the independent variables. The frequency count
rules out those categories that merely differ in their presence or absence
across the protocols. Categories that are present once in some paragraphs and
never in others will not be statistically relevant if subjected to quantitative
analysis because they lack the power of frequency. Categories must be
present at differing frequencies across the data set; in some paragraphs they
must be found at great frequency, whereas in others they ought to be barely
present, if at all. The preliminary frequency count is therefore the main tool
used to finalize the set of categories, which may then be applied to the entire
data set. In this way, a qualitative set of paragraphs can be converted into a
numerical data set that is amenable to quantitative treatment.
52 / Aesthetics and Innovation
Balancing Subjectivity and Objectivity

Successful qualitative analysis requires that the researcher be able to bal-

ance a subjective perspective when deriving categories with the rigorous and
objective application of these categories later on. Initially, a researcher must
orient to the text in a way that is open-minded and free from expectations.
Accordingly, the researcher listens to the words of the participants and is
sensitive to the meanings inherent in the narratives rather than imposing as-
sumptions and biases on the exploration of text. Once categories have been
derived, the researcher needs to switch from subjectivity to a structured and
objective mind set so that the categories can be clearly defined and applied
even-handedly to the entire data set. The complementary use of subjective
and objective orientations is essential for the production of meaningful cat-
egories and useful data. We now provide two examples of this process to
show how the generating of categories is accomplished.

Writing and Performing Poetry

(Hilscher & Cupchik, 2003)

Data Collection

The performance poetry circle in Toronto is very active and on a weekly

basis local poets flock to cafés where they take to the stage and bring their
words to life. The purpose of this study was to interview fellow poets about
episodes of writing and performance in an attempt to learn about these exper-
iences in the lived world of poets. Six male and six female poets were inter-
viewed during this study. They selected two of their own poems and de-
scribed the experience of writing and performing each one. The interviews
were conducted one-on-one at a local coffee shop and were tape recorded to
facilitate transcription. Poets spoke without interruption and only after they
had ended the narrative was follow-up questioning used to clarify details. A
total of 48 protocols were collected and to illustrate the types of experiences
described by the participants several examples excerpted from the protocols
are provided here.
Female (writing episode): On one particular night I looked out the front
window and it must have been dusk, but the sky was so big and there was this
tangerine glow because the sun was setting. I went and got my notebook and
then sat with this view of the sky and handwrote the poem. I just started to
describe what I saw and the poem just fell out. I could hardly write fast
Qualitative Research / 53
enough to keep up with the ideas coming to mind. Emotionally I put the isol-
ation I was feeling right into the poem… I was a bit emotional and feeling
close to nature in a way.
Female (performance episode): The experience was filled with many dif-
ferent emotions – anxiety and excitement. I was feeling uptight, but not
scared because I knew I’d practiced the poem before. Once I began speaking
I was thinking about how I was expressing the words – questioning myself. It
was a battle of wills not to get caught up in any mistakes I’d made as I went,
or challenging the way I’d read a certain line. The audience was so warm and
open that I relaxed and went beyond what I’d practiced, tried out new ways
of expressing certain parts of the poem. The audience response was great; it
really helped to shape the poem. Overall I loved expressing myself and being
on stage… it was fun!
Male (writing episode): I didn’t know what to do about [the situation], so
I decided to write him a poem in the attempt to give him a kick in the ass. I
decided to write the poem in a style that would really make my friend angry.
It’s written in an antagonistic way—like rap or hip-hop—sort of typical beat
poetry. It was vindictive because of the way I wrote it.
Male (performance episode): I had wanted to read this poem out loud for
a while. I was very nervous. I got up there and started out by completely con-
centrating on the poem—its words and its flow. I was also making an effort to
project my voice. The poem started out slowly, the audience wasn’t really
into it, so I started glancing at them, trying to synchronize myself to them.
Soon I felt them coming into the poem – focusing on me—and I knew that I
had them! Then I could emphasize whatever I wanted, fluctuate my voice
however I felt. It was that brief moment when you connect with them that
makes you perform again—trying to recapture the connection of that mo-

Qualitative Analysis

In order to generate a set of categories, 4 writing and 4 performance de-

scriptions were sampled from a total of 48 episodes. The writing episodes
were randomly selected from 2 male and 2 female poets and the performance
episodes were randomly selected from 2 other male and 2 other female poets.
The content of these descriptions were compared for similarities and differ-
ences with regards to such things as content, structure, and emotional tone.
Similarities and differences that seemed to distinguish writing from perform-
ance episodes based on a preliminary frequency count were rigorously
defined and 10 categories were selected which were applicable to both types
54 / Aesthetics and Innovation
of episodes (see Table 1). The frequency with which each category arose
within each of the 48 protocols was determined (see Figure 1 for an example
of the coding process) thereby transforming a qualitative set of categories
into a quantitative data set.
Figure 1 is a distilled and simplified version of the coding process as it
only shows the application of three categories (emotion, setting, actions)

Table 4-1. Definition of categories for writing and performance

Category Definition Example

Background Poet provided background “I lived in the Annex for 5

Information —information about past years, and had just moved to
events—relating to them- the Danforth…”
selves and others.
Positive Poet described a positive “I loved expressing myself
Emotion feeling experienced during and being on stage…it was
writing or performance. thrilling!”
Negative Poet described a negative “After I’d read the first few
Emotion feeling experienced during words I was just so embar-
writing or performance. rassed.”
Environment/ Poet described aspects of “I was sitting in a café…it
Setting environment or setting in was a sunny morning…”
which they wrote or per-
formed the poem.
Actions Poet described their own “I went and got my note-
actions during the creative book, and then sat and
episode. handwrote the poem.”
Conscious Poet reported that they “I made an effort to project
Intention used an intentional strate- my voice.”
gy, or that they had goals
in mind to be achieved
through writing/perfor-

Table 4-1. Continuation

Qualitative Research / 55
Spontaneity Poet reported spontaneous “When I was on the subway,
inspiration, or the feeling of the first line of the poem
an urge or impulse to write just came to me out of no
or perform. where.”
Orientation Poet reported that they fo- “I got very focused and
To cused on the poem’s words couldn’t stop thinking about
Poem and meaning. each sentence as I wrote it.”
Orientation Poet reported that they fo- “I looked out into the audi-
To Audien- cused on the response of au- ence…”
ce dience members.
Meaning Poet reported that “The poem helped me to
writing/performance clari- find some answers.”
fied some question they
had, or gave them a new

across an excerpted section of narrative. Usually frequency counts are con-

ducted for 8 to 10 categories across lengthy protocols. Furthermore, a cross-
over of categories can be expected, meaning that a single word or phrase can
be coded as belonging to more than one category.

Factor Analysis

Two factor analyses demonstrated how the same ten categories were
clustered to represent writing and performance episodes.
Writing episodes. Four factors described the basic processes underlying
the creation of poems. Factor 1 was named creative embodiment and demon-
strated that poets primary emphasized that meaningful writing was linked
with spontaneous action. Factor 2 focused on the setting in which writing oc-
curred. Factor 3 described pleasant emotions associated with the writing ex-
perience. Like Factor 2, this factor was concerned with aspects of the episode
unrelated to the actual writing process. Finally, Factor 4 referred to the inten-
tional use of writing strategies focusing thereby on effort rather than spon-
56 / Aesthetics and Innovation
Performance Episodes

Three factors were found to underlie the public performance of poetry. In-
terestingly, Factor 1 represented the complementary roles of spontaneity and
intention in the performance of poetry. It encompassed both Factors 1 and 4
that were distinct during the writing phase. Poets described the experience of
spontaneity in addition to intentional strategies, felt pleasure, found meaning,
and were oriented to the poem. Factors 2 and 3 contrasted the experiences of
negative and positive performances. Factor 2 described a negative perform-
ance experience and considered features of the audience that might be to
blame as opposed to weaknesses in their own roles as performers or poets.
Meanwhile in the case of Factor 3 poets referred to a positive performance
experience by alluding to the environment and audience.

Figure 4-1. Example of coding process for poetry study

Summary of Results

The results confirm the presence of spontaneity and strategic intention in

both creative writing and poetic performance. However, the two dynamics
exert independent influences during the writing of poetry with spontaneity
being of predominant importance. On the other hand, the two processes work
in concert to shape a poetry reading. When poets take to the stage they are
likely to have a practiced routine prepared for performance. The existence of
this plan provides a foundation upon which to attempt a spontaneous interac-
tion with the audience. Preparation provides a basis for spontaneity. Despite
this difference between creative writing and performance, emotional engage-
ment is the primary connection that a person has with poetry. The comple-
Qualitative Research / 57
mentary use of qualitative (category development) and quantitative (factor
analysis) techniques has been useful in showing the complex processes un-
derlying poetic creation and performance.

‘Being Moved’ by Design

(Cupchik & Hilscher, 2006)
Data Collection

The same general approach was applied to the question “what does it
mean to ‘be moved’ by design products?” Whereas the earlier study asked
poets to describe significant episodes involved in creative writing and per-
formance, this second study had designers focus on personally chosen design
products. The participants were industrial design students enrolled as Master-
’s students at the Delft University of Technology and professional designers
who were employed by Dutch design firms. In total, 8 design students (4
male and 4 female) and 8 professional designers (4 male and 4 female) were
interviewed one-on-one about their experience of “being moved” by design
products. Each participant described two design products which they experi-
enced as emotionally significant. A total of 32 protocols were collected. To il-
lustrate the types of experiences described by the participants several ex-
amples excerpted from the protocols will be provided:
Male student: [My guitar] is played with my fingers so there is a lot of
close contact. Every time I play it it’s not like I’m just playing a device. I feel
some kind of life—it reacts—it gives me feedback. I want to say that its
design lets me play it differently than something like an electric guitar—it
has nylon strings so it has a softer feel, it is for two hands, it is bigger so you
can hug it while you play, it is also a fragile thing so you have to take care of
it. The case is twice as expensive as the guitar, which means something.
Female student: [The lamps of Ingo Maurer] move me because they are
made of different materials and everything has a little joke in it—yeah. The
lamps are often very fragile—made from paper and he plays a lot with light
diffraction. He uses materials and objects you wouldn’t associate with a
lamp. His design puts me in a certain mood and he doesn’t do it obviously.
His designs refer to something that people like in a simple way. Allow us to
consider life. He doesn’t make things obvious, which would give me the an-
swer and not let me think for myself. I want to figure it out.
Male designer: The [MadAss motorcycle] is a prototype so you can’t buy
it and I wouldn’t buy it because it is probably not useful or practical. But the
58 / Aesthetics and Innovation
design speaks of it as a cool motorcycle. I’m enthusiastic with the idea of
how I would feel riding this motorcycle. Probably it wouldn’t be very com-
fortable because it is a motorcycle that makes you sit straight up—so it isn’t
comfortable to ride—but it looks rough, it’s a naked bike so it doesn’t have
any ornaments on it. The steam tanks aren’t closed so all the heat from the
engine comes right in your face. It’s like riding an engine and that is cool.
Female designer: I feel like I’m on top of the world when I sit on [this
couch]. It makes me feel energized and aware of the things around me. It’s a
strong feeling. This is something in the design—it is like an animal preparing
to pounce—a deer high on its feet as though it is preparing to make a little
jump at any moment. This gives the couch the appearance of being awake
and alert. When I want to read a book it makes me feel like I’m really con-
centrating because the couch is very alert.

Qualitative Analysis

In order to generate a set of categories, 10 paragraphs were sampled from

the entire data set. The independent variables (gender and experience) were
represented in the sample. Similarities and differences across the paragraphs
were noted as preliminary categories, which underwent a frequency count to
help finalize the list of categories. Nine categories (see Table 2) were derived,
formally defined and used to evaluate the entire series of protocols, resulting
in the production of a quantitative data set (see Figure 2 for an example of the
coding process). Once again, Figure 2 is a distilled and simplified version of
the coding process as it only shows the application of three categories (emo-
tion, setting, actions) across an excerpted section of narrative. Usually fre-
quency counts are conducted for 8 to 10 categories across lengthy protocols.
Furthermore, a cross-over of categories can be expected, meaning that a
single word or phrase can be coded as belonging to more than one category.
The nine categories applied across the protocols included, for example:
emotions and feelings described by the participant that related to the design;
comments by the participant indicating that the design facilitated their sense
of identity by allowing them to present themselves more truly to others or to
better explore themselves. A third category pertained to social meaning in
that participants often indicated that the design object had value because it
had been given to them by someone special, reminded them of a significant
other, or facilitated their socialization with others. Other participants spoke
about the unique qualities of a design that would make it a collectable; it
could have been part of a limited edition, or designer-signed. Another com-
mon category was the description of an interaction between the participant
Qualitative Research / 59
Table 4-2. Being moved categories

Category Definition Example

Emotions The participant describes an “…[the art exposition] made

emotion they associate with me happy because I didn’t
the design or a feeling they think it was possible…”
had while experiencing the
Sense of The participant indicates “[My I Pod] is completely
Identity that the design allowed them filled all the time so I can li-
to present themselves more sten to any music I feel like
truly; to better express their at the moment. On the train
personality; to explore who I can listen to comedy.
they are; to be flexible in When I'm riding my bike I
their lifestyle. The design want some up-tempo music.
affords a sense of unique- When I'm with my girl I
ness. Lets them introspect & want more romantic, easy-
come to know themselves. going songs...”
Social Mea- The participant indicates “[My guitar] is my sort of
ning that the design object has link between my world and
social value in that it was my family... Every time
given to them by someone when I go to my parent's
special, reminds them of house I take it out and play
somebody, or facilitates so- it.”
cializing with others.
Collector’s The participant describes “The lamp itself is quite
Edition the design object as rare, special because it is a lim-
hard-to-find, unique, part of ited edition and it has a
a collector's edition, or de- number and is signed by the
signer-signed. designer.

Table 4-1. Continuation

Interaction The participant describes in- “I didn't really notice the
teracting with the product design of [the couch] until I
60 / Aesthetics and Innovation
(or an imagined interaction). sat on it and then I began to
The object isn't just doing become more and more
something for them, rather aware of how it made me
they are doing something feel.”
with the product.
Physical The participant describes “The [Citroen DS] has nice
Features & the physical design of the lines. One line travels from
Function object or its functions. the front of the car, along
the edge of the roof and
ends in lights at the back.
This car has a nice appear-
ance but yet it is a fast car
with a strong engine.”
Symbolism The participant discusses “Maybe in years to come I
the meaning of the design. will understand the design
They may describe it as sur- of this building because
prising, intriguing, an en- right now I don't completely
igma or they may interpret understand it.”
what it means to them.
Ascribing The participant describes “When I play [my guitar]
Human the design in human terms; it's not like I'm just playing
Traits a “friend”, “cute”, a device. I feel some kind of
“friendly”, “fun”, life—it reacts—it gives me
“playful”... feedback.”

Simplicity Form follows function. The “[The lamp] needs to be

participant describes the natural, no ornaments, just a
product as simple, basic, es- plain shape.”
sential, etc.

and the design piece itself. In this case the object wasn’t merely doing
something, rather the participant felt that they were doing something with the
design. Many participants ascribed human characteristics to their chosen
design piece, as though it were a friend or companion while others followed
the idea that ‘form follows function’ by discussing the product’s simplicity.
Qualitative Research / 61

Figure 4-2. Example of coding process for episodes of being moved by design

Looking at Multilayered Art

Cupchik & Gignac (in press)
A third study shows how artworks differing in the degree of layering
stimulated the aesthetic experiences of viewers. Three different styles were
chosen from the work of Rochelle Rubinstein, an artist living in Toronto. The
Superimposed images were blended and ambiguous with the same area shar-
ing two different and yet integrated motifs. These artworks were created, for
example, by photographing images printed on fabric that were laid across the
artist’s body.
The Combined-composite images juxtaposed motifs that were clearly sep-
arated, suggesting a story and inviting connection. These motifs contrasted
adult and child related imagery in the foreground and background, respect-
ively. The Graphic images were woodcuts or linoleum block cuts created
earlier in the artist’s career which directly expressed powerful emotional
themes in a clear manner. The Graphic images were the least layered of the
three groups.
Thirty participants were presented with 6 examples of each style and were
instructed to select two artworks for an in-depth discussion. Both quantitative
and qualitative data were collected. Subjects rated the images on a series of
seven 7-point scales that measured, pleasingness, personally meaningful, lik-
ing of subject matter versus style, evoking of fresh emotions versus emotion-
al memories, locus of emotion in the artwork or in the viewer and judgments
of the artworks on the scales warm versus cold and unified versus incoherent.
62 / Aesthetics and Innovation
They also provided reasons to account for their choices which were tape re-
corded and transcribed for content analysis.
For the qualitative analysis, the overall sample space included 180 pieces
of discourse (30 participants X 6 accounts, 2 for each of the three styles). The
experimenter examined a subset of the interview material from 6 respondents
and developed a category system to describe the structure underlying the dis-
courses. Ten categories were selected that appeared with varying frequency
across the respondent accounts of reasons for selecting the artworks. The in-
terviewer assessed the frequency with which each category appeared for each
artwork with the provision that more than one category could be assigned to
the same phrase if it fit the definition. Sample categories included, for ex-
ample, projection, mystery, interpretation, anxiety, reference to subject matter
or style, and personal growth.
Separate analyses of variance were performed for the 7 verbal rating
scales and the 10 categories treating types of artworks as a within-subjects
variable. Significant main effects were found for three of the qualitative cat-
egories involving comments about, Subject Matter, Personal Growth and
Style. The Combined-composite images elicited the greatest number of com-
ments about Subject Matter and Personal Growth. The Combined-composite
images also elicited fresh emotions rather than emotional memories on one of
the 7-point scales. Having to resolve relationships between thematic layers in
the Combined-composite images appears to have stimulated thoughts about
changes in the subjects’ own lives. Subjects, who according to a question-
naire were disposed to becoming absorbed in all kinds of art, literary, music
and film works, were particularly responsive to these layered images. In con-
trast, Graphic images elicited personal memories according to the 7-point
scales and the qualitative data revealed that subjects were consciously aware
of the style of the chosen artworks. These effects suggest that the direct style
of the Graphic images stimulated episodic memories and subjects in a negat-
ive mood state appreciated them.

Final Comments Regarding Qualitative Analysis

The three studies summarized here involved very different kinds of stimu-
lus materials but together demonstrate the value of applying qualitative ana-
lysis to discourse about aesthetic materials. The studies all began with a clear
sense for the task, whether it involved poets describing episodes of creation
and performance, designers choosing products that were personally “moving”
or viewers giving reasons for selecting particular artworks. The power of the
research strategy lies in having people provide verbal accounts of events or
Qualitative Research / 63
things that are especially meaningful to them. This lends coherence to a data
set and makes it easier to find categories underlying the structure of their dis-
course. In our approach, the categories always emerge from the protocols and
are not imposed for theoretical or pragmatic reasons. There are, of course, al-
ternative strategies which identify categories a priori and examine their pres-
ence for specific subject groups in response to particular stimuli (see László
and Fülöp, this volume).
Different phenomena bring up different themes and content, and are rep-
resented by different patterns of emotion and action. As such, qualitative ana-
lysis is an ideal approach because it combines a subjective, text-oriented per-
spective that is sensitive to patterns within narratives with an objective ap-
plication of categories that formalizes and compares the frequency of these
patterns between narratives. It is important to appreciate that there is a certain
artistry underlying the naming of factors based on weights attached to the dif-
ferent categories. This leap has an interpretive aspect and there is room for
misidentification or the intrusion of hidden assumptions and hypotheses. The
development of categories and interpretation of factors should therefore be
seen as part of an overall strategy for understanding basic processes. The
value of a set of categories is established when they are meaningfully
grouped through factor analysis and these factors are then used to predict oth-
er behavioural responses collected from respondents. In sum, the theory and
studies cited here have demonstrated the value of combining qualitative and
quantitative research strategies. What seemed like irreconcilable differences
are resolved when the complementary nature of the two research approaches
is affirmed.

Arnheim, R. (1971). Art and visual perception: A psychology of the creative
eye. Berkeley, CA: University of California Press.
Avital, T. (2003). Art versus nonart: Art out of mind. New York: Cambridge
University Press.
Beach, W. A. (1990). Orienting to the phenomenon. In J. Anderson (Ed.),
Communication Yearbook, 13, 216–244.
Berlyne, D. E. (1949). 'Interest' as a psychological concept, British Journal
of Psychology, 39, 184–195.
Berlyne, D. E. (1971). Aesthetics and psychobiology. New York: Appleton-
Berlyne, D. E. (1974). The new experimental aesthetics: Steps toward an ob-
jective psychology of aesthetic appreciation. Washington: Hemisphere.
64 / Aesthetics and Innovation
Cupchik, G. C. (1988). The legacy of Daniel E. Berlyne. Empirical Studies of
the Arts, 6, 171–186.
Cupchik, G. C. (2001, February). Constructivist realism: An ontology that en-
compasses positivist and constructivist approaches to the social sciences.
Forum Qualitative Sozialforschung / Forum: Qualitative Social Research
[On-line Journal], 2. Available at:
http://qualitative-research.net/fqs/fqs-eng.htm [Date of access: 2, 12,
Cupchik, G. C., & Gignac, A. (in press). Finding meaning and expressing
emotion in response to artworks. Visual Arts Research.
Cupchik, G. C., & Hilscher, M. C. (in press) Phenomenology and the two
faces of design experience: Ease of use and expression. In H. N. J. Schif-
ferstein & P. P. M. Hekkert (Eds.), Product experience, Elsevier.
Cupchik, G. C., & Shereck, L. (1998). Generating and eceiving contextual-
ized interpretations of sculptures. Empirical Studies of the Arts, 16, 179–
Dewey, J. (1969). Art as experience. In F. A. Tillman & S. M. Cahn (Eds.),
Philosophy of art and aesthetics (pp. 252–269). New York: Harper &
Row. (Original work published, 1934).
Hilscher, M. C., & Cupchik, G. C. (2003). From the page to the stage: Poetry
writing, performance, and reception. Spiel, 22, 187–207.
László, J., & Fülöp, E. (2007). The role of attachment patterns in emotional
processing of literary narratives. In L. Dorfman, C. Martindale, & V. Pet-
rov (Eds.), Aesthetics and innovation (pp. 258–272). Cambridge: Cam-
bridge Scholars Press.
Lincoln, Y. S., & Guba, E. G. (2000). Paradigmatic controversies, contradic-
tions, and emerging confluences. In N. K. Denzin & Y. S. Lincoln (Eds.),
Handbook of qualitative research, 2nd ed. (pp. 163–188). London: Sage.
Medved, M., Cupchik, G. C., & Oatley, K. (2004). Interpretive memories of
artworks. Memory, 12, 119–128.
Smith, J. K., & Heshusius, L. (1986). Closing down the conversation: The
end of the quantitative-qualitative debate among educational inquirers.
Educational Researcher, 15, 4–12.

When the Real van Gogh is Real! Cognitive

Top-Down Effects in Art Appreciation

Helmut Leder and M. Dorothee Augustin

We discuss aesthetics and appreciation of modern art from an information-

processing viewpoint, starting from a model of aesthetic appreciation and
aesthetic judgments proposed by Leder, Belke, Oeberst, and Augustin (2004).
In particular, we focus on processes that are concerned with expertise and
top-down effects of knowledge. A study that deals with the impact of inform-
ation about the authenticity of paintings on the relation between familiarity
and liking illustrates such effects. This study rounds out the results of Leder
(2001) by showing that even for material equal in the initial level of liking
but varying in general familiarity, effects of familiarity still rely on the per-
ceivers’ belief about the authenticity of the artwork.

Art nowadays is omnipresent in many people‘s lives. Not only do news-

papers and TV formats regularly report on current exhibitions, but the exhibi-
tions themselves, such as the Melancholia in Paris or the Van Gogh &
Gauguin in Amsterdam, also attract large numbers of visitors. While block-
buster exhibitions represent the popular, more consuming side of art, the sci-
ences are particularly challenged by its more ambivalent and less obvious as-
pects. From a view of art history, examples of relevant questions are the bor-
ders of what can be considered art, as well as the role of art in a globalised
world. Psychology, on the other hand, has repeatedly addressed questions
concerning the perceptual and cognitive processes involved in art processing
(Reber, Winkielman, & Schwarz, 1998; Solso, 1994) or differences in person-
ality related to aesthetic appreciation (Furnham & Walker, 2001). Psychology
therefore aims to understand what psychologically states and affects the per-
ception of complex visual objects such as artworks, and what features and
processes determine the quality and intensity of aesthetic experiences.

66 / Aesthetics and Innovation
Recently, partly due to progress in many areas of the cognitive and neuro-
sciences, there has been an increase in studies that aim to investigate the
nature of human aesthetic experiences. While some studies make use of new
methodological possibilities, e.g. by using electroencephalograms (Jacobsen
& Höfel, 2003) or magnet-resonance measures (Jacobsen, Schubotz, Höfel,
& von Cramon, 2006; Vartanian & Goel, 2004), other approaches aim to de-
scribe art-specific cognitive and affective processes more comprehensively.
The model of aesthetic appreciation and aesthetic judgments by Leder, Belke,
Oeberst, and Augustin (2004) represents the latter approach and will be
sketched in the next sections.
As Ramachandran and Hirstein (1999) pointed out, aesthetic behaviour is
not just any domain of human behaviour – it seems to be a typical or even ex-
clusive domain of humanity. They stated that “our propensity to create and
enjoy painting and sculpture...[may be]... among the most puzzling” aspects
of human nature (p. 16). On the other hand, the reason why people produce
or appreciate art and the function of art are not apparent at first sight. This is
especially true for the realm of modern and contemporary art. Still, the per-
manence of art production and consumption in all known human cultures has
led psychologists to assume that art bears unique qualities, being not only
pleasurable, but probably also serving a certain function for the organism.
Historically, empirical aesthetics is among the most traditional fields in
experimental psychology. The founders of academic empirical psychology,
Fechner and Wundt, were both interested in perception and sensation as well
as aesthetic responses.
Fechner (1876) started his empirical investigations on aesthetics in close
relation to his psychophysics paradigms. For example, Fechner’s experiments
concerning the beauty of proportions such as the golden section revealed a
preference for rectangles that corresponded to this relation (see Höge, 1995,
for a discussion of recent views on the golden Section). In his seminal work
Vorschule der Ästhetik (1876), Fechner distinguished two forms of aesthetics,
one from below and one from above. The former is concerned with stimulus
properties, such as proportion, colour and other stimulus- inherent, often per-
ceptual, features. The latter refers to top-down influences of knowledge, edu-
cation and taste. These two forms of aesthetic variables have often been con-
sidered since then, but the majority of empirical psychological research has
been concerned with aesthetics from below.
At least three important movements of empirical aesthetics in the 20th
century should be mentioned briefly. The Gestalt-psychologists analyzed
Gestalt laws in patterns and artworks (Arnheim, 1954) and proposed prin-
ciples of aesthetic experience such as entropy, structure, order, and complex-
Top-down effects in art appreciation / 67
ity (Arnheim, 1971). A psychobiological approach was proposed by Berlyne
(1974), who saw aesthetics as serving fundamental needs of arousal and ex-
citement, that are closely related to a drive for exploration and to curiosity.
Later, Berlyne‘s approach was criticized because of its exclusive focus on ac-
tivation in terms of arousal, which, according to Martindale (1984), cannot
account for aesthetic experiences, as it neither considers semantic aspects of
artworks nor distinguishes aesthetic from non-aesthetic experiences.
Moreover, on the way to more cognitive approaches, Kreitler and Kreitler
(1972) proposed a psychodynamic approach including elements of psycho-
analysis and elements of cognitive psychology. What is essential in the Kreit-
lers’ approach is cognitive orientation, which refers to the beholder’s know-
ledge and attitudes. It is supposed to be closely related to a need for success-
ful understanding or extraction of meaning, which, in turn, are assumed to
lead to a pleasurable experience of homeostasis. Consequently, the approach
explicitly states that aesthetic appreciation cannot be understood without con-
sideration of the perceiver’s knowledge and other inter-individual variables
such as self-concept. Moreover, Kreitler and Kreitler developed a theory of
meaning in psychology – an approach that any theory of higher-order effects
on art appreciation has to consider. Their empirical work (Kreitler & Kreitler,
1984) tested hypotheses concerning the time course of extraction of meaning.
They found that the longer a stimulus was perceived, the more elements of
meaning might be activated.
In the last decade, there have been an increasing number of studies in
cognitive psychology that tested perceptual and cognitive features with re-
spect to aesthetic experiences. However, the field lacked a comprehensive
theory that considered variants of aesthetic processing at different processing
stages and integrated knowledge about different variables that have been
shown to affect aesthetic experiences. The model of aesthetic appreciation
and aesthetic judgments by Leder et al. (2004) attempted to fill this gap. It
aims to provide a framework for future research and is mainly concerned
with aesthetic appreciation of the visual arts. The model also regards top-
down knowledge, expertise, and means of understanding as essential for aes-
thetic experiences. The last section of this article empirically illustrates an ex-
ample of such top-down influences, describing a study concerning the rela-
tion between familiarity and liking.

A Cognitive Model of Aesthetic Experience

Our approach intends to explain what levels of processing are involved in
art appreciation. It seems that aesthetic objects challenge our mind in a spe-
68 / Aesthetics and Innovation
cific way. Our model endeavors to discover what these challenges are, and
how the cognitive system deals with them. Our view of the function of art
states that the information processing that underlies aesthetic experiences
somehow produces positive, self-rewarding experiences (Leder et al., 2004).
This may motivate the perceiver to search further experiences of art. One ex-
planation for this motivational function may be that art serves the purpose of
gaining knowledge about the world (Zeki, 1999). However, in our view, cog-
nitive and affective experiences during an aesthetic episode might be the es-
sential psychological part of gaining knowledge or gathering specific experi-
ences through art.
The model is based on an analysis of modern art from a psychological
perspective. According to that analysis, art is characterised by an amazing
variety of styles, which often are linked to single artists. Due to market mech-
anisms there is a strong tendency to produce innovative and unseen styles,
and, at least at the beginning of the 21th, century it seems that art perman-
ently stretches its borders and is no longer limited to a special medium.
Moreover, due to its experimental nature, 20th century art has developed a
larger need for interpretation than any art before. Concerning the psycholo-
gical understanding of aesthetic experiences, we assume that the nature and
intensity of an aesthetic emotion is determined by the subjective state of un-
derstanding of an artwork. “This is no trivial matter as the understanding is
no longer achieved with a visual representation of the ‘what is depicted’.
Conceptual ideas, abstract concepts, and reflection no longer become appar-
ent from the appearance of the artwork” (Leder & Belke, in press). As such
features are more and more essential in contemporary art, top-down influ-
ences are of growing importance for aesthetic experiences, since often only
specific declarative knowledge enables the viewer to derive meaning through
a search in concepts (Zeki, 1999).
In order to understand what cognitive and affective processes might be in-
volved when a person perceives an artwork, the model identifies a number of
processing stages (see Leder et al., 2004; or the description in Leder and
Belke, 2006). Input is an artwork, and outputs are twofold. One output of aes-
thetic processing is an aesthetic judgment concerning the artwork. The other
output is an aesthetic emotion, which is a by-product of the cognitive pro-
cessing levels. Such emotions comprise feelings of pleasure, arousal, and sat-
isfaction and can be independent of the aesthetic judgment.
What are the different levels of information processing described by the
model? We will illustrate the processing on each level by using an example of
an artwork by the sculptor Stefan Sous, who was born near Aachen in Ger-
many in 1964 and studied art at the Academy in Düsseldorf. Stefan Sous is
Top-down effects in art appreciation / 69
famous for sculptures which, like the one depicted in Figure 1, decompose
complex everyday objects to “explosion drawings” often found in technical
guidelines. Other works attempt to enrich public spaces. For example, he pro-
duced a giant scanner that seemed to scan the silhouette of the new National
Gallery in Berlin after the restoration in 2001.

Figure 5-1. Artwork by Stefan Sous: “Flunki Kiel” (1998).

The first processing stage proposed by the model of aesthetic appreciation

and aesthetic judgments consists of Perceptual Analyses. At this stage the art-
work is analysed visually by means of procedures that presumably are not
art-specific but are involved in all perceptual analyses. Contrast, colour,
shape, texture, symmetry, etc. are the variables relevant at this early pro-
cessing stage. Looking at the example in Figure 1, simply the colour and the
basic shapes of the constituting elements might be represented at Perceptual
There follows a stage of Implicit Memory Integration. Several approaches
from cognitive psychology assume that even implicit (not always verbaliz-
able) classifications affect the aesthetic value of an object (Ramachandran &
Hirstein, 1999; Zeki, 1999). Prototypicality and familiarity are among the
70 / Aesthetics and Innovation
variables relevant at this stage. Concerning the sculpture depicted in Figure 1
there is clear similarity to a well-known object, although apparently the way
the object looks is distorted. It has often been assumed that we like what we
know. However, despite the fact that mere repetitions increase liking for
many objects, the empirical findings with artworks are rather ambiguous
(Bornstein, 1989). This is not surprising if we consider the next levels of pro-
cessing which are usually involved in aesthetic experiences with art.
At the stage of Explicit Classification, perceivers are able to name what
they see in an artwork. It is presumably style and content that are most relev-
ant at this stage. We assume that with increasing art-specific knowledge style-
related classifications become more important. Thus, being very involved in
German contemporary art, the first thing that comes to your mind concerning
the work in Figure 1 might be that it is a work by Stefan Sous. In contrast, a
beholder with less domain-specific knowledge might first notice the similar-
ity to a ship. All in all, the more we know about art, the stronger we tend to
see how artworks are made rather than just what they depict.
On the next processing stage (called Cognitive Mastering) the perceiver
aims to understand and interpret the results of the previous processing stages.
Expertise in art probably has a very important role on this processing level as
well, because the criteria referred to in interpretation are formed by the view-
er’s experience. In the following stage of Evaluation the results of the cognit-
ive and the affective processes are measured. Unsatisfying, ambiguous evalu-
ations may trigger further processing of the artwork. Importantly, ambiguity
in art does not need to be fully resolved, as a certain level of ambiguity might
be experienced as positively arousing. However, the ability to experience
successful understanding is generally essential with respect to rewarding ex-
periences of pleasure. Presumably, a piece of art should be more easily classi-
fied or interpreted with increasing art-knowledge and experience. As stated
above, this is especially true for the realm of modern and contemporary art.
Let us again have a look at Figure 1 to sum up the model’s essential as-
sumptions concerning the role of knowledge in aesthetic experiences: At the
level of Explicit Classification a naïve viewer might represent the boat, while
an experienced viewer might see the space-extending sculpture, or even the
artwork by Stefan Sous. On the level of Cognitive Mastering the unexperi-
enced viewer might rely on his interest in boats or fishing, while the expert
might see a relation to cubism etc. Interestingly, the artwork in Figure 1
leaves particular space for interpretation with respect to theories of object re-
cognition. Biederman (1987) published an influential theory on how we re-
cognise objects by seeing the constituting basic elements, called geons, and
their spatial arrangements. In the exploding ship, the geons are still preserved
Top-down effects in art appreciation / 71
but the edges at which they are usually combined are made explicit through
the systematic explosion of the sculpture. However, these examples illustrate
that knowledge affects the way we see artworks, and implicitly we also have
shown that additional information might be helpful to see objects—here a
trivial boat—as artworks.
The model proposes a complex interplay of perceptual and top-down ef-
fects. The next paragraph deals with the question how such latter effects of
interpretation, knowledge and expertise can be shown empirically.

Top-down effects in art processing

As stated above, top-down cognitive processes affect most processing
levels. Depending on processing stage, different variables are involved, and
especially the stages of Cognitive Mastering and Evaluation are strongly in-
fluenced by the amount and nature of the viewer’s knowledge. To simulate
effects of knowledge in the laboratory, researchers have examined the effects
of different kinds of additional information on aesthetic processing and aes-
thetic judgments (e.g., Cupchik, Shereck, & Spiegel, 1994). One field of in-
terest in this respect has been the effect of titles, which undoubtedly stand in
a particular relation to the artworks themselves.
For example, Franklin, Becklen, and Doyle (1993) studied how viewers
responded to a painting under changing titling conditions. Their participants
saw each of two paintings twice—on one occasion with the original title, on
another occasion with a fabricated one. In a second session, for one of the
paintings an alternate title was presented. A change of title shifted the de-
scription of the artwork towards the meaning of the title, although the looking
pattern measured by registering eye movements did not change. Thus, while
the visual processing was rather unaffected by the title, the semantic pro-
cessing changed. More recently, Millis (2001) examined effects of different
titling conditions (descriptive versus elaborated). Participants rated illustra-
tions and photographs for understanding and four different qualities of the
aesthetic experience (liking, interest, elicited thoughts, and emotions). De-
scriptive and elaborative titles increased the rated understanding of both ma-
terials. Moreover, at least for illustrations, elaborative titles, which provided
an explanation or a metaphoric interpretation of the scene, increased the aes-
thetic experience more than descriptive titles. In order to understand the ef-
fects of elaboration on different dependent variables more specifically, Leder,
Carbon, and Ripsas (2006) investigated the effects of elaboration separately
for the four variables of aesthetic experience, using reproductions of art-
works. Moreover, similarly to Russel (2003), they also investigated repres-
72 / Aesthetics and Innovation
entational and abstract artworks. When both sorts of artworks were perceived
under long presentation times (90 s) in Experiment 1, then elaborative titles
increased the understanding of abstract paintings but not their appreciation.
In Experiment 2 only abstract paintings were presented, but in order to test
predictions concerning the time course of understanding and aesthetic appre-
ciation (Leder et al., 2004), two presentation times were used. For short
presentation times (1 s), descriptive titles increased the understanding more
than elaborative titles, whereas for medium presentation times (10 s), elabor-
ative titles increased the understanding more than descriptive titles. The res-
ults reveal that a presentation time of around 10 s might be needed to assign a
meaning that goes beyond mere description of the artwork.
Apart from titles, researchers have examined effects of different kinds of
additional information. With a view to the central role of style-related pro-
cessing assumed by the model of aesthetic appreciation and aesthetic judg-
ments (Leder et al., 2004), Leder, Belke, and Augustin (in press) tried to sys-
tematically induce such processing by supplying style-related information.
This information explained how the paintings had been made, for ex-
ample, that Richter had used a giant scraper to apply paint onto the canvas.
There was an increase in liking for styles with information as compared to
styles for which no such information had been given, but this effect could
only be shown for persons who were either in a relatively positive mood or
had a relatively low amount of art expertise. This is an example of how top
down influences interact with person variables and affective state, as de-
scribed in the model by Leder et al. (2004).
Concerning top-down processes, there is another factor that has to be con-
sidered as a variable of influence: social discourse. In this respect, the phe-
nomenon of so-called prestige effects has been of some interest in the literat-
ure on aesthetics. Prestige effects refer to effects of information conveying
knowledge about the artistic or social value of an artwork. Special attention
in this respect has been paid to the painter’s name as a source of information.
The exact nature of a name’s influence—whether it simply biases a judgment
or changes processing itself—has been subject to discussion. Farnsworth and
Misumi (1931) stated that artists’ names have some potency concerning pref-
erences for pictures. On the other hand, Chapman and Williams (1976) repor-
ted that ascribing music to different sources did not only change aesthetic
judgments but also descriptions of the pieces. The latter effect can be ex-
plained in light of Asch’s famous critique concerning prestige-effects in the
realm of political statements (Asch, 1948). In his view, one has to take into
account that a name has the potency to trigger a high amount of additional in-
formation that may in turn substantially change the context of a work and
Top-down effects in art appreciation / 73
thus its interpretation. In terms of the model of aesthetic experience, both
simple biases on judgments and cognitive reorganization (Asch, 1948) have
to be taken into account, and which one is more important depends upon vari-
ables such as expertise and processing stage.
All in all, we assume that different kinds of additional information change
aesthetic processing. In terms of the model of aesthetic appreciation and aes-
thetic judgments (Leder et al., 2004) they may also change relations that are
already relevant at relatively early stages of processing, for example, the rela-
tion between familiarity and liking. This relation has been shown with a num-
ber of materials (Bornstein, 1989). Concerning effects of mere (unreinforced)
repeated exposure, Kunst-Wilson and Zajonc presented evidence for en-
hancement of affective evaluations even under conditions of subliminal
presentation (1980). For the realm of artworks, the results concerning mere
exposure effects are mixed (Bornstein, 1989). Yet, Cutting (2003) presented
evidence of effects of “natural” mere exposure, in terms of significant correl-
ations between ratings of liking for paintings and their frequency of appear-
ance in library books and catalogues.
A series of experiments by Leder (2001) tried to experimentally examine
the impact of top-down effects of knowledge on the familiarity–liking rela-
tionship. In order to test whether top-down information affects the influence
of familiarity as a result of previous experience, Leder (2001, 2002) investig-
ated the relation between familiarity and liking for real artworks. He chose
van Gogh paintings for two reasons. First, van Gogh is one of the most fam-
ous painters, which makes it very likely that most participants were familiar
with at least some of his paintings. Second, there have been a number of re-
ports that not all of the works ascribed to van Gogh were painted by him.
Some experts have suggested that a significant number of forgeries hang in
museums and art collections (for example, the Hulkker Cataloge names 45
paintings in different museums which are likely to be forgeries). Because the
aim of the research was to investigate the impact of prior knowledge on the
evaluation of artworks, the ability to introduce van Gogh’s paintings as ori-
ginals or fakes made his paintings ideal for the desired experimental manipu-
lations. Concerning the considerations presented above, these manipulations
referred to knowledge bearing a connotation of social value, i.e. knowledge
concerning the authenticity of the artworks.
In a series of experiment participants first gave spontaneous ratings of
how much they liked a series of 56 van Gogh paintings. Afterwards, the per-
ceivers were asked to rate each painting again, now in respect of how familiar
they were with each of the paintings. When participants were told that all art-
works were “paintings by the famous dutch painter van Gogh” both ratings
74 / Aesthetics and Innovation
revealed a high positive correlation. The same was true when the instruction
stated that the stimuli consisted of paintings by van Gogh, among which there
might be forgeries. In contrast, when the participants were told that all stimuli
consisted of forgeries of paintings by van Gogh or of forgeries of van Gogh
paintings and works by other, unknown artists, the correlation between fa-
miliarity and liking was significantly smaller. Thus, it seems that authenticity
is an important mediator of judgments of famous pieces of art. Yet, as the ex-
periments by Leder (2001) showed, it only affected the relationship between
familiarity and liking, but did not reduce liking ratings as such. This lends
some support to the notion that one has to see effects of information related
to prestige in a more differentiated light, not simply as biases in judgment. In
contrast, one possible explanation for the effects found could be that serious
doubts concerning the authenticity of the paintings generally lead to in-
creased cognitive activity. This, in turn, might have reduced the likeliness of
relying on simple heuristics – such as familiarity. This is in line with the find-
ing that increasing inspection time for the paintings significantly reduced the
relation between familiarity and liking, too (Leder, 2001).
However, the studies reported consisted of quasi-experimental settings in
which there was large inter-individual variation of familiarity with the van
Gogh paintings under investigation. Most important, it may generally be pos-
sible that higher liking for more familiar artworks is mediated by the quality
of artworks—if one assumes that works of higher quality are chosen more of-
ten for exhibitions and publication in art books.
A complementary study (reported in Leder, 2002) tried experimentally to
control for differences in quality. On the basis of the data from the first two
of the above-described studies (Leder, 2001) artworks were chosen that had
either very low or very high ratings of familiarity but similar levels of liking.
If the relationship between familiarity and liking was still present for this set
of stimuli, this would be interpreted as evidence against quality as a mediat-
ing variable.
Twenty four students (5 males, 19 females, mean age 22.6) participated to
fulfill course requirements. Twenty-four reproductions of paintings were se-
lected from the original set of 56 on the basis of the data of 12 participants
(see Leder, 2001): twelve paintings which had received relatively high famili-
arity ratings (M = 4.93, on a seven point-scale) and twelve paintings with rel-
atively low familiarity ratings (M = 2.33). Both sets were almost equal in
mean level of liking (5.11 and 5.25, respectively). Instruction (original or
fake) was used as a between-subjects variable. In the original condition parti-
cipants were told that they would be exposed to paintings of the famous
Dutch painter van Gogh. In contrast, participants in the fake condition were
Top-down effects in art appreciation / 75
informed that new studies concerning the art of the famous painter van Gogh
had identified a number of forgeries, and that the current study aimed to in-
vestigate these paintings. They were instructed that during the following ex-
periment they would see reproductions of paintings, among which were both
van Gogh forgeries and paintings by other, unknown painters. Participants
were given either instruction according to a randomized list.
The experiment consisted of two blocks. In the first block participants
were asked to rate each image with respect to how much they liked it, and in
the second block the same paintings were rated according to how familiar the
participants were with each of the stimuli. Both ratings were given on a nine-
point scale. Since previous studies had revealed that familiarity-affect studies
are more reliable in spontaneous evaluations (Krugalski, Freund, & Bar,
1996; Leder, 2001), a short presentation time of only 200 ms was used.
Twenty-four experimental trials were preceded by four practice trials.
An analysis of variance with liking ratings as dependent variable, instruc-
tion (original versus fake) as between-subjects factor and painting set (famili-
ar versus unfamiliar) as within-subjects factor revealed a significant effect of
painting set, F(1,22) = 28.428, p < .001. The paintings in the familiar group
were liked significantly better than the paintings in the unfamiliar group. In
contrast, there was no significant main effect of instruction. Thus, the fake-
instruction did not generally lower ratings of liking.
To find out whether the instruction had affected the relationship between
familiarity and liking, the correlations between both variables (averaged over
painting sets) were Fisher-Z transformed and compared between both condi-
tions via t-test. Correlations in the fake instruction (M =.208) were signific-
antly lower than under the original instruction (M =.393), t = -2.04, p < .05.
Moreover, an analysis of variance with correlations as dependent variable, in-
struction as between-subjects factor and painting set (familiar versus unfamil-
iar) as within-subjects factor revealed a slight trend towards a main effect of
instruction, F(1,22) = 3.29, p < .10, but no other effects. Thus, there were no
significant effects related to the level of familiarity of the paintings.
All in all, the experiment lends further support to the notion that informa-
tion about the authenticity of paintings may affect the familiarity-liking rela-
tionship while leaving the general level of liking mostly unaffected. Concern-
ing the model of aesthetic experience by Leder et al. (2004) this suggests that
even mechanism that are believed to be relevant during relatively early stages
of processing can be mediated by top-down information. As to the question
of effects of prestige, the results suggest that information concerning the
value of a painting may work in a way that is much more complex than
simple biases of judgment.
76 / Aesthetics and Innovation
Conclusion and Outlook
Empirical aesthetics from the perspective of cognitive psychology have
come to a point where new approaches try to integrate knowledge about aes-
thetic processing at different levels in order to understand more closely which
processes and variables are involved in aesthetic appreciation. In this respect,
knowledge, expertise and experience are increasingly discussed as mediating
variables, leading to new approaches in the line of Fechner’s aesthetics from
above. From an empirical point of view, a number of recent studies have var-
ied the amount and quality of information and measured how such conditions
change aesthetic judgments and the subjective feeling of understanding. Such
studies directly refer to the “need for interpretation” which seems to be par-
ticularly important in modern art. Importantly, top-down information may
also affect the role of variables whose impact on aesthetic processing is as-
sumed to happen very fast or automatically. For example, information con-
cerning the authenticity of paintings reduces the relation between familiarity
and liking.
In the future the investigation of the complex interplay between aspects of
artworks and top-down processes linked to the knowledge of the perceiver
will probably lead to more insights concerning the sources of the pleasure we
derive from encounters with art.

Author Notes
The writing of this chapter was supported by a Grant (SFB 626 C5) of the
Deutsche Forschungsgemeinschaft (DFG) to Helmut Leder and by a Grant
from the Fonds zur Förderung der wissenschaftlichen Forschung (FWF, P
18910) to Helmut Leder and Claus-Christian Carbon. The data used here
have been part of the first author’s habilitation thesis at the Freie Universität
Berlin. We thank Leonid Dorfman for making this chapter possible.
Moreover, we thank Stefan Sous for the permission to use a photograph of
one of his artworks.

Arnheim, R. (1954). Art and visual perception. Berkeley: University of Cali-
fornia Press.
Arnheim, R. (1971). Entropy and art. An essay on disorder and order. Berke-
ley: University of California Press.
Top-down effects in art appreciation / 77
Asch, S. E. (1948). The doctrine of suggestion, prestige and imitation in so-
cial psychology. Psychological Review, 55, 250–276.
Belke, B., Leder, H., & Augustin, M. D. (in press). Mastering style—Effects
of explicit style-related information, art knowledge and affective state on
appreciation of abstract paintings. Psychology Science.
Berlyne, D. E. (1974). Studies in the new experimental aesthetics. New York:
John Wiley and Sons.
Biederman, I. (1987) Recognition—by—Components: A Theory of Human
Image Understanding. Psychological Review, 94, 115–147.
Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of
research, 1968-1987. Psychological Bulletin, 106, 265–289.
Chapman, A. J., & Williams, A. R. (1976). Prestige effects and aesthetic ex-
periences: Adolescents’ reactions to music. British Journal of Social and
Clinical Psychology, 15, 61–72.
Cupchik, G. C., Shereck, L., & Spiegel, S. (1994). The effects of textual in-
formation on artistic communication. Visual Arts Research, 20, 62–78.
Cutting, J. E. (2003). Gustave Caillebotte, French Impressionism, and mere
exposure. Psychonomic Bulletin and Review, 10, 319–342.
Farnsworth, P. R., & Misumi, I. (1931). Further data on suggestion in pic-
tures. American Journal of Psychology, 43, 632.
Fechner, G. T. (1876). Vorschule der Ästhetik. Leipzig: Breitkopf & Härtel.
Franklin, M. B. (1988). “Museum of the Mind”: An inquiry into the titling of
artworks. Metaphor & Symbolic Activity, 3(3), 157–174.
Franklin, M. B., Becklen, R. C., & Doyle, C. L. (1993). The influence of
titles on how paintings are seen. Leonardo, 26, 103–108.
Furnham, A., & Walker, J. (2001). Personality and judgements of abstract,
pop art and representational paintings. European Journal of Personality,
15, 57–72.
Höge, H. (1995). Fechners experimental aesthetics and the golden section
hypothesis today. Empirical Studies of the Arts, 13, 131–148.
Jacobsen, T., & Höfel, L. (2003). Descriptive and evaluative judgment pro-
cesses: Behavioral and electrophysiological indices of processing sym-
metry and aesthetics. Cognitive, Affective & Behavioral Neuroscience, 3,
Jacobsen, T., Schubotz, R. I., Höfel, L., & von Cramon, D. Y. (2006). Brain
correlates of aesthetic judgment of beauty. NeuroImage, 29, 276–285.
Kreitler, H., & Kreitler, S. (1972). Psychology of the arts. Durham: Duke
University Press.
Kreitler, H., & Kreitler, S. (1984). Meaning assigment in perception. In W. D.
Fröhlich, G. J. W. Smith, J. G. Dragung, & U. Henschel (Eds.), Psycholo-
78 / Aesthetics and Innovation
gical processes in cognition and personality (pp.173–190). New York:
Kruglanski, A. W., Freund, T., & Bar, T. D. (1996). Motivational effects in
the mere-exposure paradigm. European Journal of Social Psychology, 26,
Kunst-Wilson, W. R., & Zajonc, R. B., (1980). Affective discrimination of
stimuli that cannot be recognized. Science, 207, 557–558.
Leder, H. (2001). Determinants of preference. When do we like what we
know? Empirical Studies of the Arts, 19, 201–211.
Leder, H. (2002). Explorationen in der Bildästhetik. Lengerich: Pabst Pub-
Leder, H., & Belke, B. (in press). Art and cognition: Cognitive approaches to
art appreciation. In C. Martindale, P. Locher, and V. Petrov (Eds.), Evolu-
tionary and neurocognitive approaches to aesthetics, creativity and the
arts. Amityville NY: Baywood.
Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthet-
ic appreciation and aesthetic judgments. British Journal of Psychology,
95, 489–508.
Leder, H., Carbon, C. C., & Ripsas, A. (2006). Entitling art: Influence of title
information on understanding and appreciation of paintings. Acta Psycho-
logica, 121, 176–198.
Martindale, C. (1984). The pleasures of thought: A theory of cognitive hedon-
ics. The Journal of Mind and Behavior, 5, 49–80.
Millis, K. (2001). Making meaning brings pleasure: The influence of titles
on aesthetic experiences. Emotion, 1, 320–329.
Ramachandran, V. S., & Hirstein, W. (1999). The science of art. Journal of
Consciousness Studies, 6, 15–51.
Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual flu-
ency on affective judgements. Psychological Science, 9, 5–48.
Russell, P. A. (2003). Effort after meaning and the hedonic value of paint-
ings. British Journal of Psychology, 94, 99–110.
Solso, R. L. (1994). Cognition and the visual arts. Cambridge: MIT Press.
Vartanian, O., & Goel, V. (2004). Emotion pathways in the brain mediate aes-
thetic preference. Bulletin of Psychology and the Arts, 5, 37–42.
Zeki, S. (1999). Inner vision. Oxford: Oxford University Press.


Quantitative Estimations of Creativity: Social

Determination and Free Will

Lidia A. Mazhul and Vladimir M. Petrov

The winds, as at their hour of birth,

Leaning upon the ridged sea,
Breathed low around the rolling earth
With mellow preludes, We are free.
The streams through many a lilied row
Down-carolling to the crisped sea,
Low-tinkled with a bell-like flow
Atween the blossoms, We are free.

Alfred, Lord Tennyson (1830)

To measure the strength of social impacts on individual creative processes,

parameters of works of art (poetry, painting, music, and architecture) were
used, and six kinds of statistical estimations were made relating to different
aspects of creativity. Both the personal probability of choosing an artistic ca-
reer (e.g., to become a poet) and the probability of belonging to a definite
fashionable stylistic direction were calculated. The degree of free will of
artists increased when ascending the ladder of creative levels from absolute
(eternal) structural regularities—to relative (conventional) ones. For painting
the appropriate range of freedom changes from 25% to 64%.
Individual impacts of outstanding persons on social creative processes were
estimated on the basis of evolutionary regularities inherent in different kinds
of art. The upper limits of these impacts concerning the possibility of shifting
regular periodical processes is in the range from 13 to 22 years.

Quantitative estimations of free will / 81
It is a trivial thesis that the creative process is usually influenced by vari-
ous factors, including social ones. There are various approaches to estimate
this influence (e.g., Eysenck, 1995; Simonton, 1994). Among these ap-
proaches, those should be singled out which are derived in the framework of
the information paradigm (e.g., Golitsyn & Petrov, 1995, 2005; Maslov,
1983). We used some of the results obtained in the framework of this ap-
proach, to measure the degree of the free will of artists (painters, composers,
architects, poets, etc.) against the pressure of social and cultural factors.
When speaking of any social determination, two kinds of measures are
a) purely quantitative estimates of the probability that a person will
choose a certain career, That is, attempt to become an eminent painter, com-
poser, poet, etc.;
b) qualitative estimations, meaning primarily the content and style of
works of art which are formed under the influence of concrete social condi-
We shall start our consideration from the quantitative aspect, (a).

Quantitative Aspect: Are Artistic Careers determined by Fate

and Birth?
The most best path for estimations of this kind should be based on com-
parison of probabilities of becoming an eminent artist relating to different
time periods of social development. The raw material for such a comparison
is the evolutionary curves describing changes in the intensity of artistic life.
Examples of such curves can be found In the results of an investigation of
the evolution of literary life in Russia during the 18 th–20th centuries (Petrov &
Majoul, 2002). Quite similar curves were found for creativity in music and
painting both in Russia and West Europe during the 15th–20th centuries (Ku-
lichkin, 2004).
To study the intensity of Russian literary life, data concerning 307 poets
and 480 prose writers were extracted: years of birth together with lengths of
descriptions of their creativity (the number of lines) in an encyclopaedia. At a
preliminary stage, several literary encyclopaedias were compared with each
other to prove the statistical identity of their hierarchies in relation to the em-
inence of the authors represented, which is reflected by the length of descrip-
tions of the authors. Then all the authors were grouped in accordance with
their birthdates, into five-year intervals: 1730—1734, 1735—1739, … 1940
—1944, and the total number of lines devoted to the authors of each interval
82 / Aesthetics and Innovation
was calculated. This value was considered to be the principal indicator of the
intensity of literary life (separately for prose and verse). Appropriate evolu-
tionary curves for Russian poetry, the number of poets (n) born in each five-
year interval, and the number of lines (N) devoted to them, are presented in
Fig. 1.

Figure 6-1. Intensity of literary life in Russia, 18th—20th centuries: poetry. The number
of eminent poets (n) born in different 5-year intervals, and the number of lines (N) de-
voted to these poets, in dependence of the years of their birth (t).

Each evolutionary curve consists of two components: the long-term trend,

not monotonic, but having a hill-like form, (this trend being an artifact caused
by the procedure of measurement); and short-term sharp changes against the
background of this long-term trend.
These short-term sharp changes which take place against the long-term
background were the main focus of our analysis. They are caused by period-
ical switches between left- and right-hemispheric dominance in the entire so-
Quantitative estimations of free will / 83
cio-psychological sphere (Maslov, 1983; Petrov, 2001, 2003, 2004a, 2004b).
Each type of dominance lasts about 20—25 years. Because adjacent extreme
points of each evolutionary curve (its minima and maxima) are rather close to
each other in their time ranges, it seems reasonable to compare values of in-
tensity between these points.
The overfalls of intensity when considering adjacent extreme points,
change in a rather wide range: from 2.1 to 35 times for poetry and from 1.9 to
4.2 times for prose with average values about 13 times and 3 times, respect-
ively. This means that, first, the fate of prose writers is 4 times less dependent
upon time than the fate of poets. Second, it is possible to estimate the tempor-
al impact on the fate of creative persons. That is, the role of the year of their
Let us suppose that when intensity is maximal, there exist the most favor-
able conditions for authors to make a literary career, so the appropriate per-
sonal probability equals unity. Hence, for those authors who were born when
intensity was minimal, the probability of making a poetic career is usually 13
times less, whereas for all potential poets (born in an arbitrary year) this
probability is 13/2 = 6.5 times less. In other words, for an average potential
poet his/her ideal probability of becoming eminent equals 1/6.5 = .15. As a
result, only a rather small share of poetic fame (not more than 15%) can de-
pend on a person’s individual efforts. In reality this share is probably much
lower. As for the career of prose writers, the appropriate ideal share is not so
small—about 1/1.5 = .67, i.e. 67%. In reality, of course, it is much lower.

Qualitative Static Aspects: Subduing to Structural Regularities

Now let us turn to the second kind of estimations (b) concerning qualitat-
ive aspects of creativity. Here several methods of investigation were ex-
amined and proved fruitful. They can be ascribed to two aspects: static and
The static aspect deals with time-independent regularities, i.e. almost
eternal ones. Deviations from such regularities may be treated as evidence of
the free will of creative persons.
Some examples of eternal regularities concern color structures used in
painting. Here three regularities were deduced theoretically (Petrov & Grib-
kov, 2000) and proved empirically (Gribkov & Petrov, 1996).
A. The most strict regularity seems to be absolute, because it deals with
the very nature of the given kind of art. It is nothing else than the number of
spectral colors constituting the color structure of each painting. According to
the general theory, each painting is recommended to use from three to five
84 / Aesthetics and Innovation
spectral elements out of the seven colors the of which the spectrum consists.
Otherwise the perception of the picture would be far from optimal. To test
this theoretical conclusion, a statistical investigation was made which in-
volved expert estimations of 822 easel paintings created by 167 European
artists (French, Italian, Spanish, and Russian) of the 15th—20th centuries. Ex-
perts were asked to list all color elements which are present in each given
painting, as well as to indicate the three main color elements dominating the
perception of each painting. The frequency of usage of different colors in
Russian and French painting is shown by Fig. 2.
The distribution of the French and Russian paintings over the number of
spectral colors used is shown in Fig. 3. Color structures, exactly responding
to the above theoretical recommendations, were observed in 71% of French
paintings, 82% of Italian paintings, 82% of Spanish paintings, and 75% of
Russian paintings (Mazhul & Petrov, 2002). The background probability (i.e.
the chance to meet such structures) is 3/7 = .43. Fig. 2 presents the distribu-
tion of Russian and French paintings over the number of spectral colors used.
It is clear that the freedom of French artists when choosing the number of
spectral elements, is 100% – 71% = 29%, for Italian, Spanish, and Russian
artists it is 18%, 18%, and 25% respectively; the average freedom over these
four national schools is 25%.
B. Another regularity is not so strict but is half-absolute. It deals both with
the nature of the given kind of art and concrete sociocultural conditions of its
functioning. The requirement concerns the so-called color-and-light standard.
That is, a certain element (fragment of the depiction) which acts as a tuning
fork for perception of all other elements presented in the painting. The choice
of this standard depends upon geography. That is, the properties of the sun-
light dominating each country. [For instance, Russian painting should be
based on a white color-and-light standard, whereas French and Italian paint-
ing should be based upon a yellow standard. The second column of Table 1
presents results obtained in the investigation.
One can see that in each national school the appropriate color-and-light
standard element is met rather frequently: in 97% of French paintings, 96%
of Italian paintings, 88% of Spanish paintings, and 77% of Russian paintings.
However, much more significant results are seen when analyzing the use of
the appropriate color element as the main one. Such usage of elements con-
sistent with national color-and-light standards is also shown in Table 1 in the
third column. Typical national color-and-light standards form the core of 55%
of French paintings, 63% of Italian paintings, 55% of Spanish paintings, and
44% of Russian paintings. The background usage of the typical color can be
Quantitative estimations of free will / 85
calculated by averaging the frequencies of this color as one of the main col-
ors in non-appropriate countries. These background values are 50% for

Figure 6-2. Frequency of usage of different main colors: percentage (P) of paintings
with each given color used in the role of main; Russian painting (a)
and French paintings (b).

France and Italy, and 27% for Spain and Russia.] The freedom to violate this
requirement is 45%, 37%, 45%, and 56% for French, Italian, Spanish, and
Russian artists, respectively; the average freedom over these four national
schools is 48%.
C. An almost purely relative or conventionally cultural regularity deals
with national color triads which are inherent in most paintings of each nation-
86 / Aesthetics and Innovation
al school (serving as signs of national identity). For instance, the French and
Italian schools are based on a triad of yellow, orange, and dark blue; whereas
Spanish painting uses white, red, and black; and Russian painting uses white,
red, and green. Empirical data concerning usage of these triads in European
schools of painting, are shown in the right column of Table 1. The predicted
triads were observed in 42%, 25%, 52%, and 29% of French, Italian, Spanish,
and Russian paintings, respectively. Background probabilities (i.e., frequen-
cies of the color triads, if they were combined randomly) are 28%, 25%,
18%, and 27% for France, Italy, Spain, and Russia, respectively. Thus, the
data for triads in French and Spanish painting are statistically significant at a
level better than 95%. The freedom not to use this device, is 58%, 25%, 48%,
and 71%, respectively, with the average value about 64%.
So, creativity in painting illustrates q hierarchy of degrees of freedom cor-
related with the nature of the requirements imposed on the creative process:
the most strict are absolute requirements, followed by half-absolute and relat-
ive ones, with average values of freedom being about 25%, 48%, and 64%,
respectively (see also Mazhul & Petrov, 2002).

Qualitative Dynamic Aspect: Subduing to Fashion

The dynamic aspect of free will deals with temporal changeability of art:
how often are artists subdued to the fashion which dominates a given field?
The simplest example of temporal regularities relates to usage of definite
fashionable devices of art. For instance, in an investigation devoted to the
evolution of Japanese prints (ukiyo-e) of the 17th—19th centuries (Koptsik,
Ryzhov, & Petrov, 2004; Melamid & Petrov, 1998), the device of composi-
tional symmetry was studied. In accordance with the theoretical model, usage
of this device showed periodic behavior, with the full duration of cycles be-
ing about 35 years. The evolutionary curve is presented in Fig. 4, showing the
coefficient of asymmetry: the share of asymmetrical compositions in the en-
tire group of prints created during each four-year time interval. The curve
shows evidence in favor of the changeability of fashion.
However, most interesting is the flatness of the evolutionary curve ob-
tained. The coefficient of linear correlation between the values for odd years
and even years is about .60. It means that fashion is rather influential; the
square of this coefficient is the coefficient of determination reflecting the por-
tion of prints created in agreement with the dominating fashion. Hence, about
36% of all prints are subdued to this fashion, and the degree of painters
Quantitative estimations of free will / 87

Figure 6-3. Pictorial diversity of color structures: share (S) of paintings vs. the number
(ns) of spectral elements used; Russian painting (triangles) and French paintings

freedom is about 64%. This device, as well as all other devices mentioned in
the investigation, have frequencies of occurrence which are in agreement
with theoretical recommendations concerning optimal structures of works of
art (Petrov, 2002b).
Finally, an analogous example concerns stylistic features of music. These
features were studied in the light of left- and right-hemispheric prevalence in
creativity (e.g., Koptsik, Ryzhov, & Petrov, 2004; Lotman & Nikolayenko,
88 / Aesthetics and Innovation
1983; Maslov, 1983; Petrov, 2001, 2003, 2004a, 2004b). Each of seven mu-
sical parameters was presented in the form of a binary opposition (e.g., Op-
timism—Pessimism, Strict form—Free form), the first pole of which corres-
ponds to left-hemispheric dominance in the work of the composer studied),

Table 6-1. National schools of painting: color-and-light standards and color triads

Country (and num- Number of paintings with nation- Number of

ber of paintings al color-and-light standards paintings with
studied) national triads
as usual as main
elements elements

France (311) 301 170 131

Italy (109) 105 69 27
Spain (106) 88 58 55
Russia (296) 223 130 87

and the second pole to right-hemispheric dominance. The creativity of 102

composers (both West-European and Russian) of the 17 th—20th centuries was
described by means of a special measurement procedure. For each composer,
the indicator of his/her inclination toward the first (left) pole was calculated
from the seven scales. Then the mean value of each indicator was calculated
for each time range of the evolution by averaging the values over all com-
posers who had created their compositions during this time range. As well,
the scattering of the values of each indicator, i.e. its standard deviation, was
calculated. These values are presented in Table 2.
The standard deviation, being divided into the mean value, characterizes
the degree of composers’ freedom (Petrov & Danilova, 1992). For instance,
the rigidity of musical creativity (in accordance with the opposition Strict
form-Free form) at the beginning of the 18th century was estimated by a
mean value of about .80, with a standard deviation of about .15. Hence, the
creative freedom of composers was at that time about .15/.80 = .19. However,
Table 6-2. Average parameters of musical creativity and (in brackets) their standard deviations
Quantitative estimations of free will / 89
90 / Aesthetics and Innovation
at the beginning of the 20th century, the mean value of this indicator was .43,
with a standard deviation about .28, so creative freedom is .28/.43 = .65. In
other words, freedom increased from 19% to 65%, i.e., more than 3 times.
Quite analogous growth is inherent in all other features of musical creativity.
Exactly such a growth in different national literatures was observed earlier by
Likhachyov (1969). So here we see rather large freedom of creative activity.
In total, we computed six kinds of statistical estimations of the free will of
artists, and the results obtained were in good agreement with each other. Cre-
ative activity becomes more and more free, especially when ascending the
ladder of creative levels. This phenomenon is in good agreement with general

Figure 6-4. Evolution of the index of asymmetry (K) of Japanese prints ukiyo-e,

laws of the vertical growth in the sphere of culture (Golitsyn, 2000, Golitsyn
& Petrov, 2005; Petrov, 2002b). The higher levels of human activity are
really creative; here the personality may reveal his/her genuine potentialities
and become free.
Quantitative estimations of free will / 91
Individual Impact in Social Processes: Creators or Slaves?
Now it seems reasonable to consider the problem which is inverse to the
above analysis: what is the individual impact of creative persons on social
processes? Can a person substantially influence the reality in which he/she is
obliged to live and create his/her works? Or is a person only a slave serving
appropriate social needs and sometimes satisfying his/her own needs by
means of deviations from social regularities?
One of the ways to analyze this problem, is to resort to the help of definite
general and strong regularities of social and cultural life: changes in these
regularities can be treated as results of the efforts of some persons acting of
their own free and active will. Here the most interesting would be to estimate
not the results of the efforts themselves, but the upper limits of such possible
Undoubtedly, the most suitable kind of such strong regularities is presen-
ted by periodic changes which have been observed in various cultural phe-
nomena. There exist different models for such periodical processes. One of
the most influential models (Martindale, 1990) deals with the social as well
as individual need to increase the arousal potential carried by cultural objects
(e.g., works of art): to increase this impact value, it is necessary periodically
to change some stylistic features of the objects (e.g., works of art). Another
model is connected with the phenomenon of left- or right-hemispherical dom-
inance; here the need to switch from one pole to another is caused by the ne-
cessity to renew the current paradigm of information processing, to refresh it
(Maslov, 1983; Petrov, 1992, 2001, 2003, 2004a). However, irrespective of
the theoretical model used, such periodic processes and deviations from them
are exactly the raw material needed for our quantitative estimations of pos-
sible individual impacts on social and cultural processes.
For instance, the studies of the evolution of musical style as well as styles
of other kinds of art were based on the opposition of left- and right-hemi-
spherical creative features. A set of indicators was derived for each sphere of
creativity (e.g., the seven parameters of musical style), permitting us to meas-
ure the inclination of creativity in each sphere either to left- or right-hemi-
spheric dominance. Moreover, on the basis of these parameters, the index of
creative asymmetry was calculated for each time segment of the evolution of
each sphere of art. In this index, the data are aggregated over all the indicat-
ors derived for this sphere. This index can vary from –1 (pure right-hemi-
spheric dominance) to +1 (absolute left-hemispheric prevalence). A fragment
of the evolution of this index for music (both West European and Russian) is
presented in Fig. 5.
92 / Aesthetics and Innovation

Figure 6-5. Stylistic evolution of music (West-European and Russian): periodic changes of the index of asymmetry (K) against
the long-term monotonic trend. The long-term trend is designated by the dotted line. Some epochs of left- or right-hemispheric
dominance are shown by shading.
Quantitative estimations of free will / 93
One can easily see that the evolutionary behavior of the index of creative
asymmetry, in full agreement with theoretical predictions, reveals cyclic rises
and falls, against the background of a monotonic decreasing trend (i.e., con-
stantly growing inclination toward right-hemispheric dominance; this trend is
shown by a dotted line). The full period of such cycles is about 50 years. In
other words, after about a 20—25-year epoch of left-hemispheric dominance
(though this dominance is relative, meaning a positive difference between the
actual value of the index and its trend value), a switch to the opposite, right-
hemispheric dominance always occurs (and again this prevalence is relative,
in comparison with the long-term trend), and this epoch also lasts about 20—
25 years; then the next switch to left-hemispheric dominance comes, and so
on. Some such epochs are designated in Fig. 5 by shading (different for left-
and right-prevalence, designated by characters L and R, respectively) on the
axis of time.
Table 3 presents some examples of such epochs observed in different
spheres: socio-political climate, style of architecture, music, and painting.
Our main interest will be focused on the duration of dominance of both left-
and right-hemispheric waves. We shall try to use these data to evaluate the
upper limits of personal contributions to the process of evolution.
Let us consider an arbitrary half-wave, i.e., either a left or right epoch in a
certain sphere (let it be the socio-political climate, or architecture, or music).
To change the duration of such a half-wave, one can shift either its left border
or its right border. But arrival of the half-wave before its expected time
means nothing else than the corresponding shortening of the previous half-
wave of the opposite type. Similarly, the delay of the previous half-wave
means the corresponding shortening of the subsequent one. That is why it
seems reasonable to consider only one of these shifts as independent.
We prefer to analyze the shift of the right border, and there are at least
two reasons for such a preference. First, this shift is caused by the exhaustion
of the potentialities of the existing half-wave (see analogous motives in Mar-
tindale, 1990), and exactly this feeling may become extremely sharp due to
the activity of a certain creative person or a group of persons. Second, from
the point of view of the content of changes, we have no grounds to lean on
another effect. That is, on a shift of the left border due to the arrival of innov-
ators. The matter is that when a change has to come, there exist many candid-
ates for future leadership, and their efforts are directed rather chaotically. As
a result, the efforts of a certain person or a group of persons are lost in such a
sea. That is why even the arrival of innovators should be considered mainly
in light of their perception of the exhaustion of the previous stylistic
paradigm. Also, prolongation of the old paradigm is possible only if someone
94 / Aesthetics and Innovation
finds within it some new opportunities. But this is nothing other than a shift
of the right border.

Table 6-3. Examples of left- and right-hemispheric waves in different fields of activity

Field of activity Left-hemispheric waves Right-hemispheric waves

Dates Mean Dates Mean

duration duration
(and SD) (and SD)

Socio-political cli- 1762-1794, 24.5 (7.4) 1728-1761, 26.0 (11.4)

mate 1802-1826, 1795-1801,
(Russia) 1856-1880, 1827-1855,
1916-1930 1881-1915,
Architecture 1761-1796, 22.8 (16.5) 1728-1760 27.4 (15.4)
(Russia) 1799-1834, 1797-1798,
1870-1871, 1835-1869,
1914-1930. 1872-1913,
Music 1690-1728, 25.2 (12.4) 1729-1753, 27.0 (15.0)
(Russia and 1754-1789, 1790-1819,
Western Europe) 1820-1834, 1835-1850,
1851-1861, 1862-1912,
1913-1937. 1938-1950.
Painting 1436-1452, 22.0 (5.8) 1453-1481, 26.0 (14.3)
(Russia and 1482-1501, 1502-1523,
Western Europe) 1524-1542, 1543-1579,
1580-1600, 1601-1612,
1613-1634, 1635-1666,
1667-1695, 1696-1718,
1719-1736, 1737-1748,
1749-1765, 1766-1782,
1783-1818, 1819-1835,
1836-1857. 1858-1916.
Quantitative estimations of free will / 95
So, how big can be the shift of the right border (i.e., either shortening or
prolongation of a definite half-wave) in a definite sphere? Let us consider this
problem in application to the concrete example of left-hemispheric epochs in
a specific sphere as the socio-political climate. Concrete indicators used for
the estimations in this sphere may be found in Koptsik, Ryzhov, and Petrov
(2004), Maslov (1983), and Petrov (1992, 2004b).
According to Table 3, the mean duration of half-waves is about 24.5
years, with a standard deviation of 7.4 years. But this deviation is a result of
fluctuations both of left and right borders of many left-hemispheric epochs.
We may suppose that these two classes of deviations are statistically inde-
pendent of each other, because the starts and the finishes of these epochs are
determined by different persons. It may rarely occur that a person could be at
the top for such a long time that he/she determines both borders of a half-
wave. Hence, the standard deviation for one of the borders should be √2
times less than the full standard deviation That is, it should be about 7.4 / √2
= 5.2 years. What does it mean as far as the possible impact of separate per-
sons is concerned?
As it is well known, in normal or Gaussian distribution less than 5% of
the data deviate from the mean value more than on 2 standard deviations. So,
in 95% of the cases a possible shift (either to the left side or to the right side)
of the beginning of any left-hemispheric epoch cannot exceed 2 × 5.2 = 10.4
years. In other words, none of the outstanding politicians (meaning presid-
ents, kings, general secretaries, etc.) is capable of shortening or prolonging
the duration of the previous epoch more than for a decade. It is the upper lim-
it of possible personal or group impacts on the objective regularity, i.e., the
dynamics of the social sphere. Nevertheless, this upper limit seems to be
rather high; that is really important from the moral standpoint.
As for right-hemispheric epochs in the socio-political climate, here the
standard deviation is about 11.4 years (see Table 3). Hence, the personal im-
pact in the shift of the arrival of such epochs is about 2 × 11.4 / √2 = 16.1
Quite similarly, in the case of painting, the impacts have their upper limits
of about 2 ×5.8 / √2 = 8.2 years, and 2 × 14.3 / √2 = 20.2 years, for left- and
right-hemispheric stylistic waves, respectively. Thus, in principle, a genius
painter or a group of painters is capable either of shortening the duration of a
certain right-hemispheric style, or prolonging its life even for two decades.
Nevertheless, a voluntary shift of such a size possesses a rather low probabil-
ity, and in reality the shifts are far less. In general, the upper limits for such
shifts in painting, averaged over left- and right-hemispheric epochs, cannot
exceed (8.2 + 20.2) = 14.5 years. For other spheres of creative activity rather
96 / Aesthetics and Innovation
close estimations can be obtained using data in Table 3. Average sizes of such
shifts change in the range from 13 to 23 years, with a mean value of 17 years.
Further, we can also estimate the conditions under which those persons
should function whose efforts do really create each epoch. As soon as volun-
tary shifts, as a rule, cannot exceed half of the duration of each epoch, here
more than a half of this duration is absolutely independent of any personal ef-
forts and even any group ones, but is subdued to quite definite social circum-
stances. But most persons are usually functioning during all the epoch or
more and not only during its 50%. Hence, more than 50% duration of the
activity of each person is subdued to the inexorable course of history. That is,
it is independent of the free will of the personality. So he/she can be both a
creator of the epoch and its slave.

The results deal mainly with certain statistical subjects. But the conclu-
sions seem to be rather important, and first of all from the point of view of
the age-long philosophical contraposition between the necessity and the free-
dom of creativity and human behavior in general.

Eysenck, H. (1995). Genius: The natural history of creativity. Cambridge:
Cambridge University Press.
Golitsyn, G. A. (2000). High art and low art: On the systemic role of an elite
subculture. Journal of Russian and East European Psychology, 38, 28–44.
Golitsyn, G. A., & Petrov, V. M. (1995). Information and creation: Integrat-
ing the two cultures. Basel: Birkhauser Verlag.
Golitsyn, G. A., & Petrov, V. M. (2005). Social and cultural dynamics: Long-
range tendencies (Information approach). Moscow: KomKniga. (In Rus-
Gribkov, V. S., & Petrov, V. M. (1996). Color elements in national schools of
painting: A statistical investigation. Empirical Studies of the Arts, 14,
Koptsik, V. A., Ryzhov, V. P., & Petrov, V. M. (2004). Etydy po teorii
iskusstva: dialog yestestvennyh i gumanitarnyh nauk (Etudes on the the-
ory of art: Dialogue between sciences and humanities). Moscow: OGI
Publishers. (In Russian).
Quantitative estimations of free will / 97
Kulichkin, P. A. (2004). Evolution of artistic life and style of thinking (A trial
of a quantitative investigation). Diss. Doct. Moscow: State Institute for
Art Studies. (In Russian).
Likhachyov, D. S. (1969). Literature of the future as an object of investiga-
tion: Notes and considerations. Novy Mir, 9, 167–184. (In Russian).
Lotman, Yu. M., & Nikolayenko, N. N. (1983). Golden section and problems
of inter-brain dialogue. Dekorativnoye Iskusstvo SSSR, 9, 31–34. (In Rus-
Martindale, C. (1990). The clockwork muse: The predictability of artistic
change. New York: Basic Books.
Maslov, S. Yu. (1983). Asymmetry of cognitive mechanisms and its con-
sequences. Semiotika i Informatika, 20, 3–34.
Mazhul, L. A., & Petrov, V. M. (2002). Free will of painters: Choice of color
structures. In: R. Tomassoni (Ed.), La psicologia delle arti oggi (pp. 67–
76). Milano: Franco Angeli.
Melamid, L. A., & Petrov, V. M. (1998). Immanent periodicity from the in-
formational standpoint: Symmetry in Japanese art revisited. In G. M.
Balim, V. M. Petrov, & V. P. Ryzhov (Eds.), Information approach in em-
pirical aesthetics (pp. 251–259). Taganrog: Taganrog State University of
Radio Engineering.
Petrov, V. M. (1991). Emotional impact of art: Social aspect. In L. Ya. Dorf-
man, D. A. Leontiev, V. M. Petrov, & V. A. Sozinov (Eds.), Art and emo-
tions: Proceedings of the International Symposium (pp. 41–50). Perm:
Perm State Institute of Culture.
Petrov, V. M. (1992). Evolution of art and brain asymmetry: A review of em-
pirical investigations. In G. C. Cupchik & J. Laszlo (Eds.), Emerging vis-
ions of the aesthetic process: Psychology, semiology, and philosophy (pp.
255–268). NY: Cambridge University Press.
Petrov, V. M. (2001). Creativity in art: Stylistic waves and monotonic evolu-
tionary trends (Information approach). Bulletin of Psychology and the
Arts, 2, 30–33.
Petrov, V. M. (2002a). Devices of art: Optimal frequency of occurrence (In-
formation approach). In R. Tomassoni (Ed.), La psicologia delle arti oggi
(pp. 43–48). Milano: Franco Angeli.
Petrov, V. M. (2002b). High against low: Hierarchies in nature, culture, and
art (Information approach). Bulletin of Psychology and the Arts, 3, 3–9.
Petrov, V. M. (2003). Cyclic cultural evolution against the background of
long-range progressive trends: Information approach. Journal of Cultural
and Evolutionary Psychology, 1, 85–107.
98 / Aesthetics and Innovation
Petrov, V. M. (2004a). Quantitative methods in art studies. Handbook for stu-
dents. Moscow: Academic Project. (In Russian).
Petrov, V. M. (2004b). Wave-like social processes: Methods of forecasting.
Sotsiologiya: Metodologiya, Metody, Matematicheskiye Modeli, 18, 130–
153. (In Russian).
Petrov, V. M., & Danilova, O. N. (1992). Ideology–art–pluralism (Positivistic
notes). In A. Karyaguin (Ed.), Art and ideology: Contemporary artistic
process as an ideological problem. (pp. 226–246). Moscow: Russian In-
stitute for Art Studies. (In Russian).
Petrov, V. M., & Gribkov, V. S. (2000). National characteristics in painting:
The psychological basis of color structures. Journal of Russian and East
European Psychology, 38, 64–86.
Petrov, V. M., & Mazhul, L. A. (2002). Pulsation of literary life: Periodical
behavior of Russian poetry and prose in light of the information approach.
Rivista di Psicologia dell Arte, 13. 25–40.
Simonton, D. K. (1994). Greatness: Who makes history and why. New York:

Neuropsychological support to the Novelty

Generation Process

T. Sophie Schweizer

External evaluations, either positive or

negative, do not foster psychological
health, but rather prevent us from being
completely open to our own experience.

(Rogers, 1959)

How does support from others affect us neuropsychologically while we are

engaged in creative work? How can we support ourselves by influencing the
balance of our key neurotransmitter systems to achieve the right state for per-
forming creative work? Why can some individuals experience creative drive
under severe deadline pressure, while others experience a creative block?
Why do some loose their neuropsychological balance when becoming fam-
ous while others do not? How can we explain the phenomenon that some
famous individuals in art history have perceived different types of drugs as
supportive of their work while others have not? Why are there people who
keep on jumping to new projects without being able to finish previous ones?
The above questions refer to neurocognitive and neuropharmacological
mechanisms of social support and self-support that have hardly been paid at-
tention to in the creativity literature.
This article seeks to create a research agenda for a neuropsychology of social
support and self-support to different stages of the novelty generation process
drawing mainly from neuropsychological and social psychological research.
The neural responses of individuals to different forms of support, so it is ar-
gued here, depend on their different neuropsychological profiles and the dif-
ferent stages of the novelty generation process they are in at the moment of

100 / Aesthetics and Innovation
receiving a particular form of support. Here, the paper builds on the different
phases of the novelty generation process as conceptualized in the Novelty
Generation Model.
In describing potential neural responses to received support this article fo-
cuses on three key neurotransmitter systems dominating specific work modes
that can occur in the novelty generation process: first, the highly focused,
more agitated, stressful and reward-seeking work mode (dopamine-domin-
ated); second, the goal-directed and more mellow work mode (serotonin-
dominated); and third, work under anxiety and pressure (GABA and norad-
renergically dominated). Various kinds of support are classified in this paper
that are argued here to distinctly affect these key neurotransmitter systems in
the different stages of the novelty generation process. Also, these neural
mechanisms are also described to vary in their relevance for two different
modes of the creative process, namely on the one hand deliberate and on the
other hand spontaneous creative problem solving.

Any form of social support that creative professionals can receive, like for
instance receiving a grant, embodies evaluations of our work and a judgment
about our support-worthiness. How do evaluations, judgments, and resulting
support affect us in our novelty generation processes? An abundance of re-
search has been published on the subsidization in the arts and sciences.
However, the above questions concerning neuropsychological effects of such
support actions on the novelty generation process have not been researched,
let alone effects of other support actions such as from the very important in-
terpersonal realm or self-support.
Which are the neuropsychological parameters of motivated and poten-
tially novelty-generating work, which its stimulants and which its depress-
ants? How do their effects vary for different types of individuals and for the
different stages of the novelty generation process? If we look at the creativity
literature, we can certainly find ideal work conditions for being creative (Am-
abile, 1996). A question that has not been asked is the following: are there
different neuropsychological profiles reflecting different types of creative
workers and how do they relate with different work conditions representing
different forms of support?
This paper is a first attempt to format some of those neuropsychological
correlates of support settings by focusing on some key neurotransmitters
characterizing three typical work modes: first, the highly focused and more
agitated, stressful, and reward-seeking work mode (dopamine-dominated);
second, the goal-directed and more mellow work mode (serotonin-domin-
ated); and third, work under anxiety and pressure (GABA and noradrenergic-
Neuropsychological Support to the Novelty Generation Process / 101
ally dominated). Some people may be set to work well mainly with the help
of serotonin shots through positive forms of support. Others may need negat-
ive forms of support, and may enter a good work mode via dopamine shots or
noradrenergic pathways like under the threat of pending deadlines. In order to
be able to perform creative work, some may even artificially create such situ-
ations, like postponing work until an ultimate point before a deadline when
the pressure becomes maximal.
The following question is then: which neurotransmitter systems do differ-
ent types of individuals seek to stimulate or depress (consciously or uncon-
sciously) in the course of the novelty generation process? For instance, which
forms of drug abuse are they inclined to fall prey to? We may distinguish,
first, those who have underfed dopaminergic systems. They have a higher
likelihood to abuse drugs that stimulate these dopaminergic systems such as
nicotine, caffeine, or cocaine; second, there are the type of creative workers
who need to be high on serotonin to work well (Schweizer et al., in prepara-
tion). They may be inclined to abuse cannabis to either tone themselves down
to a more mellow work mode or tone themselves up to a less depressed, more
motivated work mode, depending on their neuropsychological profile. Third,
individuals may have an inclination to alcohol to reduce the anxiety that may
block them in their novelty generation process. Being able to recognize and
distinguish anxious work from goal-directed, mellow work and the more agit-
ated, stressful, and reward-oriented work can not only be crucial for a better
self-understanding of the creative workers themselves, but also for those who
seek to support the novelty-generating processes of others from a supervisory
How the effects of different forms of support can be explained from a
neuropsychological perspective and how they match with individual neuro-
cognitive profiles are the issues for which this paper seeks to set a research
agenda. The first step is to identify the basic neuropsychological parameters
of an individual and the particular phase of the novelty generation process in
which the individual operates at that moment. As a next step the ideal forms
of support can be identified. The theoretical framework used as a basis of this
support framework is the Novelty Generation Model (Schweizer, 2004, 2006)
which is based on an analysis of psychological theories, most importantly
drawing from neuropsychology and social psychology. This model clearly
distinguishes three components: novelty-seeking (dopamine-regulated, ex-
ploratory behavior), and creativity and innovative performance. This model
forms an ideal basis for identifying neuropsychological mechanisms that in-
terfere with or facilitate each of these components.
102 / Aesthetics and Innovation
Different forms of social support are distinguished, which may be ob-
tained from interpersonal and impersonal sources. On the other hand there is
the self-support of the individual (linked to genetic, neurocognitive, and per-
sonality patterns). Support can take the form of instrumental support (such as
material/financial aid, or drug interventions like nicotine or caffeine con-
sumption) and validational support (social esteem, affirmation). Different
modes of support transmission can be distinguished ranging from positive
support (e.g., by praise) to negative support (e.g., by way of stress through a
pending deadlines). How such support phenomena affect the neuropsycholo-
gical balance of the creative individual has not yet been paid attention to in
the neuropsychological literature on novelty generation processes. This art-
icle develops a research agenda with the aim of establishing a neuropsycho-
logy of the novelty generation process that not only has utmost theoretical,
but also practical relevance.

Basic Support Mechanisms: Self-Support and Social Support to

the Novelty Generation Process
Specific neurotransmitter activity and resulting motivational states fed by
temperaments, personality, and genetic and social set-ups affect the three key
components of the novelty generation process as conceptualized in the Nov-
elty Generation Model (Schweizer, 2004, 2006) below (Fig. 1).
This model allows researchers and practitioners to identify more accur-
ately the switches in the novelty generation process where neuropsychologic-
al dysfunctions may occur. Novelty-seeking forms the first component in the
onset of the novelty generation process, followed by creativity as a second
component consisting of two main processes: novelty-finding and novelty-
production, which in turn is followed by innovative performance in which a
product is placed in a wider social environment. Within the Novelty Genera-
tion Model notion of creativity we can speak of novelty-finding when an in-
dividual is capable of finding novel problems and novel solutions. When, in
addition, an individual transforms novel findings into observable products we
can say that someone has been creative, which in turn is the necessary condi-
tion for novelty to enter the innovation process afterwards.
It is important to note that there are also two different neurocognitive
types of creative problem solving: first, the deliberate mode, and second, the
spontaneous mode, the first being related to an up-regulation and the second
to a down-regulation of the prefrontal cortex (Dietrich, 2004a). In the case of
spontaneous creative processing, creative thinking also involves intuitive
Neuropsychological Support to the Novelty Generation Process / 103
leaps, that are facilitated in the higher stages of self-aware consciousness
(Cloninger, 2004, p.329). Relaxation, low levels of cortical and frontal-lobe
activation, more right than left hemisphere activation were found to accom-
pany the state of creative inspiration (Martindale, 1999; Martindale & Hasen-
fus, 1978; Martindale & Hines, 1975). The possible neurotransmitter activity
corresponding to different states and phases of the novelty generation process
will be sketched in this article.

Figure 7-1. The Novelty Generation Model (NGM) (Schweizer, 2006)

Social support as a variable influencing novelty-seeking and creative be-

havior as well as innovative performance has long been neglected in social
psychological research. Even the research history of social support in general
is a relatively short one in social psychology (Stroebe & Stroebe, 1996). Only
in the last two decades has interest in social support increased. To study sup-
port to novelty-seeking, creativity and innovative performance, social support
has to be distinguished from self-support. Novelty-generating individuals can
find self-support in their novelty-seeking genetic, neurocognitive, and per-
sonality parameters and in the intrinsic motivation they experience. Social
support can be distinguished into two forms: interpersonal support and imper-
sonal support. Interpersonal support can occur in families or organizations,
104 / Aesthetics and Innovation
whereas public support institutions, governments and markets are sources of
impersonal support.
Any component of the Novelty Generation Model can be the target of
support: the process of novelty-seeking, the creative process or the innova-
tion process. Social stimulation and social recognition are probably the most
important mechanisms of social support to the novelty-generating individual.
Just like some specific drugs can function as pharmacological stimulants or
depressants of novelty-seeking and creative behavior, social stimulation or
social recognition, in its most extreme form called fame, may also be condu-
cive or aversive to the different components of the novelty generation pro-
cess. Social support in the form of social recognition is per definition inher-
ent in the innovation process. If novelty-generating professionals do not
make the transition from mere novelty-seeking behavior to novelty-finding
and production, they are unlikely to receive continued social support, neither
in the form of social recognition, nor stimulation for their activities. The in-
novative performance of individuals partly reflects their capability to secure
social support for themselves.
Social support can facilitate, but can also interfere with novelty-seeking,
creativity as well as innovative performance. Of interest in this context is
work from various research streams: the motivation literature, especially re-
search on the receipt of rewards and their effects on intrinsic creative interest
and performance (see for instance Eisenberger & Armeli, 1997); the innova-
tion and creativity literature, where the term ‘support’ for innovation is used
mainly synonymous to the notion of an appropriate climate for creativity and
innovation (Amabile, 1996); and literature on the arts and sciences including
biographical information on famous novelty-seekers, which provide rich ex-
amples of the psychological mechanisms typical of the novelty-seeking pro-
Social support refers to all forms of support given or received from the
social environment (see for instance Stroebe et al., 1996). The focus in the
social psychological literature has been on interpersonal forms of support and
the different ways in which different types of support can affect mental and
physical health (Stroebe & Stroebe, 1996) Four types of support, which have
widely been agreed to contribute to the health and well-being of the support
receiver (House, 1981) are: (1) emotional support, which would include the
satisfaction of esteem needs and needs of intimacy; (2) validational or ap-
praisal support, which would satisfy the need for affirmation, feedback, and
social comparison; (3) informational support, which would simply be advice
of any kind; (4) instrumental support, which would include any material or
financial aid mainly (Stroebe & Stroebe, 1996, p. 599). Emotional and in-
Neuropsychological Support to the Novelty Generation Process / 105
formational support forms are usually associated with interpersonal settings.
Instrumental support embodied in rewards or grants such as financial or other
material support create time and freedom to seek novelties and to pursue the
transformation of findings into innovative performance. The same thing may
happen via validational support: When conferred publicly in the form of an
honorary reward for instance, it may build up status for the receiver of this
support. The use of legal (nicotine, caffeine) or illegal drugs (cannabis, co-
caine) can be described as a form of instrumental self-support.
Individuals with different neurocognitive profiles may differ in their pref-
erences for different types of support, and also in their neuropsychological re-
sponses to them. For instance, to strive for social recognition or fame as an
external motivator reflects an individual’s achievement needs. This can be ar-
gued to relate to a need for validational support. Just as biological psycholo-
gists have identified differences in individuals’ biological reward systems
(Blum et al., 2000), we may argue that the social reward systems of individu-
als differ in a similar way.
Whether a specific type of intended support indeed turns out to be factual
support also depends on the receiver. After all, giving and receiving is a form
of social negotiation. Also, imposed versus self-controlled support have very
different psychological effects. The need for support has also been identified
as a key factor moderating the individual’s negative reaction to imposed
(even if well-intended) support (Deelstra et al., 2003). The supportiveness of
external impulses arguably depends strongly on the timing with which they
enter the novelty generation process. Looking at the stages conceptualized in
the Novelty Generation Model, at some stages extrinsic impulses may add the
necessary moment to an intrinsically motivated work, which may reach stag-
nation at some point in time. Pressures generated by an impersonal aggregate
of competitors may support the novelty-seeking process by increasing the
self-support of the individual, namely the intrinsic motivation to seek
something completely new. Later in the innovation process when the indi-
vidual seeks social recognition for her/his potential novelties, such external
pressures may also have a supportive effect. However, external pressures are
less likely to facilitate the creative conceptual stage of idea formation, where
task focus is the most conducive. It may be argued that a U-shaped relation-
ship exists between the maturity of the project and the desirability of social
support in addition to self-support. In the first phase, novelty-seekers will
need some degree of social comparison to see what is there, what others have
done, and how they stand in relation to them and then need a phase of clos-
ure. In the conceptual phase, the novelty-seeking temperament and personal-
ity as well as intrinsic motivation can provide the individual with crucial por-
106 / Aesthetics and Innovation
tions of self-support. What follows is an elaboration phase in which the per-
severance trait becomes key. Amabile and Collins for instance argue that in
the often tedious process of working out the details of a project, developing
fully and presenting the product to others extrinsic motivators may play a fa-
cilitating role as opposed to the early stages of creative thinking (Collins,
1999, p. 306). As soon as the elaboration of the novelty is finished the full
portion of social support is desirable. Here the presentation of novelty to a
public begins: social recognition for the novelty is the form of social support,
which is most desirable in this phase.
When it comes to self-support, the most valuable source for this support is
the novelty-seeker’s intrinsic motivation to seek the new. Individuals who
need to be propelled by others to generate novelties can be said to be less
self-supportive than individuals who have the intrinsic motivation to produce
something. High scores in TCI Novelty-Seeking combined with high Scores
in Self-Directedness (Cloninger et al., 1994) are key elements for self-sup-
port. An individual is intrinsically motivated if he or she ‘performs an activity
for no apparent reward except the activity itself. Extrinsic motivation, on the
other hand, refers to the performance of an activity because it leads to extern-
al rewards (e.g.. status, approval, or passing grades) (Deci, 1972, p.113). So-
cial support in its most adequate form can be argued to offer the potential
novelty-generator an extension to his/her own human capabilities, that is it
complements the individual’s self-support. In situations where the capability
of one human party has reached its limits or is not sufficient to master the
situation, social support is efficient. The self-support that individuals may
find in their genetic and personality characteristics as well as intrinsic motiv-
ation patterns, combine with the different types of support coming from the
social circles surrounding an individual. Together they influence the degree to
which individuals engage in novelty-seeking behavior and the degree of in-
novative performance they may achieve. In the next section the neural correl-
ates of the different forms of support described in this section will be added
to the picture.

Key Neurotransmitters and Support Constellations Affecting

Different Phases of the Novelty-Generation Process
Psychological antecedents and consequences of social support have been
investigated (see for instance Sarason et al., 1989). However, they have not
been investigated in a neuropsychological perspective. Self-supporters may
be distinguished from individuals who mainly operate on interpersonal sup-
Neuropsychological Support to the Novelty Generation Process / 107
port and again others who mainly draw on impersonal support. A balanced
support status is ideally one that draws from all circles. Support deficits in
one of the circles may lead to compensatory behaviors of the novelty-gener-
ator trying to draw excess support from other circles. The psychological com-
plications attached to such imbalances may affect the novelty-generator,
his/her social environment as well as the outcome of the novelty generation
process. A number of key neurotransmitter systems active in such support
constellations will be discussed in the following. What happens to the novelty
generation process when specific neurotransmitter levels are too high or low
and require regulation? How can the individuals themselves regulate them?
Clear answers are difficult to find at this stage, but abundant are the research
questions to be identified for this new field, which is the aim of this article:
setting a research agenda.

Support to the Dopaminergic Dimensions of the Novelty

Generation Process: About Agitation, Reward-seeking, Stress,
and Being Hyper
Dopamine is the neurotransmitter that has been argued to increase creat-
ive arousal in that it is key to highest goal-directedness and reward-seeking
activity (Flaherty, 2005). In the Novelty Generation Model dopamine is the
relevant neurotransmitter for the novelty-seeking component. When the need
for new stimulation and the urge to find something new is strong and the
craving for a reward is a dominant feature, dopamine is at work. It has indeed
been discovered that novelty-seeking behavior is related to individual differ-
ences in specific neurotransmitter activity in our synapses. The novelty-seek-
ing personality is said to be modulated by the transmission of the neurotrans-
mitter dopamine (Cloninger, 1994; Cloninger et al., 1993). Specific genes de-
termining its transmission (DRD4, DRD2-A2, SLC6A3-9) have been labeled
‘novelty-seeking genes’ (Benjamin et al., 2002; Benjamin et al., 1996; Eb-
stein et al., 1996; Hamer et al., 1999; Keltikangas-Järvinen et al., 2003; Ler-
man et al., 1999; Noble et al., 1998; Prolo & Licinio, 2002). An individual’s
novelty-seeking personality crucially depends on the individual’s dopamine
levels, which in turn is related to the chances of finding something new and
being creative (Schweizer, 2006).
Nicotine, Caffeine, and Cocaine as dopaminergic stimulants are often
used by creative professionals to support focused deliberate creative prob-
lem-solving. From a neuropharmacological perspective the physical need for
nicotine during the novelty generation process can be explained by its facilit-
108 / Aesthetics and Innovation
ative effect in deliberate and focused creative problem-solving (similar to
caffeine). In the context of impersonal validational support an interesting
condition is the extreme dose of in the accumulation of world fame. This
form of support has to be handled with caution, since it has been found to in-
terfere with intrinsic motivation and self-support to novelty generation. Why
is it that some novelty-generators who achieve world fame stay completely
normal and keep producing, while others become exalted and actually cannot
handle their fame and go mad? An explanation can be offered, which com-
bines neuropsychological and social insights: Novelty-seeking personality
has been postulated to be modulated by dopamine transmission (Cloninger,
1994; Cloninger et al., 1993) and the role of dopamine in the rewarding effect
of external stimuli such as drugs has also been demonstrated in pharmacolo-
gical experiments (Wise & Rompre, 1989). Also we have seen that individu-
als differ in their dependence on rewards. It has never been investigated
whether social stimuli like social recognition travel the same dopamine routes
and may substitute for stimuli derived from novelty-seeking. So, a very dif-
ferent way of looking at fame and its effects is to take into account the differ-
ent levels of dopamine-modulated novelty-seeking in individuals. Psychotic
states of delusions about their own grandeur related to excessive dopamine
levels have been witnessed in some composers, writers, and painters. Famous
examples are Tschaikowsky, Rachmaninoff, Hemingway, and Jackson Pol-
lock. Furthermore, high dopamine levels are related with low latent inhibi-
tion, which in turn has been found to be related to higher creative perform-
ance (Carson et al., 2003). Interestingly also, dopamine antagonists suppress
free association which is a key element of creative processes (Flaherty,
Social support clearly can also turn out to be non-supportive, counter-pro-
ductive or, very importantly, may turn into negative support and cause stress.
What is negative support to the novelty generation process? Factual support
can come in a negative mode via stereotypes, ignorance, resistance, excess
support, and even destruction. All this can motivate individuals with a partic-
ular personality set-up to do their very best and use the resistance as a form
of negative support. Resistance from the social environment means that this
environment pays attention and invests energies into the target it resists. Zero
support or indifference may be the most useless in supporting novelty-seek-
ing and innovative performance, except for strong egos, who may even per-
ceive mere indifference as a form of provocation and cause dopaminergic
agitation. Negative stereotypes like women had to face in scientific work can
turn out to interact with the refusal to offer support and can form a source of
negative support for the stereotyped: For instance stereotyping can generate
Neuropsychological Support to the Novelty Generation Process / 109
the energies for social change processes engaged in by those who are disad-
vantaged (Reynolds et al., 2000)
Instrumental support like for instance a grant has its greatest impact on
the novelty-seeking and creative process. The following responses to such re-
wards have been described in the psychology literature: first, rewards may in-
crease motivation, and also rewards may help novelty-seekers to learn about
their social position by the mechanism of social comparisons, the outcome of
which may encourage or discourage them (Festinger, 1954). Second, rewards
which already contain a higher degree of social recognition may interfere
with the individual’s self-supportive level via the complacency/overjustifica-
tion effect (Bem, 1972; Kelley, 1973). In the social psychological literature
the overjustification effect suggests a negative relationship between external
reward upon an individual’s activity and the individual’s intrinsic motivation
to engage in that activity. As soon as a reward has been received, the dopam-
ine level that is the motor behind reward seeking changes. Arguably the nov-
elty-seeking impulses are also reduced then. Creative individuals may start
asking themselves whether they are creating something for these external re-
wards or because they are really intrinsically motivated to do so. This ques-
tion may sometimes be followed by a need to create a situation in which the
answer is unambiguous, which either means blocking external rewards or
stopping the activity itself. These are constraints to creativity via constraints
to intrinsic motivation (Amabile, 1987; Deci, 1971, 1972; Lepper & Greene,
Evidence that this problem exists among creative people can be found in
(auto)biographies and many letters sent between individuals in the novelty-
seeking professions. I would like to return to the examples with which I
opened this chapter: the letter of the poet Anne Sexton written to her friend
W.D. Snodgrass shortly after he had won the Pulitzer Prize for poetry. There
she warns him not to lose his intrinsic motivation for writing his poetry: ‘To
hell with their prize and their fame. You have got to sit down now and write
more “real”...write me some blood. That is why you were great in the first
place. Don’t let prizes stop you from your original courage, the courage of an
alien.’ (Sexton & Ames, 1977, pp.109-110) Clearly, the Pulitzer Prize for po-
etry is an honorary reward. It comes with an immense amount of publicity
and social recognition. After receiving such a reward it can be argued that
some reward-seeking is satisfied and that the related dopaminergic activity
that drove this reward-seeking is decreased.
Cases in which receiving an honorary reward leads to a drastic reduction
in the work motivation of the winner may also bring the painful insight that
previous work motivation may in large parts have been based on achievement
110 / Aesthetics and Innovation
needs rather than the intrinsic motivation of seeking and creating something
novel. As soon as this achievement is there, the work motivation drops. In
that sense, rewards can also function as a painful mirror probing the receiver
to think about his or her motivations in the novelty generation process. The
neuropsychological correlates of such motivational patterns are a highly rel-
evant point for the research agenda set up in this chapter.
Governments or any other public institutions have the possibility to put
strong incentives in place for individuals such as researchers or artists to en-
gage in the novelty generation process. Impersonal support providers motiv-
ate their support for novelty-seeking individuals among others with the ex-
pectation of benefits for whole occupational fields and society at large. Such
institutions may be international, national, or regional. Science, technology,
and the arts are probably the most frequently investigated sectors with respect
to impersonal support forms like subsidies. Setting incentives can appeal to
the dopaminergic reward systems of the producing individual to go and try to
obtain a certain reward.
A final issue that indirectly affects the dopaminergic side of the novelty
generation process is the issue of stress. Here it is important to distinguish
between acute stress situations and chronic stress. Given the role of dopamine
in the novelty generation process described above it is important to under-
stand how glutamatergic mechanisms that characterize stress situations can
cause an exaggerated striatal dopamine response to stress (Moghaddam,
2002). This is the mechanism that has also been related to stress-induced
symptom exacerbation in schizophrenia (Benes, 1997). So here it will be in-
teresting to investigate how individuals with different dopaminergic levels re-
act to acute stress situations and up to which levels stress can be supportive to
the novelty generation. The dopamine-increasing effects of such acute stress
situations can be hypothesized to stand in clear contrast to the effects of
chronic stress which has been related with a decrease in nucleus accumbens
dopamine (Gambarana et al., 1999). It may be hypothesized then that chronic
stress depresses the novelty-seeking impulse and other dopamine-related
components of the novelty-generation process. The role of both mechanisms
deserves the utmost attention in neurocognitive and neuropsychological re-
search on the novelty generation process in professional environments.
Neuropsychological Support to the Novelty Generation Process / 111
Support to the Serotonergic Dimensions of the Novelty
Generation Process: The Goal-Directed, Mellow Work Mode
A lack of serotonin that is characteristic of depression has been found to
be related to creative block in that it causes frontal deficits in the brain such
as abulia (lack of impulse), decrease in motivation, cognitive flexibility, goal-
directed activities (Flaherty, 2005). Especially when it comes to the novelty-
producing phase within the novelty generation process, sufficient serotonin
levels may be argued to be very conducive. Novelty-producing is the phase
when the seeking and finding of a new idea is done and this new idea has to
be transformed into a product in an often tedious work process that requires
persistence and tolerance of frustration. In these phases higher dopamine
levels that are related with increased reward-seeking are rather disturbing,
since they are more likely to make the individual jump to other new projects
instead of finishing the older ones. So for these phases higher serotonin levels
make for the inner restfulness needed to get through the often less exciting
and rather operational processes of transforming ideas into a finished
product. Receiving interpersonal positive support can be a powerful sero-
tonergic stimulator that can enable a steady and satisfying work mode but as
will be discussed in a later section it can also turn into a negative mode. But
first its positive form: Such support may come from the private interpersonal
realm, such as the familial environment or (in)formal relationships. They may
act as support sources and often are the individual’s main sources of emotion-
al support, a dimension not to be underestimated in discussions of support
systems of novelty generation. Furthermore, interpersonal support can come
in the validational form, that is recognition and feedback to generate a feeling
of competence in the individual receiving it, and constructive information on
how to improve (Deci & Ryan, 1985). Also instrumental support can be con-
ferred in this realm. An in-depth Diagnosis of Human Relationships
(Cloninger, 2004, p.168) can also serve to get a clearer picture about the in-
terpersonal support conditions under which a novelty-generator operates.
Many novelty generators in the history of art and science have managed
to secure familial or professional positions for themselves in which they
were–-ften within silent consensus--the ones who consumed most of the at-
tention and absorbed most of the available support. From research with mon-
keys and humans we know that this kind of privileged positions are related to
increased serotonin levels. In practice, this may for instance, be instrumental
support. Sigmund Freud, or instance, is said to have been the only one
amongst all his siblings being granted a study room on his own (Gay, 1999).
This may also be validational support such as being praised and being the
112 / Aesthetics and Innovation
pride of the parents (Albert, 1983). Developmental psychology suggests that
privileged positions in childhood may prepare the individual for equivalent
roles in adult life: The early experience of tolerance towards support absorp-
tion within a family setting may equip the absorbers with a predisposition for
high self-confidence and an expectation of privileged positions also in later
interpersonal relationships. Also access to first-rate mentors is very likely to
be facilitated by previous interpersonal support in the familial setting. Using
longitudinal data, it was found that continued support of a mentor during the
period of idea generation in an occupational sphere correlated strongly with
the number of recognized creative contributions produced by a novice includ-
ing differences between male and female protégés’ needs for different kinds
of support (Torrance, 1983).
In the next concentric circle professional interpersonal support has its
place. A whole list of factors has been offered in the creativity literature as to
what organizational climates in the arts, sciences, nd industry should not be
like, if they were supposed to support creativity. Fostering fear of failure,
which results in a reluctance to take risks; a preoccupation with order and tra-
dition; a failure to see the strengths of the individuals working in an organiza-
tion; an over-reliance on ineffective algorithms; employees who are reluctant
to assert their own ideas; a reluctance to play; an excessive use of salient re-
ward (Amabile, 1983, 1996). All these negative factors represent serotonin-
level-reducing conditions. In contrast, the conditions cited in the literature as
favorable to innovation are a climate conducive to new ideas combined with
an established process for developing these ideas into products; an organiza-
tional structure flexible enough to bend with the requirements that innovation
may bring; support for innovation from the highest levels of management
(Amabile, 1983, p. 02, 1996, p 256). Such a serotonergic activity-stimulating
atmosphere can be sensed on the work floor in an elated and content staff.
In the social support literature hardly any reference is made to the novelty
generation process. Closest to the issue of creativity, social support has been
linked to intrinsic motivation (Van Yperen & Hagedoorn, 2003). Intrinsic mo-
tivation has been identified as key to creativity. Investigating the interaction
between job demands, job control, and job social support on fatigue and in-
trinsic motivation simultaneously, van Yperen and Hagedoorn identify work
conditions, which minimize job strain and maximize intrinsic motivation in
highly demanding jobs. Confirming earlier studies on intrinsic motivation, it
has been demonstrated that a controlling style of interaction towards their
subordinates had a detrimental effect on their intrinsic motivation. Autonomy
with respect to what and how organizational members carry out their work
supports their intrinsic motivation (Richer & Vallerand, 1995). Also it is ar-
Neuropsychological Support to the Novelty Generation Process / 113
gued that instrumental support defined here as ‘help from others to get the
job done when things get tough’ increases intrinsic motivation because it in-
creases the individual’s confidence that the work will be finished, creating
feelings of relatedness with others (Vallerand, 1997; Van Yperen & Haged-
oorn, 2003, p. 5). High levels of intrinsic motivation were found in less de-
manding jobs when job social support was high. Regardless of job control
and job demands, high job social support turned out to be the most effective
way of enhancing intrinsic work motivation (Van Yperen & Hagedoorn,
2003). So, it has been concluded that instead of reducing job demand and
thereby decreasing productivity, it was better to improve job control and job
social support to reduce strain and keep employees intrinsic work motivation
high. Self-confidence and being in control of ones own work clearly have
high serotonergic correlates.
Feeling honored, proud, self-confident and in control increases serotonin
levels; in contrast depression-triggering negative stereotypes like women ex-
perienced in the history of science are related with lower serotonin levels.
Excess interpersonal support such as excess control or evaluation has been ar-
gued to interfere with the intrinsic motivation individual and their capability
to support themselves (Schweizer, 2004). Particular norms and values can
predominate in the social environment of an actor, which withhold the actors
living in these settings to become self-supportive, but rather keep them de-
pendent, as has historically been the case with female novelty-seekers (Mil-
let, 1969). The motivation to keep others from becoming self-supportive and
providing them with excess support instead, is often a question of power
and/or insecurity of the support provider. In its most extreme form, it keeps
the dependent from becoming self-supportive while in addition negating so-
cial support as well. Both supporting an actor or withholding support from an
actor, means to exert a certain degree of control over this actor (Foucault,
1965). It means higher chances for the supporter to make the actor comply
with the aims of the supporter. These are situations, which may fire intrinsic
motivation of those discouraged from becoming self-supportive via the neg-
ative support mode.
Professional social rewards clearly have serotonergic correlates as is also
illustrated by a behavioral explanation of depression in people who have
stopped working in their sixties and then experience neuropsychological im-
balances: It is amongst others the lack of social rewards and validation exper-
ienced in a job which leads to decreased serotonin levels and depression
(Comer, 2002). Outside the concentric circles of private and professional in-
terpersonal support there is the realm of impersonal support sources including
public institutions, foundations, government, and what economists call “the
114 / Aesthetics and Innovation
market”. In the realm of self-support neuropharmacological interventions
(like in drug abuse) are a good example of impersonal support. Cannabis is
often used as a serotonergic stimulator.
There may also be downsides to excessive serotonin levels in the work
process, which is a big concern of many working in creative professions. Too
happy and mellow- does it erase the urge to express oneself in creative work?
In the history of art it is often argued that artists try to avoid to become “too
happy”, since this would keep them from good creative work. Depression
which is related to serotonergic deficits has been researched as a prerequisite
for good creative work (Kaufmann & Vosburg, 2002). However, whereas
mild depression seems to be well integrated into the novelty generation pro-
cess, severe depression that comes with stronger serotonergic deficits, dis-
ables any kind of productive work. So it has also been contended that it is not
the hard-core manic-depression but the milder cyclothymia that is supportive
to creative work (Jamison, 1993).
Finally, in terms of the Novelty Generation Model it can be argued that
apart from the often tedious process of transforming ideas into complete
products, it is also the innovative performance phase that is served mainly by
higher serotonin levels; after all, self-confidence and proactivity provide a
good basis for presenting one’s ideas to a wider public and convince them of
the quality of one’s products, in other words receive the social recognition
that is inherent in the innovative performance.
What will be of major interest in this context are the interactions of sero-
tonin and dopamine in the transitions between the different phases of the nov-
elty generation process. It can for instance be argued that for particularly high
level novelty-seekers increased dopaminergic activity disrupts the successful
completion of novelty-producing processes, which for instance may catapult
them back to the novelty-seeking phase of a new project, instead of finishing
the older less novelty-rewarding ones, a mechanism that keeps them from
reaping innovative performance for their work.

Support to the GABA- and Noradrenergic Side of the Novelty

Generation Process: About Anxiety and Panic
Anxiety-related neurotransmission has been found to be related to creat-
ive block (Flaherty, 2005). Combining this evidence with social-psychologic-
al support issues it can be argued that power and dependence issues that may
arise under all social support conditions have arguably the strongest neuro-
psychological implications for the GABA anxiety neurotransmitter systems in
Neuropsychological Support to the Novelty Generation Process / 115
the interpersonal forms of support. What are the effects of depending on oth-
Insights from the extrinsic versus intrinsic motivation debate (for instance
Eisenberger & Armeli, 1997) may be relevant here, as well as knowledge
about the personality factors and genetic set-up constituting the novelty-
seeker’s self-support. Any influence can produce extrinsic motivations in the
novelty-seeking individual. However, the good side of external impulses is
highly underrated in the creativity-intrinsic motivation debate. If intrinsic
motivation were the only good in the creative process, any interaction of nov-
elty-generating individuals with their environment would be undesirable.
However, there have always been researchers arguing that extrinsic motiva-
tion does not only have undermining effects to intrinsic motivation, but that it
may also be additive (Vroom, 1964). Even if the external motivation distracts
novelty-seekers from intrinsically motivated direction they work towards, the
social support may still give impulses, which may retrospectively be judged
as constructive support to the process of novelty generation. Creativity re-
searchers have implicitly written about such social support to creativity using
terms like ‘synergistic’ or ‘informational or enabling extrinsic motivation’
which is described as conducive to creativity as opposed to ‘controlling ex-
trinsic motivation’ which is argued to destroy intrinsic motivation (Amabile,
1996, p.119). Fears of being evaluated, controlled, criticized, or under the
pressure of external expectations then comes in as GABA-related paralyzing
distracters from relaxed and good-humored creative work.
Such effects have for instance been experienced by writers after having
launched a successful first novel. Like after receiving a prestigious reward,
first novelists report to be far more aware of this sense of implicit expecta-
tions from the public, which may easily result into a ‘writer’s block’. As the
novelist Thomas Wolfe wrote ‘Almost a year and a half had elapsed since the
publication of my first book and already people had begun to ask that ques-
tion which is so well meant, but which as year followed year was to become
more intolerable to my ears than the most deliberate mockery: “Have you fin-
ished your next book yet?” “When is it going to be published?”...now, for the
first time, I was irrevocably committed so far as the publication of my book
was concerned. I began to feel the sensation of pressure, and of naked desper-
ation, which was the become almost maddeningly intolerable in the next
three years.’(Wolfe, 1936, pp.49-50).
What sets off such psychological effects with almost panic-like symptoms
is the implicit agenda of social expectations and pressures behind grants and
rewards, which are comparable to those raised by the reception of a gift
(Schwartz, 1967). Rewards and grants both come with an obligatory response
116 / Aesthetics and Innovation
component, so psychologically they oblige receivers to realize and to make
use of the goodwill that others have given them, or to raise up to the expecta-
tions that others have with respect to them. It seems to be the burden of hav-
ing to confirm the characteristics, which social judges assumed when decid-
ing to confer the reward, also the expectations, which the reward created
among the audience of an reward event. The receiver’s response to such ex-
ternal obligations indeed varies according to the personality of the receiver.
High TCI Self-Directedness and Self-Transcendence (Cloninger et al., 1994)
is arguably the best protection against the potential threats of such external
There is an important difference between being able to obtain support and
being able to accept it. An individual’s ability to obtain support of others in-
cludes that the individual has to be able to convince others that he or she is
worthy of being supported. It also means having the self-confidence concern-
ing one’s work (see in the Novelty Generation Model and an accurate evalu-
ation of its potential performance with regard to the norms of the supporters.
This also implies a certain dependence on the judgment of the supporters. In
this sense, every form of support could be a potential threat to the intrinsic
motivation of the supported as discussed above, especially in the case of
strong differentials between the norms of supporters and the supported. Fears
of being controlled or dominated in the novelty generation process as well as
issues of pride may interfere with a novelty-seeker’s ability to accept support
offered by others. The need for independence may lead the novelty-creator to
refuse social support, which can be an expression of defensive behavior. A
shining example is the support-refusing behavior of Baudelaire in such con-
texts (Bourdieu, 1992). Depending on the situation, such defenses may in-
deed facilitate self-support, but may also be a sign of excessive self-support.
The decision to give uncompromising precedence to self-support eliminates
the possibility of constructive stimulation from the social environment. Work-
ers in novelty-seeking professions are per definition ideally highly self-sup-
portive, however they can be excessively self-supportive within their specific
domain of novelty-seeking, which expresses itself in stubbornness and fear of
and inability to accept any advice or criticism from others. This often stands
in gross contrast to their absorption of social support in auxiliary domains of
practical life, which do not directly affect their novelty-seeking activities, but
enable it practically as discussed in an earlier section on interpersonal sup-
Back to neuropsychological self-support often used by creative profes-
sionals to support focused deliberate creative problem-solving, relevant to the
GABA-related neural circuits there is the consumption of alcohol to achieve
Neuropsychological Support to the Novelty Generation Process / 117
a down-regulation of the pre-frontal cortex and the relaxation that supports
more spontaneous creative problem-solving. Alcohol abuse is also known to
be widely spread in the creative professions and a highly neglected topic in a
neuropsychology of creative workers. Alcohol leads to a down-regulation of
the prefrontal cortex which is more related to the spontaneous mode of creat-
ive problem solving than to the deliberate creative process. Evidence on the
relationship between the neuropsychological profile of the creative worker on
the one hand, and the situational support parameters on the other hand are es-
sential in researching support effects to the two different creative processing

Some Concluding Thoughts on Methodological Issues

In this article it has been argued that the overall support constellation of
individuals, their genetic, neurocognitive, and personality predispositions op-
erate together in an individual’s neural responses to different forms of inten-
ded support. This determines whether a support action turns out to be a factu-
al stimulant or depressant to on the one hand different phases of the novelty
generation process and on the other hand the two different creative pro-
cessing modes of deliberate or spontaneous problem solving. An inventory
has been presented here that comprises different support types, modes of sup-
port transmission, key neurotransmitter systems, and neuropsychological pro-
files of those engaged in the novelty generation process.
From a methodological point of view, a first step is to determine the
neuropsychological profile of an individual by means of administering neuro-
psychological test batteries. Such data provide evidence concerning activity
and balance between an individual’s key neurotransmitter systems, in other
words, which are the deficient neurotransmitter systems and which ones the
overactive ones in different phases of the novelty generation process. On the
basis of such knowledge it can become clearer how different support types
could establish, maintain or restore the neuropsychological balance needed
for each of the stages of the novelty generation process. This way it could for
instance also be explained how instrumental self-support in the form of drug
interventions affects this individual and why some novelty creators engage in
drug abuse like alcohol, caffeine and nicotine, cannabis, or cocaine and oth-
ers do not.
As a next methodological step, neuropharmacological intervention studies
in combination with neuroimaging methods, in particular functional Magnet-
ic Resonance Imaging (fMRI), are the ideal approach for the realization of
the research agenda set forth in this article. The first research projects on the
118 / Aesthetics and Innovation
effects of self-support have been conducted in the following contexts: first,
different personalities, like an fMRI study on creative processes in serotonin-
deficient obsessive-compulsives (Schweizer et al., 2006); second, the fear of
making errors, the neural correlates of which are recently receiving increased
attention (Nieuwenhuis et al., in press); and third, a serotonergic intervention
study on Cannabis-smoking before creative tasks (Schweizer et al., in prepar-
Far more difficult to research are the neuropsychological effects of other
forms of support, such as interpersonal support, for instance how direct inter-
personal praise works on the neurotransmitter systems of an individual. Argu-
ably even more difficult to grasp in controlled neuro-experimental settings is
impersonal support, for instance in the form of acquired fame. As described
in this article, art history teaches that fame can indeed have an effect on the
novelty-generating individual. It will take all our scientific creativity to de-
vise research designs that can capture such support effects on a neuropsycho-
logical level.
Furthermore, promising research ideas in the realm of self-support could
be to investigate other human states that are marked by comparable experi-
ences of ‘flow’—those transient states of hypofrontality that are so character-
istic of creative work processes and also occur in other human activities such
as athletic performance (Dietrich, 2004b). An interesting topic in this context
that is also far from understood is how nutrition high on either dopaminergic
or serotonergic precursors affects us in the novelty generation process. Fi-
nally, what touches all the above mentioned issues is the importance to re-
search the interactions between serotonergic and dopaminergic processes dur-
ing the novelty generation process and how they prepare the individual for
entering the different components of this process as depicted in the Novelty
Generation Model.
The theoretical basis provided in this article is meant to pave the way for
a thorough scientific investigation of such neuropsychological issues. Many
hypotheses can be derived from this article that need to be tested first to be
able to start practically applying such knowledge to the identification of an
optimal support mix for different individuals in different stages of the novelty
generation process. Starting to view social support interaction and self-sup-
port in the light of the regulation of neurotransmitter systems can affect all of
us who are engaged in research work, who receive support, but also provide
support to others. This may be a fruitful way to optimize our support interac-
tions by cross-fertilizing a neuropsychological perspective with a social psy-
chological one. In our understanding of how we ourselves and others tick in
support situations at the critical switches of the novelty generation process,
Neuropsychological Support to the Novelty Generation Process / 119
there seems to be plenty of room for neuropsychologically informed im-

Albert, R. S. (Ed.) (1983). Genius and eminence. The social psychology of
creativity and exceptional achievement (Vol. 5). Oxford: Pergamon Press.
Amabile, T. M. (1983). The social psychology of creativity. New York:
Amabile, T. M. (1987). The motivation to be creative. In S. G. Isaksen (Ed.),
Frontiers in creativity research: Beyond the basics (pp. 223–254). Buf-
falo, NY: Bearly.
Amabile, T. M. (1996). Creativity in context: Update to the social psychology
of creativity. Boulder, CO: Westview.
Bem, D. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in
experimental social psychology (Vol. 6, pp. 1–62). New York: Academic
Benes, F. M. (1997). The role of stress and dopamine-gaba interactions in the
vulnerability for schizophrenia. Journal of Psychiatry Research, 31, 257–
Benjamin, J., Ebstein, R. P., & Belmaker, R. H. (2002). Molecular genetics
and the human personality. Washington, D.C.: American Psychiatric As-
Benjamin, J., Li, L., Patterson, C., Greenberg, B. D., Murphy, D. L., &
Hamer, D. H. (1996). Population and familial association between the d4
dopamine receptor gene and measures of novelty seeking. Nature Genet-
ics, 12, 81–84.
Blum, K., Braverman, E. R., Holder, J. M., Lubar, J. F., Monastra, V. J.,
Miller, D., Lubar, J. O., Chen, T. J. H., & Comings, D. E. (2000). Reward
deficiency syndrome: A biogenetic model for the diagnosis and treatment
of impulsive, addictive and compulsive behaviors. Journal of Psychoact-
ive Drugs, 32(Suppl.), 1–68.
Bourdieu, P. (1992). Les règles de l'art. Genèse et structure du champ lit-
térair. Paris: Éditions du Seuil.
Carson, S. H., Peterson, J. B., & Higgins, D. M. (2003). Decreased latent in-
hibition is associated with increased creative achievement in high-func-
tioning individuals. Journal of Personality and Social Psychology, 85,
Cloninger, C. R. (1994). Temperament and personality. Current Opinion in
Neurobiology, 4, 266–273.
120 / Aesthetics and Innovation
Cloninger, C. R. (2004). Feeling good: The science of well-being. New York:
Oxford University Press.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994).
Temperament and character inventory: A guide to its development and
use. Missouri: Center for Psychobiology of Personality, Department of
Psychiatry, Washington University St.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiolo-
gical model of temperament and character. Archives of General Psychi-
atry, 1993, 975–990.
Collins, M. A., & Amabile, T. M. (1999). Motivation and creativity. In R. J.
Sternberg (Ed.), Handbook of creativity (pp. 297–312). Cambridge: Cam-
bridge University Press.
Comer, R. J. (2002). Fundamentals in abnormal psychology. New York:
Worth Publishers.
Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motiv-
ation. Journal of Personality and Social Psychology, 21, 105–115.
Deci, E. L. (1972). Intrinsic motivation, extrinsic reinforcement, and inequity.
Journal of Personality and Social Psychology, 22, 113–120.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determina-
tion in human behavior. New York: Plenum.
Deelstra, J. T., Peeters, M. C. W., Schaufeli, W. B., Stroebe, W., Zijlstra, F. R.
H., & Van Doornen, L. P. (2003). Receiving instrumental support at work:
When help is not welcome. Journal of Applied Psychology, 88), 224–231.
Dietrich, A. (2004a). The cognitive neuroscience of creativity. Psychonomic
Bulletin and Review, 11, 1011–1026.
Dietrich, A. (2004b). Neurocognitive mechanisms underlying the experience
of flow. Consciousness and Cognition, 13, 746–761.
Ebstein, R. P., Novick, O., Umansky, R., Priel, B., Osher, Y., Blaine, D., Ben-
nett, E. R., Nemanov, L., Katz, M., & Belmaker, R. H. (1996). Dopamine
d4 receptor (d4dr) exon iii polymorphism associated with the personality
trait of novelty seeking. Nature Genetics, 12, 78–80.
Eisenberger, R., & Armeli, S. (1997). Can salient reward increase creative
performance without reducing intrinsic creative interest? Journal of Per-
sonality and Social Psychology, 72, 652–663.
Flaherty, A. W. (2005). Frontotemporal and dopaminergic control of idea
generation and creative drive. The Journal of Comparative Neurology,
493, 147–153.
Foucault, M. (1965). Madness and civilization. A history of insanity in the
age of reason. New York: Pantheon Books.
Neuropsychological Support to the Novelty Generation Process / 121
Gambarana, C., Masi, F., Tagliamonte, A., Scheggi, S., Ghiglieri, O., &
Montis, d. M. G. (1999). A chronic stress that impairs reactivity in rats
also decreases dopaminergic transmission in the nucleus accumbens: A
microdialysis study. Journal of Neurochemistry, 72, 2039–2046.
Gay, P. (1999). Psychoanalyst Sigmund Freud. Time, March 29, 66–69.
Hamer, D. H., Sabol, S., Nelson, M. L., Fisher, C., Gunzerath, L., Brody, C.
L., Hu, S., Sirota, L. A., Greenberg, B. D., Lucas, F. R., Benjamin, J.,
Murphy, D. L., & Marcus, S. E. (1999). A genetic association for cigarette
smoking behavior. Health Psychology, 18, 7–13.
House, J. S. (1981). Work stress and social support. Reading, MA: Addison-
Jamison, K. R. (1993). Touched with fire: Manic depressive illness and the
artistic temperament. New York: The Free Press.
Kaufmann, G., & Vosburg, S. K. (2002). Mood in early and late idea produc-
tion. Creativity Research Journal, 14, 317–333.
Kelley, H. (1973). The processes of causal attribution. The American
Psychologist, 28, 107–128.
Keltikangas-Järvinen, L., Elovainio, M., Kivimäki, M., Lichtermann, D.,
Ekelund, J., & Peltonen, L. (2003). Association between the type 4
dopamine receptor gene polymorphism and novelty-seeking. Psychoso-
matic Medicine, 65, 471–476.
Lepper, M., & Greene, D. (1978). Overjustification research and beyond: To-
ward a means-end analysis of intrinsic and extrinsic motivation. In M.
Lepper & D. Greene (Eds.), The hidden costs of reward (pp. 78–97). Hill-
sdale, N.J.: Lawrence Erlbaum Associates.
Lerman, C., Audrain, J., Main, D., Boyd, N. R., Caporaso, N. E., Bowman, E.
D., Lockshin, B., & Shields, P. D. (1999). Evidence suggesting the role of
specific genetic factors in cigarette smoking. Health Psychology, 18, 14–
Martindale, C. (1999). Biological bases of creativity. In R. J. Sternberg (Ed.),
Handbook of creativity (pp. 137–152). Cambridge: Cambridge University
Martindale, C., & Hasenfus, N. (1978). EEG differences as a function of cre-
ativity, stage of creative process, and effort to be original. Biological Psy-
chology, 6, 157–167.
Martindale, C., & Hines, D. (1975). Creativity and cortical activation during
creative, intellectual, and eeg feedback tasks. Biological Psychology, 3,
Millet, K. (1969). Sexual politics. New York: Doubleday.
122 / Aesthetics and Innovation
Moghaddam, B. (2002). Stress activation of glutamate neurotransmission in
the prefrontal cortex: Implications for dopamine-associated psychiatric
disorders. Biological Psychiatry, 51, 775–787.
Nieuwenhuis, S., Schweizer, T. S., Mars, R. B., Botvinick, M. M., & Hajcak,
G. (in press). Error-likelihood prediction in the medial frontal cortex: A
critical evaluation. Cerebral Cortex.
Noble, E. P., Ozkaragoz, T. Z., Ritchie, T. L., Zhang, X., Belin, T. R., &
Sparkes, R. S. (1998). D2 and d4 dopamine receptor polymorphisms and
personality. American Journal of Medical Genetics, 81, 257–267.
Prolo, P., & Licinio, J. (2002). Drd4 and novelty seeking. In J. Benjamin, R.
P. Ebstein & R. H. Belmaker (Eds.), Molecular genetics and the human
personality (pp. 102–115). Washington, DC: American Psychiatry Pub-
Reynolds, K. J., Oakes, P. J., Haslam, S. A., Nolan, M. A., & Dolnik, L.
(2000). Responses to powerlessness: Stereotyping as an instrument of so-
cial conflict. Group dynamics: Theory, research, and practice, 4, 275–
Richer, S. F., & Vallerand, R. J. (1995). Supervisors' interactional styles and
subordinates' intrinsic and extrinsic motivation. The Journal of Social
Psychology, 135, 707–722.
Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal rela-
tionships as developed in the client-centered framework. In S. Koch (Ed.),
Psychology: A study of a science (Vol. 3, pp. 184–246). New York: Mc-
Sarason, B. R., Sarason, I. G., & Pierce, G. R. (1989). Social support: An in-
teractional view. New York: Wiley.
Schwartz, B. (1967). The social psychology of the gift. American Journal of
Sociology, 73, 1–11.
Schweizer, T. S. (2004). An individual psychology of novelty-seeking, cre-
ativity and innovation. ERIM Ph.D. Series, Nr. 48.
Schweizer, T. S. (2006). The psychology of novelty-seeking, creativity and
innovation: Neurocognitive aspects in a work-psychological perspective.
Creativity and Innovation Management, 15, 63–89.
Schweizer, T. S., Deijen, J. B., Heslenfeld, D., Nieuwenhuis, S., & Talsma, D.
(2006). Functional magnetic resonance imaging of brain activity during
rigid versus creative thought processes in obsessive-compulsive patients.
Paper presented at the Cognitive Neuroscience Society Conference, San
Francisco, CA, USA.
Schweizer, T. S., Hasselbaink, S., & Deijen, J. B. (in preparation). Serotoner-
gic stimulation and creative cognition: A tetrahydrocannabinol study.
Neuropsychological Support to the Novelty Generation Process / 123
Sexton, L. G., & Ames, L. (1977). Anne Sexton: A self-portrait in letters. Bo-
ston: Houghton Mifflin.
Stroebe, W., & Stroebe, M. (1996). The social psychology of social support.
In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology handbook
of basic principles (pp. 597–621). New York & London: The Guildford
Stroebe, W., Stroebe, M., Abakoumkin, G., & Schut, H. (1996). The role of
loneliness and social support in adjustment to loss: A test of attachment
versus stress theory. Journal of Personality and Social Psychology, 70,
Torrance, E. P. (1983). Role of mentors in creative achievement. Creative
Child and Adult Quarterly, 8, 8–15.
Vallerand, R. J. (1997). Toward a hierarchical model of intrinsic and extrinsic
motivation. Advances in Experimental Social Psychology, 29, 271–360.
Van Yperen, N. W., & Hagedoorn, M. (2003). Do high job demands increase
intrinsic motivation or fatigue or both? Academy of Management Journal,
46, 339–348.
Vroom, V. (1964). Motivation and work. New York: Wiley.
Wise, R. A., & Rompre, P. P. (1989). Brain dopamine and reward. Annual Re-
view of Psychology, 40, 191–225.
Wolfe, T. (1936). The story of a novel. New York: Scribner.

The Plural Self, Plural Achievement Motives,

and Creative Thinking

Leonid Dorfman and Anastasia Ogorodnikova

In this study we examined the link between the plural self, plural achieve-
ment motives and creative thinking among university students. Based on the
socioindividual world theory, the plural self was discriminated into subselves
and the plural achievement motives into submotives either of motives to
achieve success or to avoid failure. Hypothesized models were tested using
structural equation modeling. After a measurement model with an acceptable
fit to the data was found, a series of nonnested structural models was used
systematically to test the hypothesized models. Structural equation modeling
was also used to estimate the mediational effects of submotives between sub-
selves and creative thinking. The data indicated paths with significant posit-
ive coefficients from the autonomous and the dominant subselves to the dom-
inant submotive of the motive to achieve success to creative thinking. These
subselves and submotive enable creative thinking. There also were paths with
significant or nearly significant negative coefficients from the autonomous
subself to the autonomous submotive of the motive to avoid failure to creat-
ive thinking. The autonomous subself can enable creative thinking and under
some conditions the autonomous submotive of the motive to avoid failure is
decreased. The findings suggest parallel paths from subselves to submotives
to creative thinking. The autonomous and the dominant subselves can be al-
ternatives in concert with submotives as mediators. They are parallel ways to
think of diverse valuable alternatives (i.e., creative thinking).

Creativity is a cognitive ability, but it seems to occur within a matrix of

associated motivational, attitudinal, and personalogical traits (Martindale,
1989). Three domains can be converged upon to extend creativity studies:
creative thinking, self, and motivation (specifically, achievement motives).

Subselves, submotives, and creativity / 125
Creative or Divergent Thinking
Some cognitive studies of creativity focus upon divergent thinking abilit-
ies (Guilford, 1950, 1967). Divergent thinking, or the ability to think of di-
verse useful responses to a novel situation, forms a major part of what is of-
ten called creativity (Runco, 1991). Typical of the creator, divergent thinking
tends toward the novel or unknown. Of particular importance are cases where
a problem has yet to be defined or discovered, and where no set way of solv-
ing it exists (Dellas & Gaier, 1970). Divergent thinking tests ask an examinee
to think of as many appropriate uses for a common object (such as a brick) as
possible. Conversely, convergent thinking is oriented toward a known or cor-
rect solution. Convergent tests have a single correct answer, whereas diver-
gent tests do not.

The Self
As McCrae (1987) put it, although cognition and personality have tradi-
tionally been seen as distinct domains, creativity, seems to hold an intermedi-
ate position. In parallel to the cognitive approach to creativity, the personality
approach deals with personality and motivational variables as sources of cre-
ativity. We would emphasize the self as a part of the personality domain (al-
though the self is not the same thing as personality, e.g. Mischel & Morf,
2003). Just as personality and motivational variables, the self can also be seen
as a source of creativity. Well known findings show that the self relates to
creativity. Cattell (Cattell & Drevdahl, 1955; Drevdahl & Cattell, 1958)
found ego-strength to be associated with creativity in science and art. Barron
(1969) found superior ego-strength in creative writers and architects. Sum-
marizing many studies, Eysenck (1995) came to conclusion that high ego-
strength correlates highly with creativity. Self-esteem also seems to be linked
to creative thinking and creative performance (Barron & Harrington, 1981;
Osche, 1990; James et al., 1992; James & Eisenberg, 2004). Self-confidence
is related to creativity although assumptions vary. On the basis of copious
biographical and autobiographical material, Martindale (1989, 2001) and oth-
ers have held that high self-confidence is a creative trait. However, Kaufman
(2002) argued that creative eminence is associated with low self-confidence.
Self-evaluation and creativity are also related (Szymanski & Harkins, 1992).
Following psychoanalytic and humanistic theories, Wink (1999) emphasized
the dynamic nature of the self and its potential for growth. He drew attention
to relationships between creativity, on the one hand, and the actualized self,
self-individuation, self-cohesion, and the social self, on the other. Based upon
126 / Aesthetics and Innovation
several humanistic theories, Runco (1999) argued that self-actualization and
creativity are strongly related. Self-actualization hypothetically leads directly
to openness to experience and creativity.

Creativity consists of combining previously unrelated mental elements in
a new and useful fashion. In order to do this, the presence of certain motiva-
tional factors is necessary (Martindale, 1989). Many studies suggest that
motives and creativity are related. Much of this research has been within an
intrinsic motivation framework. It initially involved the argument that cre-
ativity is contingent in large part on intrinsic motivation and it can be under-
mined by reward and other extrinsic motives (Amabile, 1996; Amabile et al.,
1986; Hennessey, 1998; Hennessey & Amabile, 1988). There have been some
changes in the intrinsic motivation principle of creativity. Amabile (1996)
now argues that extrinsic motivation can be conducive to creativity, particu-
larly if intrinsic motivation is high. Other studies (e.g. Baer, 1998; Cropley,
1999; Eisenberger et al., 1998) found evidence in favor of extrinsic motiva-
tion as well.
Barron (1963) has emphasized the relevance of need for order in creativ-
ity. MacKinnon (1960) described creative individuals in terms of breadth of
interests and an absence of repression and suppression. Martindale (1989)
linked creativity with high levels of motivational factors such as interest,
curiosity, and ambition.
One of the oldest topics in psychology is the relationship between creativ-
ity and achievement. Beard (1874) was the earliest contributor to the study of
achievement and creativity. According to him, creativity is a function of two
underlying factors, enthusiasm and experience. Enthusiasm provides the mo-
tivational force to yield original work. Experience gives the achiever the abil-
ity to express original ideas in an intelligible fashion. Galton (1883) struggled
to explain individual differences in outstanding achievement in various do-
mains of creativity. After a century, Mumford and Gustafson (1988) focused
on creative behavior and innovative occupational achievement. Helson et al.
(1995) examined creative achievement and looked at the personality of the
creative achiever. In particular, the motivational-identity system was spe-
cified. It was characterized by energy for self-chosen work, career ambition,
persistence, and commitment to creative endeavor. Friedman and Förster
(2005) showed an influence of approach and avoidance anticipatory states on
Subselves, submotives, and creativity / 127
It comes as a surprise that except for a few accounts (e.g. McClelland et
al., 1953) there has been almost no systematic research on the need for
achievement and creativity. The classic achievement motivation approach of
McClelland and Atkinson (Atkinson, 1957; McClelland et al., 1953) identi-
fied motive dispositions as central determinants of achievement behavior. A
distinction was made between the desire for success (i.e., need for achieve-
ment) as an approach motive and the desire to avoid failure (i.e., fear of fail-
ure) as an avoidance motive.

Self and Achievement Motives related to Creativity: Some

Perspective Outlines
Of particular importance with a focus on creativity would be at least two
issues. First, are self and motives (the need for achievement) related? Second
is the plural account of the self and motives (again, an emphasis is given to
the need for achievement)?

Self and Motives

For the purpose of this section, we need not prefer one theory or study
over the other. Rather, we shall try to point out some theories and studies in
which the self and motives are related. Given this, our major emphasis is first
on motives treated from the viewpoint of the self and second, inversely, on
the self taken from the perspective of the need for achievement.
Research and theory on the self of the 1970s dealt primarily with the “self
as known” or as an object. That is, it focused on the self-concept or what Wil-
liam James (1890) called the “me.” In the 1980s the self construct acquired
personal agency such as self-evaluation, self-enhancement, self-defense, self-
regulation, self-efficacy, and self-control. The self was portrayed as a “doer”
as well as “thinker” and “feeler.” That is, the focus turned to what James
(1890) had called the “I.” This framework implied motivation. It raised issues
concerning the self’s agency and associated motives (for details see Mischel
& Morf, 2003). Agentic qualities of the self led to the view that it possesses
motivational power (Gaertner, Sedikides, & Graetz, 1999; Martindale, 1980;
Pyszczynski et al., 2004) and even motivational primacy (Gaertner et al.,
1999; Gaertner et al., 2002).
In his social cognitive theory, Bandura (1999) drew attention to the dicho-
tomy separating the self into agent and object. People are agents when they
act on the environment but objects when they reflect and act on themselves.
128 / Aesthetics and Innovation
In the agentic sociocognitive view, people are self-organizing, proactive, self-
reflecting, and self-regulating. The self is also agentic when it is motivating
and creative. Conversely, the self is reactive when individuals exert no motiv-
ational, self-reflective, creative, or directive influence on a process. But self-
reactors possess a capacity for self-direction. Once the capability for self-dir-
ection is developed, self-demands and self-sanctions appear as major guides,
motivators, and deterrents of human behavior. Thus, there is at least a two-
faced direction in which the self system can regulate the level of motivation.
Generally, rather than splitting the self into agent and object, social cognitive
theory treats this static dichotomy as a dynamic system operating interact-
Martindale (1980) dealt with this question from another perspective; that
is from a cognitive model. He put forward the view that the hierarchically
structured action system is a basic mental module. The action system consists
of a large number of action units. Each action unit codes one of the actions
that we can perform (e.g., walking, opening a door). Input to them comes
from two sources. First, there are inputs from perceptual and semantic mod-
ules. Thus, we do not attempt to open a door if we do not perceive one to be
present. Second, action units also receive inputs from deeper levels of the ac-
tion system. Martindale postulated several deeper strata in the action system.
Immediately below the level of action units are nodes that code plans or
scripts (e.g., going to a restaurant). We do not open a door unless doing so is
part of the script we are presently enacting. At the next lower level are units
coding dispositions. These units were called motives to the extent they are
connected to the emotional system. At the bottom level are subself units.
Each of these is vertically connected to a set of motive or dispositional units.
The same motive may be connected to several subself units. The regnant sub-
self selects which dispositions or motives will be given high priority. In gen-
eral, a subself was seen as a mediating or chunking unit that allows the con-
nection of units in other analyzers to dispositional units. Deeper-level units in
other analyzers cannot be directly connected to dispositional units but must
be linked to them via mediating subself units. Given this, by the term self
Martindale referred to what James (1890) called the I or the self as actor
rather than to the me or self-concept. He also agreed with James that we do
not have a single self but a multiplicity of them.
Researchers hold that some motives are a ground of and are rooted in the
self not only in a theoretical fashion but also when conducting empirical stud-
ies. For instance, Andersen and Chen (2002) proposed an interpersonal so-
cial-cognitive theory of the relational self (knowledge about the self is linked
with knowledge about significant others). In this framework they showed that
Subselves, submotives, and creativity / 129
an individual’s repertoire of relational selves is a source of interpersonal pat-
terns involving affect, motivation, self-evaluation, and self-regulation. Some
models of self-regulation include motivation also in terms of goals (e.g., San-
sone & Thoman, 2005). The self-evaluation maintenance model assumes self-
enhancement motives in domains in which another person outperforms the
self (Tesser, 1988). Sedikides (1993) compared three major self-evaluation
motives: self-assessment (people pursue accurate self-knowledge), self-en-
hancement (people pursue favorable self-knowledge), and self-verification
(people pursue highly certain self-knowledge). He found that the self-en-
hancement motive emerged as the most powerful determinant of the self-
evaluation process, followed by the self-verification motive. Erez and Judge
(2001) studied core self-evaluations (self-esteem, locus of control, general-
ized self-efficacy, and neuroticism as a broad personality trait) related to mo-
tivation. They found that this broad trait is related to motivation and perform-
ance. The four dispositions loaded on one higher-order factor. The higher or-
der trait was related to task motivation and performance in a laboratory set-
Another line of inquiry stems from studies of achievement motivation that
focus on contributions of the self. Heckhausen (1982) identified centering on
a self-produced outcome as the first element of the achievement motive. A
consequence of the self reorientation of motive theory has been to call into
question the meaning of the need for achievement. McClelland and his asso-
ciates (McClelland et al., 1989; Weinberger & McClelland, 1990; see also:
Brunstein & Maier, 2005) suggested that “self-attributed” motives (question-
naire measures) should be distinguished from “implicit” motives (TAT pro-
jective measures) because they reflect two different motivational systems.
The contemporary achievement goal approach pioneered by Dweck (1986)
and Nicholls (1984) identified two main goal orientations: task orientation
and ego orientation (Duda, 1993; Nicholls, 1989; Skaalvik, 1997). In particu-
lar, different dimensions of ego orientation (self-defeating and self-enhan-
cing) have been given particular attention as referred to achievement motives
(Skaalvik, 1997). In goal-setting research (Bandura, 1989, 1991; Bandura &
Wood, 1989) stronger self-efficacy has been shown to lead to higher self-set
goals (Locke & Latham, 1990). Self-efficacy and need for achievement were
positively related to goal level (Phillips & Gully, 1997). Dutton and Brown
(1997) started another line of research. They have shown that global self-es-
teem (the way people generally feel about themselves) and attribute-specific
self-esteem (the way people evaluate their specific attributes and abilities)
guide people’s cognitive and emotional reactions to success and failure. It
was found that specific self-views predict people’s cognitive reactions to task
130 / Aesthetics and Innovation
performance whereas global self-esteem predicts people’s emotional reac-
tions to task performance. Success and failure can be viewed as closely re-
lated to the need for achievement. Sedikides and Strube (1997) pointed out
self-assessment or self-enhancement views as more appropriate in accounting
for the choices of achievement-motivated individuals. Thompson et al. (1995)
showed self-worth protection in achievement motivation and subsequent per-
formance effects and attributional behavior.

A Plural Account of Self and Motives

Like divergent thinking as opposed to convergent thinking, the plural self

can be opposed to the unitary self and the plural achievement motives to a
unitary achievement motive.
Very few studies took into account the plural self as a source of creativity.
Barron and Harrington (1981) noted that in the self–concepts of creative per-
sons opposite or conflicting traits may coexist. Martindale (1980) postulated
that a person may have many subselves which can be linked to creativity. Be-
sides, one can see some indirect evidence in favor of the plural self as a
source of creativity. Cropley (1999) noted the paradox that stems from the
simultaneous coexistence in creative people of psychological elements that
seem logically to be mutually contradictory. As Csikszentmihalyi (1997) put
it, creativity is the property of a complex system, and none of its components
alone can explain it. The creative person has contradictory extremes—instead
of being an “individual,” he or she is a “multitude.”
To set the context, it is essential first to underscore the plural self account
without reference to creativity. James (1890) considered the self as composed
of the I (the self-as-actor), the Me (the self-concept), and the Mine (one’s
possessions). In the framework of a narrative approach Hermans (1996) has
put forward the view that Me is the protagonist in a story while the I is an au-
thor who can imaginatively construct that story. This view of the self system
focuses on its diverse aspects and functions which are not isolated or uncon-
nected components but rather interacting facets of a coherent system that op-
erates at multiple levels concurrently. It suggests that the self may be usefully
conceptualized as a coherent organization of cognitive-affective representa-
tions (Mischel & Morf, 2003). On the other hand, social cognitive theory
(Bandura, 1999) rejects the fractionation of human agency into multiple
selves. Bandura argues that diversity of action arises not from a collection of
agentive selves but from the different options of the same agentive self.
Likewise, few attempts have been systematically made to study the plural
account of motives and the plural achievement motive specifically. In his the-
Subselves, submotives, and creativity / 131
ory of end goals (“doing something for its own sake”), Reiss (2004) found
evidence in favor of the multifaceted nature of intrinsic motives. They pertain
to what some have called ego motives. Curiosity, autonomy, and play are ex-
amples of intrinsic motives. The multifaceted theory holds that intrinsic
motives are largely unrelated to each other. Multifaceted theorists include
psychologists who have put forth evolutionary theories of motivation (e.g.,
McDougall, 1926). Similarly, Reiss discerned unrelated motives as genetic-
ally distinct sources of motivation with different evolutionary histories.
Atkinson (1957) posited a model explaining how the motive to achieve and
the motive to avoid failure influence behavior. It was assumed that strength
of motivation is a multiplicative function of motive, expectancy, and incent-
ive. Performance level should be greatest when there is greatest uncertainty
about outcome. In applied psychology, researchers developed multidimen-
sional measures of achievement motivation and related dispositions (Cassidy
& Lynn, 1989; House et al., 1996; Johnson, 1990; Ward, 1997).

Major Issues of Self and Achievement Motives related to Creativity

Apparently, self and motives are closely related as the above mentioned
studies show. The former and the latter would be linked with creativity, as
well. However, this concern is too general. It would be interesting to disen-
tangle self and achievement motive determinants of creativity more subtly.
We would put aside views on motives as rooted in the self. It would seem,
instead, much confusion will be avoided if we clearly distinguish motives
from the self but consider them as interacting with the self. Given this, a plur-
al account of the self and achievement motives would be of particular in-
terest. There has been no systematic research on these questions, but it would
be well worth undertaking.
Creative thinking must occur with the self and achievement motives in
combination. As was mentioned above, creative thinking is also the ability to
think of diverse useful alternatives. When we put forward the plural account
to the self and achievement motives, a new issue arises. One focal point is
which self-aspect and achievement motive-aspects should be taken together
and incorporated with creativity. Another concern is with self-aspects and
achievement motive-aspects as diverse useful alternatives which could lead
in parallel to thinking of diverse useful alternatives.
Oddly enough, there is diversity in theory and research but no coherent
body of research on divergent or creative thinking related to the plural self
and plural achievement motives. Furthermore, there is no compelling ra-
tionale for selecting a coherent theory among those mentioned above.
132 / Aesthetics and Innovation
However, such a coherent theory is necessary. From this perspective, of par-
ticular importance are also subselves and achievement submotives as predict-
ors of creative thinking.
We proceed to lay a tentative foundation for such a coherent theory. Of
relevance would be the socioindividual world theory (Dorfman,1993). It
opens room for divergent or creative thinking, the plural self, and plural
achievement motives taken together.

The Socioindividual World Theory

The socioindividual world

Systems ideas are valuable because they give rise not only to an under-
standing of how people function in a coherent fashion but they also provide
some directions for empirical research. Some researchers have put forward a
systems view of creativity (e.g., Csikszentmihalyi, 1999; Gruber, 1999;
Rathunde, 1999). Most systems perspectives have been expanded to include
personality research (e.g., Magnusson, 2001; Mayer, 1998; Pervin, 2001).
The eminent Russian psychologist Volf Merlin (1986) elaborated the “in-
tegral study of individuality” from an hierarchical multi-level systems per-
spective. He treated individuality as self-regulated, hierarchical levels struc-
tured and arranged in a complex integrated system.
The socioindividual world theory (Dorfman, 1993, 1995) is derived from
the Merlinian theory. The crucial question to be debated is how does the in-
tegral individuality, being a complex system and operating for its own sake,
respond to and incorporate at the same time social rules, demands and ex-
pectations. A perspective raised in the socioindividual world theory is the
nature of the relationships between integral individuality and the social envir-
onment from a multi-systems perspective.
Integral individuality taken in the social-cultural context would be seen as
the intersection of personal and social-cultural life. This broad scope was
called the “socioindividual world.” The view is similar to Sullivan’s (1953)
definition of personality as “the relatively enduring pattern of recurrent inter-
personal situations which characterize a human life” (pp. 110–111). The so-
cioindividual world theory has been described in detail elsewhere (e.g., Dorf-
man, 1993, 1995, 1997, 2004, 2005). In general, it deals with the individual-
ity system and the social-cultural system (in which individuals dwell) interac-
tions from a multisystems perspective. These systems function for their own
sake but also extend to each other. The system extensions are fixed using the
concept of socioindividuality.
Subselves, submotives, and creativity / 133
First, an individual invests himself or herself in socio-cultural events. The
target can be not only other people but also objects that are material or imma-
terial in nature. Changes produced by an individual in social-cultural events
are called “socioindividuality.” Its source is the individuality system but it
operates in the realm of social-cultural events (individuality is a system and
socioindividuality is its subsystem). Second, an individual can be depicted as
being shaped, regulated, and controlled by the socio-cultural system. Changes
the latter produces in the individual are also called “socioindividuality.”
However, its source is the social-cultural system and appropriate individual
changes are its realm. In this case, the socio-cultural realm is a system and
socioindividuality is its subsystem. Thus, socioindividuality is viewed as
double-sided, because it relates to either the individuality system or the socio-
cultural system. In sum, the socioindividual realm is viewed as the plural het-
erogeneous multidimensional multisystem entity that describes an individual-
social settings interaction.
For purposes of research, the notion of regions is of particular importance.
This notion is used to emphasize functionally different units of the socioindi-
vidual world if it falls into systems and their subsystems. Second, it is intro-
duced to address functionalist concerns conceptualized in a structuralist
framework. Third, the socioindividual realm is a kind of confluence of theor-
ies. The notion of regions is used to emphasize collections of properties
rather than single variables. Fourth, a regulation criterion can provide a way
in which the socioindividual world is differentiated. Given this, the socioindi-
vidual realm may be divided into four broad regions.
The first major region is contingent upon common internal or internally
restricted self-regulation. It may be called “autonomy.” Hypothetically, this
region would include a wide set of different variables. To cite but a few ex-
amples, they would be sense of personal identity and sameness through time
(James, 1890), internal locus of control (Rotter, 1990), internal perceived
locus of causality and autonomy orientation (Ryan & Deci, 2000), the causal
agent (Snyder & Higgins, 1997), and autonomy and independence as key
components of creative individuals (Feist, 1999).
The second major region appears to be due to common internal but exten-
ded self-regulation. It is commonly termed “dominance.” Hypothetically, this
region includes a wide set and different variables. To cite but a few examples,
they would be the assimilation of reality to the mind (Piaget, 1975/1985),
psychological ownership (Pierce et al., 2003), dominance (Wiggins, 1995),
extraversion as a major dimension of personality (Eysenck, 1970), and im-
age-guided orientation of artists (Cupchik, 1999).
134 / Aesthetics and Innovation
The third major region appears to be due to external other-regulation and
is based on other-acceptance. It is commonly termed “empathy.” Hypothetic-
ally, this region would include a wide set and different variables. They would
be, for instance, external perceived locus of causality (Deci & Ryan, 1985),
external criteria for evaluation as a subscale of extrinsic motivation (Harter,
1981), and evaluation of other people, as well as experts’ evaluations and aes-
thetic appraisals according to professional criteria (Csikszentmihalyi, 1999).
The fourth major region appears to be due to internal or relational to oth-
er-regulation. It is commonly termed “submission.” Hypothetically, this re-
gion includes a wide set and different variables. To cite but a few examples,
they would be experience of pressure and tension, just as being controlled by
external events or response on the promise of a reward (Deci & Ryan, 1985),
submission (Wiggins, 1995), relatedness to others (Hodgins et al., 1996), and
rule-guided orientation of artists (Cupchik, 1999).
Some data obtained by means of principal components analysis using
Varimax normalized rotation were consistent with these regions functioning
as latent factors and guiding their manifest variables. Dorfman and Ogorod-
nikova (2004) examined the personality domain using the adjective checklist
developed by Gough and Heilbrun (1983) and three sets of scales—autonomy
and aggression, achievement and affiliation, succorance and abasement. The
loadings of the first, second, and third pairs of variables appeared on three or-
thogonal factors. The first was called autonomy, the second dominance, and
the third submission. Dorfman and Liakhova (2004) examined dominance
and nurturance measured by the Interpersonal Adjective Scales (Wiggins,
1995) and extraversion and social desirability measured by the Eysenck Per-
sonality Questionnaire (EPQ-R Adult) (Eysenck & Eysenck, 1994). It was
shown that dominance and extraversion were grouped in one component and
nurturance and social desirability appeared on a second component. The first
factor was labeled dominance and the second factor was called submission.
Dorfman and Gasimova (2004) examined extraversion and social desirability
measured by the Eysenck Personality Questionnaire (Eysenck & Eysenck,
1994), sensation seeking measured by the Sensation Seeking Scale (Zucker-
man, 1979), and some dimensions of temperament measured by the Rusalov
temperamental questionnaire (Rusalov, 1997). The loadings of extraversion,
sensation seeking, and people-related ergicity were grouped in one factor
called dominance and the loadings of social desirability and object-related er-
gicity appeared on another factor called submission.
Of course, the data obtained are preliminary and need to receive addition-
al empirical support. Besides, variables related to the empathy region are still
to be tested. Also, the studies should be extended to test all of the hypothes-
Subselves, submotives, and creativity / 135
ized variables mentioned above. Notwithstanding, the beginning seems prom-
ising and fruitful.

The plural self

Another aspect of the socioindividual realm theory is the notion of the

plural self. Kihlstrom et al. (1997) construed the self as a person’s mental
representation of his or her own personality. Likewise, Dorfman (2004)
defined the plural self as a person’s mental representation of his or her own
socioindividual realm. Then, one can specify a number of subselves as they
represent the socioindividual world regions. Four subselves are postulated.
The first is called the autonomous subself. It is a mental representation of the
autonomy region. The second is called the dominant subself. It is a mental
representation of the dominance region. The third was called the empathic
subself, and the fourth the submissive subself. The empathic subself would be
considered as a mental representation of the empathic region and the sub-
missive subself as a mental representation of the submission region.
The autonomous subself is close to the I or the self-as-knower, a sense of
personal identity and a sameness through time (James, 1890) or the self as
actor (Martindale, 1980), or the self as author (Hermans, 1996). The domin-
ant subself is similar to “mine and property” that attaches people to objects
(Furby, 1991; James, 1890; Pierce et al., 2003). The empathic subself is like
an internalized generalized other (Mead, 1934). The submissive subself is
similar to the Me or self-as-known (James, 1890; Mead, 1934) or self-
concept (Martindale, 1980) or the Me as a complex, narratively structured
self (Hermans, 1996).
When the plural self is defined as a person’s mental representation of his
or her own socioindividual realm, the construct of mental representation is
not operationalized and would seem to be speculative. Nevertheless, evidence
for variables of the socioindividual world regions and subselves appears in
the common latent factors. Dorfman (2006) reanalyzed data obtained by his
associates using principal component analysis using Varimax normalized ro-
tation. He tested hypotheses that variables of the socioindividual world re-
gions and subselves would be clustered on the same factors. In addition to
variables of the socioindividual world regions measured by the question-
naires mentioned above, subselves were measured by the Perm Plural Self
Questionnaire (Dorfman et al., 2000).
According to Ogorodnikova’s data, three orthogonal factors appeared. On
Factor 1, the loadings of autonomy, aggression, and the autonomous subself
were highest and significant (autonomy). On Factor 2, the loadings of
achievement, affiliation, and the dominant subself were highest and signific-
136 / Aesthetics and Innovation
ant (dominance). On Factor 3, succorance, abasement, and the submissive
subself were highest and significant (submission). Based on Liakhova’s data,
it was found that the loadings of dominance, extraversion, and the dominant
subself were included in one component (dominance). The loadings of nur-
turance, social desirability, and the submissive subself were included in an-
other component (submission). Dorfman’s associate Tokareva examined ex-
traversion and social desirability measured by the Eysenck Personality Ques-
tionnaire (EPQ-R Adult) (Eysenck & Eysenck, 1994), and Leary’s (1957) in-
terpersonal octant categories, such as Managerial—Autocratic, Competitive
—Narcissistic, Rebellious—Distrustful, Docile—Dependent, Cooperative—
Overconventional measured by the Interpersonal Check List (Laforge &
Suczek, 1955; Leary, 1956). She found two orthogonal factors. On Factor 1,
the loadings of extraversion, Managerial—Autocratic, and Competitive—
Narcissistic octants, and the autonomous subself were highest and significant
(autonomy). On Factor 2, the loadings of social desirability, Rebellious—Dis-
trustful, Docile—Dependent, and Cooperative—Overconventional octants,
and the submissive subself were highest and significant (submission).
However, some mixing was revealed between the autonomous and the dom-
inant subselves. The point is when the autonomous subself was replaced with
the dominant subself a similar structure appeared but the first factor would be
titled dominance.
Dorfman, Fenogentova, et al. (2000) used a one-way ANOVA to study
subself effects on Rotter’s locus of control measured by Bazhin et al.’s (1984)
modified questionnaire. It was found that the autonomous subself produced
significant increased effects on internal control related to either general con-
trol expectations or achievement behavior. The submissive subself produced
significant decreased effects on internal control related to general control ex-
pectations. The data obtained suggest that the autonomous subself is in con-
cert with internal control, whereas the submissive subself acts in concert with
external control.
Schebetenko (2005) used the General causality orientations scale (Deci &
Ryan, 1985) adopted to a Russian population (Dergacheva et al., 2002) and
the Perm Plural Self Questionnaire (Dorfman, Riabikova, et al., 2000) to
study by means of a three-way ANOVA (within-subjects design) whether
autonomy, control, and impersonal orientations subscales and the subselves
scales interact. He grouped the autonomous subself and the autonomy causal
orientation subscale in one within-subjects factor, the dominant subself and
the control causal orientation subscale in a second within-subjects factor, and
the submissive subself and the impersonal causal orientation subscale in a
third within-subjects factor. The within-subjects factors interacted signific-
Subselves, submotives, and creativity / 137
antly. The first factor was called autonomy, the second dominance, and the
third submission. This means, in particular, that these factors incorporate the
mentioned subselves and causal orientations in some but not all combina-
tions. Still, these factors contrast with one another.
These findings show that subselves as they are articulated within the plur-
al self and the socioindividual world regions are closely related. However, the
empathic subself remains unvalidated because the variables related to the em-
pathic socioindividual world region are still to be tested.

The plural achievement motives

There are essential differences between the self and achievement motives.
The self is an experiencing subject, beliefs about oneself, and an executive
agent (Leary & Tangney, 2003). On the other hand, achievement motives
consist of energizing a person to attain some performance level. On this level,
the achievement motives would be seen as dispositions oriented toward
achievement behavior, to excel or attain success in some field of endeavor.
On another level, achievement motives are woven in self-related phenomena
(e.g., Heckhausen, 1982). Hence, it is not surprising that achievement
motives could be mediators between the self and creative behavior or related
Good results with the plural self give rise to the idea of extending the
plural account to achievement motives. The latter would relate to each sub-
self and then be differentiated as appropriate submotives. Given this, each
submotive can be defined through one of the subselves as they are specified
in the plural self model (subself-related achievement submotives). Based
upon the plural self model (Dorfman, 2004) and classic achievement motiva-
tion theory (Atkinson, 1957; McClelland et al., 1953; Mehrabian, 1969,
1994/1995) the concept of plural achievement motives was elaborated (Dorf-
man et al., 2002). It was suggested that either the motive to achieve success
(e.g., need for achievement, an approach motive) or the motive to avoid fail-
ure (e.g., fear of failure, an avoidance motive) fall into four submotive cat-
egories, that is the autonomous, the dominant, the empathic, and the sub-
missive achievement submotives.
Dorfman et al. (2002) tested two research hypotheses. First, the four-
factor model can provide a description of subself-related achievement sub-
motives within either the motive to achieve success or the motive to avoid
failure. Second, subselves and subself-related achievement submotives cor-
relate and interact with either the motive to achieve success or the motive to
avoid failure. Subselves were measured with the Perm Plural Self Question-
naire (Dorfman et al., 2000) and submotives of the motive to achieve success
138 / Aesthetics and Innovation
and the motive to avoid failure with the Perm Plural Achievement Motives
Questionnaire (Fenogentova, 2002). Principal components analysis with Var-
imax normalized rotation for the motive to achieve success scores yielded
four orthogonal components. The autonomous, the dominant, the empathic,
and the submissive submotive scores loaded on different factors. Similar data
were obtained for the motive to avoid failure. These submotive structures
were supported in Dudorova’s (2004) study. Dorfman et al. (2002) conducted
also a 3 x 4 ANOVA treating subselves, submotives of the motive to achieve
success, and the motive to avoid failure as three within-subjects factors. The
results indicated their significant interaction.

A socioindividual model of creative thinking

Based on the socioindividual theory a model of creativity was developed

(Dorfman, 2005). Instead of looking at personality (individuality) from a cre-
ativity systems perspective, creativity was examined from a personality (indi-
viduality) systems perspective. Creative thinking cannot be viewed as a re-
gion of the socioindividual realm itself. Rather, creative thinking may be seen
as a result or function of how the socioindividual realm is arranged.
Moreover different kinds of regulation and regions would differ in their con-
tribution to creative thinking. Crucial research questions that arise are: Where
is creative thinking located in the socioindividual world, and how do the re-
gions of the socioindividual realm contribute to creative thinking.
At least three levels of creativity research in the socioindividual realm
were applied. Specifically, three kinds of hypothesized models were tested.
The first does not deal directly with creative thinking; it deals with variables
that relate to different regions of the socioindividual realm. The subject of the
hypothesized models here is which variables would present regions of the so-
cioindividual realm as latent factors. The second model addresses the issue of
whether creative thinking can be treated as rooted within the socioindividual
realm and guided by its compounds. If so, exogenous factors should be dis-
tinguished from endogenous factors. The subject of the hypothesized models
here is which variables referred to different regions (exogenous factors) can
reveal and predict the variability of creative thinking (the endogenous factor).
The third types of models deal with mediation. The subject of the hypothes-
ized models here is assessing regional latent constructs (mediators) interven-
ing in the relationships between another region’s latent construct (an exogen-
ous factor) and creative thinking (an endogenous factor). That is, regions of
the socioindividual realm would be considered as a chain regulating creativ-
ity. However, endogenous factors (for instance, intelligence) which can medi-
Subselves, submotives, and creativity / 139
ate links between variables referred to several regions (exogenous factors)
and creative thinking (an endogenous factor) were hypothesized, as well.
In Dorfman and Gasimova’s (2004) study extraversion, sensation seeking,
and people-related ergicity were indicated as manifest variables of the domin-
ance region and social desirability and object-related ergicity as manifest
variables of the submission region. Fluency, flexibility, and originality were
indicated as manifest variables of creative thinking, measured by the Altern-
ate Uses Test (Wallach & Kogan, 1965). In terms of path analysis (structural
equation modeling) a path was hypothesized creative thinking from the dom-
inance exogenous factor but not from the submission exogenous factor. The
hypothesized model fit the data in such a way that most indices were signific-
ant and some very close to being significant. The model summary indicated
that the dominance exogenous factor included extraversion, sensation seek-
ing, and people-related ergicity as its manifest variables. The submission exo-
genous factor included social desirability and object-related ergicity. Exogen-
ous factors correlated positively. The creative thinking endogenous factor in-
cluded flexibility and originality as its manifest variables. The path with a
significant coefficient was to the creative thinking endogenous factor from
the dominance exogenous factor but not from the submission exogenous
factor. These findings are consistent with results obtained by Martindale and
Dailey (1996) for extraversion and by Vartanian, Martindale, and Kingery
(2002) for Cloninger’s (1987) novelty seeking dimension and cognitive disin-
Dorfman and Liakhova (2004) found a similar path from the dominance
exogenous factor to the creative thinking endogenous factor although extra-
version and dominance were manifest variables of the dominance region and
social desirability and nurturance were manifest variables of the submission
region. Besides, it was found that the dominance factor can mediate the path
from the submission exogenous factor to the creative thinking endogenous
Dorfman and Gasimova (2004) tested intelligence, measured by the Ger-
man Intelligence Structure Test (IST-70) developed by Amthauer, 1973, as a
mediating factor between the dominance exogenous factor (extraversion, sen-
sation seeking, and people-related ergicity were its manifest variables) and
the submission exogenous factor (social desirability and object-related ergi-
city were its manifest variables) and the creative thinking endogenous factor
(fluency, flexibility, and originality were its manifest variables). The paths
with significant coefficients were from the submission and the dominance
exogenous factors to the verbal intelligence factor to the creative thinking en-
dogenous factor.
140 / Aesthetics and Innovation
Preconditions and research hypotheses

It could be that creative thinking is contingent on the dominance region,

besides, the dominance region would be a mediator between the submission
region and creative thinking (Dorfman & Gasimova, 2004; Dorfman & Liak-
hova, 2004). Because the dominant subself relates to the dominance region
and the submissive subself to the submission region (Dorfman, 2006) we
might guess that the dominant subself enables creative thinking and the sub-
missive subself enables it through the dominant subself. The dominant and
the submissive subselves were related but neither the dominant subself nor
the submissive subself correlated significantly with creative thinking (Koval-
eva, 2000).
It would seem that subselves do relate to creative thinking but some medi-
ating conditions should be taken into consideration. Ogorodnikova (2006)
found no significant correlations between subselves and creative thinking.
However, significant correlations were found first between some submotives
of the motive to achieve success and creative thinking. The dominant and the
submissive submotives and creative thinking variables correlated positively.
Second, significant correlations were found between some submotives of the
motive to achieve success and subselves. In particular, the autonomous sub-
motive and the autonomous subself variables correlated positively. The dom-
inant submotive and the dominant subself correlated positively. The empathic
submotive and the empathic subself correlated positively. Likewise, signific-
ant correlations were found first between some submotives of the motive to
avoid failure and creative thinking. The autonomous and the empathic sub-
motives and creative thinking correlated negatively. In turn, significant cor-
relations were found between some submotives of the motive to avoid failure
and subselves. The autonomous submotive and the autonomous subself vari-
ables correlated negatively.
Based on data obtained by Ogorodnikova (2006) three hypothesized mod-
els were developed in terms of path analysis.
Hypothesized model # 1. There are paths from the autonomous and the
dominant subselves to the dominant submotive of the motive to achieve suc-
cess to creative thinking.
Hypothesized model # 2. There are paths from the autonomous subself to
the autonomous submotive of the motive to avoid failure to creative thinking.
Hypothesized model # 3 (combined). There are paths (a) from the
autonomous and the dominant subselves to the dominant submotives of the
motive to achieve success to creative thinking and (b) from the autonomous
subself to the autonomous submotive of the motive to avoid failure to creat-
ive thinking.
Subselves, submotives, and creativity / 141


Participants were 154 undergraduates at Perm Teacher Training Uni-

versity. The sample consisted of 72 men and 82 women. Age ranged from 16
to 22, M = 18.30, SD = .52. After outliers on creative thinking variables bey-
ond 2 SD were eliminated, 142 participants’ raw data were taken into consid-


Self-reports. Each participant got measures on her/his subselves and sub-

motives. Subself subscales (the autonomous, the dominant, the empathic, and
the submissive) were measured by the Perm Plural Self Questionnaire. Each
subscale included 8 items. Responses were given on a 6-point scale. The ex-
treme values on the subscales were Strongly Disagree (1) and Strongly Agree
(6). The questionnaire was validated on its construct validity and external
validity (Dorfman et al., 2000).
The submotives of the motive to achieve success and the motive to avoid
failure were measured by the Perm Plural Achievement Motives Question-
naire. The autonomous, the dominant, the empathic, and the submissive sub-
motive subscales were specified either for the motive to achieve success or
the motive to avoid failure. Each subscale included 6 items. Responses were
given on a 6-point scale. The extreme values on the subscales were Strongly
Disagree (1) and Strongly Agree (6). The questionnaire was validated on its
construct validity and external validity (Fenogentova, 2002).
Paper-and-pencil tasks. Creative thinking was assessed by the Alternate
Uses Test (Wallach & Kogan, 1965) which asks subjects to think of as many
unique uses for common objects as possible other than their typical uses.
Three objects (brick, newspaper, and pencil) were used. Fluency, flexibility,
and originality measures were computed.
Methods of Analysis. Structural equation modeling using SEPATH in Stat-
istica (Steiger, 1995) was used to test path models. Analysis of the proposed
models followed the two-step procedure suggested by Anderson and Gerbing
(1988). The first step is to use confirmatory factor analysis to develop a
measurement model with an acceptable fit to the data. The second step is the
structural models to be tested.
After a measurement model with an acceptable fit to the data was met, a
series of nonnested structural models was used systematically to test hypo-
thesized models. The first model was a null model, M n. There were no pos-
142 / Aesthetics and Innovation
ited relations of the constructs to one another. The second model, M 1, tested
hypothesized paths from the autonomous subself to the autonomous sub-
motive of the motive to avoid failure to creative thinking. The third model,
M2, tested hypothesized paths from the autonomous and the dominant sub-
selves to the dominant submotive of the motive to achieve success to creative
thinking. The fourth model, M3, tested hypothesized paths (a) from the
autonomous and the dominant subselves to the dominant submotives of the
motive to achieve success to creative thinking, and (b) from the autonomous
subself to the autonomous submotive of the motive to avoid failure to creat-
ive thinking. In doing so, the autonomous and the dominant subselves were
included in the model as correlated. Finally, the best well-fitting and plaus-
ible model was assessed.
We also provided structural equation modeling to estimate properly the
mediational effects. The mediator variable M explains the relation between
the exogenous variable X and the outcome variable Y. As such, M is endo-
genous relative to X, but exogenous relative to Y. For our models, submotives
served as mediator variables, subselves were exogenous variables, and creat-
ive thinking was the endogenous variable. We followed the mediation
strategy for structural equation modeling suggested by Holmbeck (1997) and
supported by Frazier et al. (2004). It tested the fit of the X—Y model and the
fit of the X—M—Y model, as well as the X—M and M—Y paths. A medi-
ational model was supported if the model with the direct path between X and
Y does not provide a better fit to the data (i.e., the direct path between X and
Y is not significant) than paths in the X—M—Y model.
Initial parameters were as follows. Based on exploratory factor analysis
each of the autonomous and the dominant exogenous factors included the
three indicators with highest loadings extracted from the appropriate subscale
items. Likewise, based on exploratory factor analysis each of the dominant
submotives of the motive to achieve success and the autonomous submotive
of the motive to avoid failure as mediating factors included the three indicat-
ors with the highest loadings extracted from the appropriate subscale items.
The creative thinking endogenous factor included three indicators. They were
fluency, flexibility, and originality.
For each model the method of discrepancy function estimation used was
Generalized Least Squares. The line search method used was Cubic Interpol-
ation. Five indices were used to assess the model fit. First, the chi-square stat-
istic was used. Typically, one seeks a nonsignificant chi-square to indicate no
significant difference between the hypothesized model and the measured
variables. However, the chi-square statistic is affected by sample size such
that it can be significant for models that fit the data relatively well (Bentler,
Subselves, submotives, and creativity / 143
1990). Second, the chi-square /df ratio was used. It adjusts the chi-square test
to control for sample size, with values exceeding 2.00 suggesting poorer fit-
ting models (Byrne, 1989). Third, Steiger and Lind’s root mean square error
of approximation (RMSEA) was used. Greater values indicate a poorer fit.
Values of .05 or less indicate a good fit, values of .06–.08 a reasonable fit,
values close to .10 a poor fit (Steiger, 1990). Fourth, the goodness-of-fit in-
dex (GFI) was used. The GFI can range from 0.00 to 1.00, with values closer
to 1.00 indicating better fitting models. Values of .95 or greater indicate that
the model adequately fits the data. Fifth, the adjusted goodness-of-fit index
(AGFI) was used. The AGFI can range from 0.00 to 1.00, with values closer
to 1.00 again indicating better fitting models. Values of .95 or greater indicate
that the model adequately fits the data. Finally, the chi-square difference test
was used to compare models.

Descriptive statistics

The means, standard deviations, coefficient alphas and intercorrelations

among the study variables are shown in Table 1.
Table 8-1. Descriptive statistics, measurement reliabilities and correlations among measures
144 / Aesthetics and Innovation

Note: ** p < .01; *** p < .001

Subselves, submotives, and creativity / 145
The average Alternate Uses Test score (creative thinking) was 31.45
(SD = 10.24), averaged across all objects (i.e., brick, newspaper, and pencil)
and fluency, flexibility, and originality measures. It was quite high compared,
for instance, with the average (M = 28.51) and SD (7.84) obtained by Mar-
tindale (2002). In part this may be because participants were given five
minutes, as opposed to the three minutes in Martindale’s study, to complete
the task.
The value of the autonomous subself variable indicated a quite high mag-
nitude (M = 4.67, SD = .74). The value of the dominant subself variable was
lower (M = 4.31, SD = .83) as compared to the autonomous subself, t(141) =
4.59, p < .001. The value of the dominant submotive of the motive to achieve
success indicated a quite high magnitude (M = 4.50, SD = .78). The value of
the autonomous submotive of the motive to avoid failure was lower (M =
3.48, SD = .88) as compared to the dominant submotive of the motive to
achieve success, t(141) = 6.78, p < .001.
Creative thinking correlated positively with the dominant submotive of
the motive to achieve success, r(141) = .21, p < .01) and negatively with the
autonomous submotive of the motive to avoid failure, r(141) = -.19, p < .01.
Creative thinking did not correlate significantly with the autonomous subself
or the dominant subself.

Measurement model

Fit indexes of the measurement model (confirmatory factor analysis) are

provided in Table 2. The results displayed in Table 2 show that the measure-
ment model fit the data quite well: χ2 (90) = 132.10, p < .001, χ2/df = 1.47,
RMSEA = .06, GFI = 1.00, AGFI = 1.00.

Table 8-2. Combined confirmatory factor analysis of the Measurement model and in-
terfactors correlations

Latent factor and indicators Parameter SE t p


Creative thinking
1 (Fluency) .99 .01 172.61 .001
2 (Flexibility) .94 .01 80.23 .001
3 (Originality) .96 .01 108.99 .001
146 / Aesthetics and Innovation
Table 8-2. Continuation
Autonomous subself
1 .80 .04 17.84 .001
2 .79 .04 17.57 .001
3 .89 .04 23.95 .001
Dominant subself
1 .90 .04 25.02 .001
2 .78 .04 17.50 .001
3 .77 .05 16.54 .001
Dominant submotive of the
motive to achieve success
1 .84 .04 19.32 .001
2 .70 .05 12.68 .001
3 .79 .05 16.56 .001
Autonomous submotive of
the motive to avoid failure
1 .70 .08 8.80 .001
2 .64 .08 7.65 .001
3 .50 .09 5.46 .001

Interfactors correlations

1 2 3 4 5

Creative thinking –
Autonomous subself .02 –
Dominant subself .08 .21* –
Dominant submotive of the .33*** .39*** .47*** –
motive to achieve success
Autonomous submotive of -.31*** -.42*** -.03 -.52*** –
the motive to avoid failure

Note: * p < .05, *** p < .001

Subselves, submotives, and creativity / 147
Each latent factor included all of its indicators at level p < .001. When in-
terfactor correlations were tested data were found similar to those obtained
by correlations among measures (compare the interfactor correlations shown
in Table 2 with the correlations among measures shown in Table 1).

Structural models

Fit indices of the structural models are shown in Table 3. Each of the four
models fit the data quite well. Comparing the second model, M1, with the null
model, Mn, the former model significantly fit the data better than the later:
Δχ2 (4, N = 142) = 2.84, p < .001. The third model, M2, significantly fit the
data better than the second model, M1: Δχ2 (1, N = 142) = 18.42, p < .001.
The fourth model, M3, fit the data better than the third model, M2: Δχ2 (2, N =
142) = 10.23, p < .001. The fit of the fourth model, M3, was significantly bet-
ter than those of the preceding models. The fit of this model was χ2 (82) =
100.12, p > .05, χ2/df = 1.22, RMSEA = .04, GFI = 1.00, AGFI = 1.00. It is
the preferred model because it is more parsimonious.
To estimate mediational effects, the model with direct paths between the
autonomous and the dominant subselves and creative thinking was evaluated.
This model fit the data as follows, χ2 (84) = 101.44, p > .05, χ2/df = 1.21, RM-
SEA = .04, GFI = 1.00, AGFI = 1.00. However this model did not provide a
better fit to the data than the fourth model: Δχ2 (2, N = 142) = 1.32, p > .05.
Moreover, the direct paths between the autonomous and the dominant sub-
selves and creative thinking were not significant. This is to say that the medi-
ational effects of the fourth model can be accepted.
The final structural mediational model with path coefficients appears in
Figure 1.
Parameter estimates revealed statistically significant positive path coeffi-
cients from the dominant subself to the dominant submotive of the motive to
achieve success (.48, p < .001) to creative thinking (.37, p < .05). There were
statistically significant positive path coefficients from the autonomous sub-
self to the dominant submotive of the motive to achieve success (.39, p <
.001) to creative thinking (.37, p < .05) as well. There were also negative path
coefficients from the autonomous subself to the autonomous submotive of the
Table 8-3. Fit indices for measurement and nonnested structural models
148 / Aesthetics and Innovation

Note: N = 142. RMSEA—root-mean-square error of approximation; GFI—comparative fit index; AGFI—adjusted comparative fit index. M0—Null
model, M1—Paths from the autonomous subself through the autonomous submotive of the motive to avoid failure to creative thinking, M2—Paths from
the autonomous and the dominant subselves to the dominant submotive of the motive to achieve success to creative thinking, M3—Paths (a) from the
autonomous and the dominant subselves to the dominant submotives of the motive to achieve success to creative thinking, (b) from the autonomous
subself to the autonomous submotive of the motive to avoid failure to creative thinking.
*** p < .001.
Subselves, submotives, and creativity / 149
motive to avoid failure (-.56, p < .001) to creative thinking (-.31, p < .07).
Besides, the dominant subself and the autonomous subself were positively
correlated (.19, p < .07).

Figure 8-1. Path diagrams for submotives of the achievement motive’s mediation of
subselves on creative thinking. Full lines indicate positive paths and dashed lines neg-
ative paths. Indicators are omitted.

Note: 1—the dominant subself, 2—the autonomous subself, 3—the dominant sub-
motive of the motive to achieve success, 4—the autonomous submotive of the motive
to avoid failure, 5—Creative thinking.* p < .10, ** p < .05, *** p < .001

Comparing the hypothesized models, it was found that the combined hy-
pothesized model # 3 (numbered in the Method and Results sections as the
fourth model) better fit the data than did the other hypothesized models. This
model is to be preferred because it is more parsimonious. The mediational ef-
fects of this model can be accepted as well. There were significant or very
close to significant paths with positive coefficients from the autonomous and
the dominant subselves to the dominant submotives of the motive to achieve
success to creative thinking and paths with negative coefficients from the
autonomous subself to the autonomous submotive of the motive to avoid fail-
ure to creative thinking.
150 / Aesthetics and Innovation
The data indicate that the autonomous and the dominant subselves do not
have direct access to creative thinking but they enable creative thinking
through the dominant submotives of the motive to achieve success. Besides,
the autonomous subself is related to creative thinking through the autonom-
ous submotive of the motive to avoid failure. The autonomous subself can en-
able creative thinking if the autonomous submotive of the motive to avoid
failure is decreased.
How may we explain the data obtained? Atkinson (1957) considered fear
of failure as a motive that dampens the tendency to succeed. However, the
motives to achieve success and avoid failure can be seen as too broad. Dis-
crimination between the dominant submotive (the motive to achieve success)
and the autonomous submotive (the motive to avoid failure) may result in a
better understanding of achievement motivation. Herein, significant negative
correlations between these submotives were obtained (see Table 1).
Mediational effects of submotives can be captured by referring to
Skaalvik’s (1997) suggestion that ego orientation falls into self-defeating and
self-enhancing dimensions. A self-defeating ego orientation is an attempt to
protect self-esteem and is associated with high anxiety and negatively related
to achievement and self-perceptions. Self-enhancing ego orientation is posit-
ively related to achievement, self-perception, and intrinsic motivation. Aca-
demic achievement, self-concept, and self-efficacy are positively related to a
self-enhancing ego orientation and negatively related to a self-defeating ego
orientation. Based on Skaalvik’s (1997) approach and findings we can guess
that the autonomous submotive of the motive to avoid failure expresses a
self-defeating orientation toward the autonomous subself, on the one hand,
and constrains creative thinking on the other. It is also plausible that the dom-
inant submotive of the motive to achieve success expresses a self-enhance-
ment of the autonomous and the dominant subselves, on the one hand, and in-
creases creative thinking on the other.
Of particular importance with respect to the data obtained is our sugges-
tion of parallel paths from subselves to submotives to creative thinking. The
autonomous and the dominant subselves may be conceived of as alternatives
in concert with their submotives as mediators. They constitute parallel ways
of thinking of diverse useful alternatives (creative thinking).
From the socioindividual model of creativity (Dorfman, 2005) perspect-
ive, of particular relevance is the autonomous subself and the autonomous
submotive of the motive to avoid failure, which can be referred to the
autonomy region of the socioindividual realm, and the dominant subself and
the dominant submotive of the motive to achieve success, which are related
to the dominance region. One can see that these regions are major latent con-
Subselves, submotives, and creativity / 151
structs. They can be discriminated by their manifest variables, subself-repres-
entations, and appropriate submotives of achievement motives. We discover
new exogenous and mediational factors linking with creativity as an endo-
genous factor. Given this, we observe intraregional effects within the domin-
ance region when the dominant subself, through the dominant submotive of
the motive to achieve success, relates to creative thinking. Likewise, the in-
traregional effects take place within the autonomy region when the autonom-
ous subself, through the autonomous submotive of the motive to avoid fail-
ure, relates to creative thinking. We observe interregional effects between the
Autonomy and dominance regions when the autonomous subself through the
dominant submotive of the motive to achieve success relates to creative
thinking. This claim would be important in testing new hypotheses about cre-
ativity based on other variables manifesting and mediating the autonomy and
dominance regions of the socioindividual realm, as well as their interrela-

Amabile, T. M. (1996). Creativity in context: Update to the social psychology of cre-
ativity. Boulder, CO: Westview.
Amabile, T. M., Hennessey, B. A., & Grossman, B. S. (1986). Social influences on
creativity: The effects of contracted-for reward. Journal of Personality and Social
Psychology, 50, 14-23.
Amthauer, R. (1973). Intelligenz-Struktur-Test L S. T. 70. Handanweisung fur die
Durchfahrung und Auswertung. Gőttingen: Verlag fũr Psychologie.
Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal social-cog-
nitive theory. Psychological Review, 109, 619–645.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103, 411–
Atkinson, J. (1957). Motivational determinants of risk-taking behavior. Psychological
Review, 64, 359–372.
Bandura, A. (1989). Human agency and social cognitive theory. American Psycholo-
gist, 55, 1175–1184.
Bandura, A. (1991). Social cognitive theory of self-regulation. Organizational Beha-
vior and Human Decision Processes, 50, 248–287.
Bandura, A. (1999). Social cognitive theory of personality. In L. A. Pervin and O. P. John
(Eds.), Handbook of personality: Theory and research (pp. 154–196). 2nd ed. New
York: Guilford Press.
Bandura, A., & Wood, R. E. (1989). Effect of perceived controllability and perform-
ance standards on self-regulation of complex decision making. Journal of Person-
ality and Social Psychology, 56, 805–814.
152 / Aesthetics and Innovation
Barron, F. (1963). The needs for order and for disorder as motivation in creative activ-
ity. In C. W. Taylor & F. Barron (Eds.), Scientific creativity: Its recognition and
development. New York: Wiley.
Barron, F. (1969). Creative person and creative process. New York: Holt, Rinehart &
Barron, F. & Harrington, D. M. (1981). Creativity, intelligence, and personality. An-
nual Review of Psychology, 32, 439−476.
Baer, J. (1998). Gender differences in the effects of extrinsic motivation on creativity.
Journal of Creative Behavior, 32, 18–37.
Bazhin, E. F., Golinkina, E. A., & Etkind, A. M. (1984). The method to examine the
level of subjective control. Psykhologicheskyi Zhurnal, 5, 152–162.
Beard, G. M. (1874). Legal responsibility in old age. New York: Russell Sage Found-
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological
Bulletin, 107, 238–246.
Byrne, B. (1989). A primer of LISREL: Basic applications and programming for con-
firmatory factor analytic models. New York: Springer-Verlag.
Brunstein, J. C., & Maier, G. W. (2005). Implicit and self-attributed motives to
achieve: Two separate but interacting needs. Journal of Personality and Social
Psychology, 89, 205–222.
Cassidy, T., & Lynn, R. (1989). A multifactorial approach to achievement motivation:
The development of a comprehensive measure. Journal of Occupational Psycho-
logy, 62, 301–312.
Cattell, R. В., & Drevdahl, J. E. (1955). A comparison of the personality profile
(16 PF) of eminent researchers with that of eminent teachers and adminis-
trators, and of the general population. British Journal of Psychology, 46,
Cloninger, C. R. (1987). A systematic method for clinical description and classifica-
tion of personality variants. Archives of General Psychiatry, 44, 573-588.
Cropley, A. J. (1999). Definitions of creativity. In M. A. Runco and S. R. Pritzker
(Eds.), Encyclopedia of creativity, 1, 511–524. San Diego et al.: Academic Press.
Csikszentmihalyi, M. (1997). Creativity. Flow and the Psychology of Discovery and
Invention. 51–76. New York: Harper.
Csikszentmihalyi, M. (1999). Implications of a systems perspective for the study of
creativity. In R. J. Sternberg (Ed.), Handbook of creativity (pp. 313–335). Cam-
bridge: Cambridge University Press.
Cupchik, G. C. (1999). The thinking-I and the being-I in psychology of the arts. Cre-
ativity Research Journal, 12, 165–173.
Deci, E. L., & Ryan, R. M. (1985). The General causality orientations scale: Self-de-
termination in personality. Journal of Research in Personality, 19, 109–134.
Dellas, M., & Gaier, E. L. (1970). Identification of creativity: The individual. Psy-
chological Bulletin, 73, 55–73.
Dergacheva, O., Dorfman, L., & Leontiev, D. (2002). Measurement of autonomy in
Russian culture. 8th International Conference on Motivation (Workshop on
Achievement and Task Motivation). Abstracts (p. 27). Moscow.
Subselves, submotives, and creativity / 153
Dorfman, L. Ya. (1993). Metaindividual world: Methodological and theoretical is-
sues. Moscow: Smysl. (in Russian, English Summary).
Dorfman, L. (1995). Metaindividual world. Studia Psychologica, 37, 4, 279–286.
Dorfman, L. Ya. (1997). Emotion in art: Theoretical and empirical considerations.
Moscow: Smysl. (in Russian, English Summary).
Dorfman, L. Ya. (2004). Self–conception: Differentiation and integration. In L. Ya.
Dorfman (Ed.), Integral individuality, self–conception, and personality (pp. 96–
123). Moscow: Smysl. (in Russian).
Dorfman, L. (2005). A metaindividual model of creativity. In P. Locher, C. Mar-
tindale, & L. Dorfman (Eds.), New directions in aesthetics, creativity, and
the arts (pp. 105–122). Amityville, NY: Baywood.
Dorfman, L. Ya. (2006). Relationships between subselves and regions of metaindi-
vidual world. In B. A. Viatkin (Ed.), XXI Merlinian readings. Perm: Perm State
Teacher Training University. (in press). (in Russian).
Dorfman, L., Fenogentova, O., & Dudorova, C. (2002). The plural self and the plural
achievement motives. 8th International Conference on Motivation (Workshop on
Achievement and Task Motivation). Abstracts (p. 28–29). Moscow.
Dorfman, L. Ya, Fenogentova, O. P., & Dziuba (2000). The plural self and locus of
control. In B. A. Viatkin & M. R. Schukin (Eds.), XV Merlinian readings (pp. 72–
73). Perm: Perm State Teacher Training University. (in Russian).
Dorfman, L. Ya., & Gasimova, V. A. (2004). Metaindividual world and creative think-
ing: Intelligence as a mediator. In L. Ya. Dorfman, Eu. A. Malianov, & E. M.
Berezina (Eds.), Metaindividual world and the plural self: Creativity, the arts,
and ethnos (pp. 40–51). Perm: Perm State Institute of Art and Culture. (in Russi-
Dorfman, L. Ya., & Liakhova, N. A. (2004). Regions of the metaindividual world and
creative thinking. In L. Ya. Dorfman, Eu. A. Malianov, & E. M. Berezina (Eds.),
Metaindividual world and the plural self: Creativity, the arts, and ethnos (pp. 9–
17). Perm: Perm State Institute of Art and Culture. (in Russian).
Dorfman, L. Ya., & Ogorodnikova, A. V. (2004). Regions of the metaindividual world
and personality creative potential. In L. Ya. Dorfman, Eu. A. Malianov, & E. M.
Berezina (Eds.), Metaindividual world and the plural self: Creativity, the arts,
and ethnos (pp. 18–29). Perm: Perm State Institute of Art and Culture. (in Russi-
Dorfman, L. Ya., Riabikova, M. V., Goldberg, I. M., & Vedrov, A. A. (2000). A new
version of Perm Plural Self Questionnaire. In Eu. Malianov, L. A. Shipitsina, L.
Ya. Dorfman, S. I. Kornienko, & C. Martindale (Eds.), Creativity in education,
culture, and art (pp. 179–183). Perm: Perm State Institute of Art and Culture. (in
Drevdahl, J. E., & Cattell, R. B. (1958). Personality and creativity in artists and
writers. Journal of Clinical Psychology, 14, 107–111.
Duda, J. L. (1993). Goals: A social-cognitive approach to the study of achievement
motivation in sport. In R. N. Singer, M. Murchey, & L. Keith Tennant (Eds.),
Handbook of research on sport psychology (pp. 421–436). New York: Macmillan.
154 / Aesthetics and Innovation
Dudorova, E. V. (2004). Plural achievement motives and plural self as factors of aca-
demic performance. In L. Ya. Dorfman, Eu. A. Malianov, & E. M. Berezina
(Eds.), Metaindividual world and plural self: Creativity, art, and ethnos (pp. 95–
110). Perm: Perm State Institute of Art and Culture. (in Russian).
Dutton, K. A., & Brown, J. D. (1997). Global self-esteem and specific self-views as
determinants of people's reactions to success and failure. Journal of Personality
and Social Psychology, 73, 139–148.
Dweck, C. S. (1986). Motivational processes affecting learning. American Psycholo-
gist, 41, 1040–1048.
Eisenberger, R., Armeli, S., & Pretz, J. (1998). Can the promise of reward increase
creativity? Journal of Personality and Social Psychology, 74, 704–714.
Erez, A., & Judge, T. A. (2001). Relationship of core self-evaluations to goal setting,
motivation, and performance. Journal of Applied Psychology, 86, 1270–1279.
Eysenck, H. J. (1970). The structure of human personality. London, G.B.: Methuen.
Eysenck, H. J. (1995). Genius: The natural history of creativity. Cambridge: Cam-
bridge University Press.
Eysenck, H. J., & Eysenck, S. B. G. (1994). Manual of the Eysenck Personality Ques-
tionnaire (EPQ-R Adult) comprising the EPQ-Revised (EPQ-R) & EPQ-R Short
Scale. San Diego, Ca: EdITS.
Feist, G. J. (1999). Autonomy and independence. In M. A. Runco and S. R. Pritzker
(Eds.), Encyclopedia of creativity (Vol. 1, pp. 157–163). San Diego: Academic
Fenogentova, O. P. (2002). Questionnaires of plural achievement motives to achieve
success and avoid failure: Theoretical consideration and preliminary psychometric
results. In E. M. Berezina (Ed.), 85 years of higher professional education in Ur-
al’s region (pp. 126–130). Perm: Perm State Institute of Art and Culture. (in Rus-
Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing moderator and mediator ef-
fects in counseling psychology research. Journal of Counseling Psychology, 51,
Friedman, R. S., & Förster, J. (2005). Effects of motivational cues on perceptual
asymmetry. Implications for creativity and analytical problem solving. Journal of
Personality and Social Psychology, 88, 263–275.
Furby, L. (1991). Understanding the psychology of possession and ownership: A per-
sonal memoir and an appraisal of our progress. Journal of Social Behavior and
Personality, 6, 457–463.
Gaertner, L., Sedikides, C., & Graetz, K. (1999). In search of self-definition. Motiva-
tional primacy of the individual self, motivational primacy of the collective self,
or contextual primacy? Journal of Personality and Social Psychology, 76, 5–18.
Gaertner, L., Sedikides, C., Vevea, J. L., & Iuzzini, J. (2002). The "I," the "we," and
the "when": A meta-analysis of motivational primacy in self-definition. Journal of
Personality and Social Psychology. 83, 574–591.
Galton, F. (1883). Inquiries into human faculty and its development. London: Mac-
Subselves, submotives, and creativity / 155
Gough, H. G., & Heilbrun, A. B. Jr. (1983). The adjective check list (ACL): Manual.
Palo Alto, CA: Consulting Psychologists Press.
Gruber, H. E. (1999). Evolving systems approach. In M. A. Runco and S. R. Pritzker
(Eds.), Encyclopedia of creativity (Vol. 1, pp. 689–693). San Diego et al.: Aca-
demic Press.
Guilford, J. P. (1950). Creativity. American Psychologist, 5, 444–454.
Guilford, J. P. (1967). The nature of human intelligence. New York: McGraw-Hill.
Harter, S. (1981). A new self-report scale of intrinsic versus extrinsic orientation in the
classroom: Motivational and informational components. Developmental Psycho-
logy, 17, 300–312.
Heckhausen, H. (1982). The development of achievement motivation. In W. W. Har-
tup (Ed.), Review of child development research (Vol. 6, pp. 600–668). Chicago:
University of Chicago Press.
Hennessey, B. A. (1998). Reward, intrinsic motivation, and creativity. American Psy-
chologist, 53, 674–675.
Hennessey, B. A. & Amabile, T. M. (1988). The conditions of creativity. In R. J.
Sternberg (Ed.), The nature of creativity (pp. 11–38). New York: Cambridge Uni-
versity Press.
Hermans, H. J. M. (1996). Voicing the self: From information processing to dialogical
interchange. Psychological Bulletin, 119, 31–50.
Helson, R., Roberts, B., & Agronick, G. (1995). Enduringness and change in creative
personality and the prediction of occupational creativity. Journal of Personality
and Social Psychology, 69, 1173–1183.
Hodgins, H. S., Koestner, R., & Duncan, N. (1996). On the compatibility of autonomy
and relatedness. Journal of Personality and Social Psychology Bulletin, 22, 227–
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in
the study of mediators and moderators: Examples from the child-clinical and pedi-
atric psychology literatures. Journal of Consulting and Clinical Psychology, 65,
House, R. J., Shane, S. A., & Herold, D. M. (1996). Rumors of the death of disposi-
tional research are vastly exaggerated. Academy of Management Review, 21, 203–
James, K., Chen, J., & Goldberg, C. J. (1992). Organizational conflict and individual
creativity. Journal of Applied Social Psychology, 22, 545–566.
James, K. & Eisenberg, J. (2004). Personal identity and group identity influences on
algorithmic and original task performance. Creativity Research Journal, 16, 91–
James, W. (1890). Principles of psychology. New York: Holt, Rinehart & Winston.
Johnson, B. R. (1990). Toward a multidimensional model of entrepreneurship: The
case of achievement motivation and the entrepreneur. Entrepreneurship: Theory
and Practice, 14, 39–54.
Kaufman, J. C. (2002). Creativity and confidence. Price of achievement? American
Psychologist, 57, 375–376.
156 / Aesthetics and Innovation
Kihlstrom, J. F., Marchese-Foster, L. A., & Klein, S. B. (1997). Situating the self in
interpersonal space. In U. Neisser & D. A. Jopling (Eds.), The conceptual self in
context. Culture, experience, self-understanding (pp. 154−175). Cambridge: Cam-
bridge University Press.
Kovaleva, G. V. (2000). Creative thinking and its effects produced on the plural self.
In Eu. Malianov, L. A. Shipitsina, L. Ya. Dorfman, S. I. Kornienko, & C. Mar-
tindale (Eds.), Creativity in education, culture, and art (pp. 133–136). Perm: Perm
State Institute of Art and Culture. (in Russian).
Laforge, R., & Suczek, R. F. (1955). The interpersonal dimension of personality: III.
An interpersonal check list. Journal of Personality, 25, 94–112.
Leary, M. R., & Tangney, J. P. (2003). The self as an organizing construct in the beha-
vioral and social sciences. In M. R. Leary and J. P. Tangney (Eds.), Handbook of
self and identity (pp. 3–14). New York: Guilford Press
Leary, T. (1956). Multilevel measurement of interpersonal behavior. Berkeley, CA:
Psychological Consultation Service.
Leary, T. (1957). Interpersonal diagnosis of personality. New York: Ronald Press.
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance.
Engelwood Cliffs, NJ: Prentice Hall.
MacKinnon, D. W. (1960). The highly effective individual. Teachers College Record,
61, 376–378.
Magnusson, D. (2001). The holistic–interactionistic paradigm: Some directions for
empirical developmental research. European Psychologist, 6, 153–162.
Martindale, C. (1980). Subselves. The internal representation of situational and per-
sonal dispositions. In L. Wheeler (Ed.), Review of personality and social psycho-
logy (Vol. 1, pp. 193–218). Beverly Hills: Sage.
Martindale, C. (1989). Personality, situation, and creativity. In J. A. Glover, R. R.
Ronning, & C. R. Reynolds (Eds.), Handbook of creativity (pp. 211−232). New
York: Plenum.
Martindale, C. (2001). Oscillations and analogies: Thomas Young, MD, FRS, geni-
us. American Psychologist, 56, 342–345.
Martindale, C. (2002). Creativity, attention, and cognitive disinhibition. In R. Tomas-
soni (Ed.), La psicologia delle arti oggi (pp. 89–99). Milano: Angeli.
Martindale, C., & Dailey, A. (1996). Creativity, primary process cognition and person-
ality. Personality and Individual Differences, 20, 409–414.
Mayer, J. D. (1998). A systems framework for the field of personality. Psychological
Inquiry, 9, 118–144.
McClelland, D. C., Atkinson, J. W., Clark, R. A., & Lowell, E. L. (1953). The
achievement motive. New York: Appleton-Century-Crofts.
McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and
implicit motives differ? Psychological Review, 96, 690–702.
McCrae, R. R. (1987). Creativity, divergent thinking, and openness to experience.
Journal of Personality and Social Psychology, 52, 1258–1265.
McDougall, W. (1926). An introduction to social psychology. Boston: John W. Luce.
Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press.
Mehrabian, A. (1969). Measures of achieving tendency. Educational and Psycholo-
Subselves, submotives, and creativity / 157
gical Measurement, 29, 445−451.
Mehrabian, A. (1994/95). Individual differences in achieving tendency: Review of
evidence bearing on a questionnaire measure. Current Psychology, 13, 351–364.
Merlin, V. S. (1986). Essay on the integral study of the individuality. Moscow: Ped-
agogika. (in Russian).
Mischel, W., & Morf, C. C. (2003). The self as a psycho-social dynamic processing
system: A meta-perspective on a century of the self in psychology. In M. R. Leary
and J. P. Tangney (Eds.), Handbook of self and identity (pp. 15–43). New York:
Guilford Press.
Mumford, M. D., & Gustafson, S. B. (1988). Creativity syndrome. Integration, applic-
ation, and innovation. Psychological Bulletin, 103, 27–43.
Nicholls, J. G. (1984). Achievement motivation: Conceptions of ability, subjective ex-
perience, task choice, and performance. Psychological Review, 91, 328–346.
Nicholls, J. G. (1989). The competitive ethos and democratic education. Cambridge,
MA: Harvard University Press.
Ogorodnikova, A. V. (2006). Plural self, plural achievement motives, and creative
thinking: A correlational analysis. In B. A. Viatkin (Ed.), XXI Merlinian readings.
Perm: Perm State Teacher Training University. (in press). (in Russian).
Osche, R. (1990). Before the gates of excellence: The determinants of creative genius.
New York: Cambridge University Press.
Pervin, L. A. (2001). A dynamic systems approach to personality. European Psycholo-
gist, 6, 172–176.
Phillips, J. M., & Gully, S. M. (1997). Role of goal orientation, ability, need for
achievement, and locus of control in the self-efficacy and goal-setting process.
Journal of Applied Psychology, 82, 792–802.
Piaget, J. (1975/1985). The equilibration of cognitive structures. Chicago: University
of Chicago Press.
Pierce, J. L., Kostova, T., & Dirks, K. T. (2003). The state of psychological owner-
ship: Integrating and extending a century of research. Review of General Psycho-
logy, 7, 84–107.
Pyszczynski, T., Greenberg, J., Solomon, S., Arndt, J., & Schimel, J. (2004). Why do
people need self-esteem? A theoretical and empirical review. Psychological Bul-
letin, 130, 435–468.
Rathunde, K. (1999). Systems approach. In M. A. Runco and S. R. Pritzker (Eds.),
Encyclopedia of creativity (Vol. 2, pp. 605–609). San Diego: Academic Press.
Reiss, S. (2004). Multifaceted nature of intrinsic motivation: The theory of 16 basic
desires. Review of General Psychology, 83, 179–193.
Rotter, J. B. (1990). Internal versus external control of reinforcement: A Case history
of a variable. American Psychologist, 45, 489–493.
Runco, M. A. (1991). Divergent thinking. Norwood, NJ: Ablex Publishing Co.
Runco, M. A. (1999). Self-actualization. In M. A. Runco and S. R. Pritzker (Eds.),
Encyclopedia of creativity (Vol. 2, pp. 533–536). San Diego: Academic Press.
Rusalov, V. M. (1997). A questionnaire of temperamental traits of the individuality
(OFDSI). Moscow: Russian Academy of Sciences Institute for Psychology (in
158 / Aesthetics and Innovation
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. American Psychologist,
55, 68–78.
Sansone, C., & Thoman, D. B. (2005). Interest as the missing motivator in self-regula-
tion. European Psychologist, 10, 175–186.
Schebetenko, S. (2005). Merging vs. detachment attitude of readers to folktale charac-
ters. In Eu. Malianov, C. Martindale, E. Berezina, L. Dorfman, D. Leontiev, V.
Petrov, & P. Locher (Eds.), Proceedings. International Congress on Aesthetics,
Creativity, and Psychology of the Arts (pp. 345–347). Moscow: Smysl.
Sedikides, C. (1993). Assessment, enhancement, and verification determinants of the
self-evaluation process. Journal of Personality and Social Psychology, 65, 317–
Sedikides, C., & Strube, M. J. (1997). Self-evaluation: To thine own self be good, to
thine own self be sure, to thine own self be true, and to thine own self be better. In
M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 29, pp. 209–
269). New York: Academic Press.
Skaalvik, E. M. (1997). Self-enhancing and self-defeating ego orientation. Relations
with task and avoidance orientation, achievement, self-perceptions, and anxiety.
Journal of Educational Psychology, 89, 71–81.
Snyder, C. R., & Higgins, R. L. (1997) Reality negotiation: Governing one's self and
being governed by others. Review of General Psychology, 1, 336–350.
Steiger, J. H. (1990). Structural model evaluation and modification: An interval estim-
ation approach. Multivariate Behavioral Research, 25, 173–180.
Steiger, J. H. (1995). Structural equation modeling in Statistica (Version 5) [Computer
software]. Tulsa, OK: Statsoft.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.
Szymanski, K., & Harkins, S. G. (1992). Self-evaluation and creativity. Personality
and Social Psychology Bulletin, 18, 259–265.
Tesser, A. (1988). Toward a self-evaluation maintenance model of social behavior.
In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp.
181–227). New York: Academic Press.
Thompson, T., Davidson, J. A., & Barber, J. (1995). Self-worth protection in achieve-
ment motivation performance effects and attributional behavior. Journal of Edu-
cational Psychology, 87, 598–610.
Vartanian, O., Martindale, C., & Kingery, L. (2002). Creativity and disinhibition. In
Eu. Malianov, N. Zakharov, E. Berezina, L. Dorfman, V. Petrov, and C. Mar-
tindale (Eds.), Personality, creativity, and art (pp. 204–215). Perm: Perm State In-
stitute of Art and Culture.
Wallach, M. A., & Kogan, N. (1965). Modes of thinking in young children: A study of
the creativity-intelligence distinction. New York: Holt, Rinehart, and Winston.
Ward, E. A. (1997). Multidimensionality of achievement motivation among employed
adults. Journal of Social Psychology, 137, 542–544.
Weinberger, J., & McClelland, D. C. (1990). Cognitive versus traditional motivational
models: Irreconcilable or complementary? In E. T. Higgins & R. M. Sorrentino
Subselves, submotives, and creativity / 159
(Eds.), Handbook of motivation and cognition (Vol. 2, pp. 562–597). New York:
Guilford Press.
Wiggins, J. S. (1995). IAS: Interpersonal adjective scales. Professional manual. PAR.
Wink, P. (1999). Self processes and creativity. In M. A. Runco and S. R. Pritzker
(Eds.), Encyclopedia of creativity (Vol. 2, pp. 537–541). San Diego: Academic
Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of arousal. Hills-
dale, NJ: Erlbaum.

Perversion and creativity in the language of war

Robert Hogenraad

We examine the degree of association between creativity and the risk of war
in political documents issued before or during international conflicts. The
documents are speeches, conversations, and diplomatic memos prepared or
recorded before or during WW I, WW II, the Cuban missile crisis, and the
Anglo-American intervention in Iraq. We analyze the texts using computer-
readable semantic filters. The filters touch on several indicators of creativity
and one indicator of the risk of war (using the motives of need for affiliation
and need for power). We evaluate creativity through thesauri that weigh the
levels of imagery, concreteness, meaningfulness, and regressive thought con-
tents in texts. We assess the risk of war in texts by measuring the gap
between contents of affiliation and contents of power. In the past, such a gap
consistently preceded the outbreak of wars. In the main, when the risk of war
increases, words for war reflect a management of simplification. When war is
in the making, political leaders communicate to the public with simple con-
crete words that everybody can understand. This strategy reflects a manage-
ment of simplification that brushes aside opportunities of change.

Under pressure toward novelty, car makers build new models of cars.
They know that marketing the same car years on end makes customers bored
and leads them to abandon the brand. For the same reason, novelists and po-
ets write literary works containing ever more novel images (Martindale,
1975). For the same reason too, scientists write scientific works ever more
abstract and complex (Hayes, 1992; Hogenraad, McKenzie, Morval, & Duch-
arme, 1995). This is because the rule of science involves grouping concrete
facts under a single abstract heading. Political leaders also produce words. It
is an important part of their job. How political leaders deal with pressure to-
ward novelty is low on the list of questions for most of us. Except in two

Perversion of words / 161
cases: When leaders engage in war with another nation and when they fix
conflicts with other nations. This is the question we address in this chapter.
Pressure toward novelty may not be the major constraint on political lan-
guage, but it is a constant one. How does pressure toward novelty shape the
language of political leaders? There are as many reasons to expect more im-
agery and concrete words in political language when war is in the making as
when a conflict is on the way to a solution. On the one hand, we can achieve
a compromise on concrete issues, not on matters of principles. This effect of
rhetoric (Perelman & Olbrechts-Tyteca, 1969) is enough to cause us to expect
more images and concrete words while negotiating a peace. On the other
hand, to find solutions to a conflict demands one to make up new ways of
looking at the political reality. Take the Cuban missile crisis of October 1962.
“Our principal problem is to try and imaginatively to think what the world
would be like if we do this, and what it will be like if we don’t” (McGeorge
Bundy, President’s Assistant, in May & Zelikow, 1997, p. 102). Finding a
peaceful solution to a conflict supposes managing complex cognitive mech-
anisms that are the contrary of the clichéd gale-force speeches of war (Gut-
tieri, Wallace, & Suedfeld, 1995). To that we can oppose Cattaneo’s (1963)
theory according to which new ideas come from conflicting rather from har-
monious relations:

“A quell'opera di nemico era necessario un altro intelletto. È perciò

che i grandi pensatori (…), i quali ruppero il circolo della tradizione
(…), si mostrano quasi sempre accinti con tutte le forze loro come ad
un'impresa di Guerra” (To behave as an enemy, another mind is ne-
cessary. This is why great thinkers, who broke with tradition, were al-
most always engaged as if in a war enterprise).

Dmitry Ushakov (2005) pointed out that French-speaking psychologists

like Perret-Clermont, Mugny, and Doise (1976) further developed Cattaneo’s
theory, exploring experimentally the socio-cognitive conflict as a motor of
cognitive development.
One difficulty is the influence of war on language. For Virginia Woolf,
World War I was responsible for the collapse of the old language of rational
control (Hussey, 1991). The military sound of words is present in several of
her novels, as in “Jacob’s Room” (Woolf, 2000). The French linguist Brice
Parain (1934; see also Merton, 1976) denounced the dizzying feeling of inex-
actness of language. Words are sick, he claimed, in which peace means ag-
gression, freedom, oppression, and socialism, a regime of social inequalities.
In the poem, “The White Man’s Burden”, Rudyard Kipling (1899) wrote of
162 / Aesthetics and Innovation
“savage wars of peace” after the US take over of the Philippines following
the Spanish-American war of 1898. It is as if a form of perversion, as a way
of pushing further the limits of what is possible, made it possible for political
elites to blur the differences between the meaning of words (Chasseguet-
Smirgel, 1996). And to alter reality. The poignant (and perhaps justified)
Armageddon style of the speeches of President G. W. Bush and Prime Minis-
ter Blair is a carry-over from biblical language to idealized secular matters:

“The ideal of America is the hope of all mankind… That hope still
lights the way. And the light shines in the darkness. And the darkness
will not overcome it”. (Accessed June 2, 2006

Transferring language from one realm (the Bible) to another (reality) res-
ults in a transfigured wonderfully shining reality from which death is ex-
cluded. Anyway, perversion of language is not one-sided:

“But our words have no impact upon you, therefore I’m going to talk
to you in a language that you understand. Our words are dead until
we give them life with our blood”. (From a videotape of British-born
Mohammad Sidique Khan, West Yorkshire, supposed ringleader of the
7/7 London bombing –Guardian Weekly, Sept 9-15 2005, p. 8, acessed
June 2, 2006:

Using the language of the Apocalypse to justify a war is another way to

respond to pressure toward novelty –by dubiously crossing the frontiers of
reality (Bacevich, 2005, especially chapter 5 “Onward” on Evangelicalism,
pp. 122-146). Orwell’s (1990) “Newspeak” in Nineteen eighty-four is a liter-
ary example of the same. Yet, if war eclipses the word, without words, polit-
ical leaders could not go to war, contrary to what character Raunce replies in
Henry Green’s novel, Loving: “It's what we're going to do whatever the name
you give to it” (2000, p. 197). In this sense, the argument cuts two ways,
which comes as something of a relief.
Sometimes, the lack of innovativeness in leaders hinges on passivity. To
an irresolute Kaiser expressing whims of changing military plans that had
been worked on for a year, General Moltke tersely replied on August 1st
1914 “Your Majesty, it cannot be done…once settled, it cannot be altered”
(Tuchman, 1994, p. 79). And during the Cuban missile crisis of October
1962, to an aide expressing his hope that “the boat does not capsize”, Chair-
Perversion of words / 163
man Nikita Khrushchev answered “Now it’s too late to change anything”
(May & Zelikow, 1997, p. 681). In both examples, there is an implicit ac-
knowledgment that some ideal moment for acting had long passed. In the
first example, the fluffed opportunity caused 15,000,000 deaths, of which
8,500,000 were military ones (http://users.erols.com/mwhite28/warstat1.htm
- WW1). In the second one, a restraint on force led finally to a satisfactory
Perhaps we are part of the many bored with inconclusive theories on the
way war and creativity interact. Political leaders can be innovative whether
using images or complex abstract thought contents. Circumstances decide.
Our own commitment is with predicting the outbreak of meat grinder con-
flicts. Predicting the outbreak of wars is possible using a dependable model
of motivation. Assessing creativity is also possible through various indicat-

Power and affiliation: A motivational model of war

McClelland's predictive model of war (1975) involves two needs, which

are the need for affiliation and the need for power. Intimacy, friendship, and
positive emotional contacts with a person, as well as liking and wanting to be
liked, define the need for affiliation. The will to power, to have an impact on
another person or to get or to keep control over people, forms the essential of
the need for power. These two criss-cross threads, power and affiliation, are
always difficult to patch up with each other. Niccolò Machiavelli experienced
this difficulty –in prison and under torture–, about which he wrote later: “A
prince is always compelled to injure those who have made him the new
ruler… As a result, you (…) cannot keep the friendship of those who have
put you there” (1999, p. 8). Wrote Auden (1989, p. 62): “We tend to deprive
of their faces any person whom we believe to be at the mercy of our will”,
which is not a bad description of the unstable balance between affiliation and
power. The use of one's own power to save others is often the link between an
“imperial motivation pattern” (that is, the gap created by high need for power
and low need for affiliation) and later wars (McClelland, 1975, pp. 314–359).
The wider the gap, the greater the risk of war. No doubt a naïve view, but
possible. McClelland’s motivation theory of war stops short of predicting the
moment beyond which war becomes unavoidable. That would be giving the
model more than its due.
The violent logic that leads from reformist fervor for social justice to war
has to do with the nature of spiritual perfection. Perfection does not exist. It
breeds hypocrisy, as urging easily slips into the imperative. The awareness of
164 / Aesthetics and Innovation
a spiritual perfection coupled with the impossibility of reaching it causes one
to indulge oneself with pretending to be perfect. Subject to that, one vents the
awareness of one’s flaws through more militant calls to flawlessness in oth-
ers, thus pressuring them to a near-Masonic severity. “Just think how many
crimes have been committed because their author could not bear to be
wrong!”, this from Camus in La chute (1956, p. 23). Even the outbreak of
World War I, the precipitant of which one often imputes to the murder of
Archduke Ferdinand on June 28, 1914, was about moral and religious ideas
(Strachan, 2001). Puritan Germany’s viewed its ‘Tugend’ (virtue) threatened
by British materialism and French atheistic rationalism. Another example of
compulsory salvation is in Graham Greene's (2001, p. 18) The Quiet Americ-
an: “He was determined to do good, not only to any individual person, but to
a country, a continent, a world”. Whatever is virtuous in universalism always
feeds a non-commendable form of imperialism (Bacevich, 2005, p. 75). Uni-
versalism replaces whatever was good in the past by something supposed to
be even better, blurring differences again. Many answer positively to the
question “Should democracy be universal?”. Yet, other creeds may work in
other times and places.

Assessing creativity

Literary and scientific writers, going at great lengths not to repeat them-
selves, share the same question “What can I do that has not yet been done?”.
To keep interest in their readers, writers use concrete words that bring out im-
ages (Cohen, 1966; Martindale, 1975). At the other side of the scale, the rule
of certain styles, as in science, involves grouping varied concrete facts under
a single abstract heading. This is because the rule in science requires it to be-
come ever more abstract. In psychological jargon, images and metaphors, we
call them primary process thought contents (Kris, 1952). Their opposite, that
represent degrees of mental formulas, such as law and order, abstract think-
ing, time references, and moral imperatives, we call secondary process
thought contents. Martindale (1990) set up the primary process and second-
ary process thought contents into a measuring instrument for content analys-
is, the regressive imagery dictionary. The difference between primary process
and secondary process thought contents is that between the sensate and no-
tional schemes of culture (Sorokin, 1985). In sensate schemes, reality is that
which is present to the sense organs, in notional schemes, one considers that
it is the inner meaning that gives value to the world. Sensate contents are
“found in the world” (love, sex, food, chaos, dream, flying, for example). No-
Perversion of words / 165
tional contents are “built into the world” (money, work, discipline, police,
time, justice, law).
How does regressive imagery with its variants apply to the language of
political leaders on the brink of war? There are limits to the use of abstract
notions, in political language as elsewhere. First, abstract words are more dif-
ficult to understand than concrete ones (Hayes, 1992). Secondly, there are in-
trinsic limits to the capacity for abstraction of human thought (Thorngate,
1990). For what little control political leaders have over laws laid down for
them, managing a peaceful solution to a conflict supposes, as we pointed out
earlier, complex cognitive mechanisms. We expect in this case expressions of
abstract contents, allowing political leaders to group varied concrete events
and facts under a single abstract heading, as if in a scientific language. From
leaders intent to go to war, we would not expect an abstract language. Words
of primary process thought contents (love, food, sex, body parts) are more
those of literature than of political language. Yet, in the 172 speeches of Pres-
ident G. W. Bush (Hogenraad, 2005), the total frequency of usage of primary
process versus secondary process thought contents is 12,037 words (and 770
different ones) versus 28,383 words (and 764 different ones). What counts
however is the ratio of the ones to the others over time. Why indeed draw on
abstract words difficult to understand while concrete incidents may serve to
justify a war. In Saperstein’s chaos theory of war (1995), a single event (the
murder of Archduke Ferdinand) accounts for the death of millions of people
in World War I.
We analyzed political documents using five semantic filters. The role of
the filters is to transform a psychological motive or content into a tool –we
dub it a dictionary– fit to isolate and evaluate the presence of that motive or
content in a document. One such filter is the Motive dictionary (Hogenraad,
2003, 2005) that allows us to set power words to solidarity words into a se-
mantic filtering and use the filter to estimate the risk of war from a document.
The other dictionaries are the Regressive imagery dictionary (Martindale,
1975, 1990) and the Dictionaries of concreteness, imagery, and meaningful-
ness (Paivio, Yuille, & Madigan, 1968). We use the Martindale and Paivio
dictionaries to evaluate creativity in documents. We then control if any in-
crease in the risk of war in political documents goes with a similar increase –
or decrease– in one of the indicators of creativity.
166 / Aesthetics and Innovation


1. The first corpus is about James W. Headlam’s (1915) “History of

Twelve Days”. Headlam’s “History” concerns the diplomatic history of the
fortnight between the Austro-Hungarian ultimatum to Serbia of July 24, 1914
and the British ultimatum to Germany of August 4, 1914. Headlam’s count-
down to World War I covers, in chronological order, the exchanges of diplo-
matic telegrams and reports of conversations and negotiations. For the
present analysis, Headlam’s History corpus covers chapters II to XIII.
Chapter I is an introduction to the previous relations between Serbia and the
Austro-Hungarian Empire (see Table 1 for details on this and the other docu-
2. The next corpus is a 421-page document called “Events leading up to
World War II. 1931—1944” (1944). It is a commented chronology set up by
the Legislative Reference Service of the Foreign Affairs Committee of the
House of Representatives. The document covers the period from 1931 to
1944; it concerns all the world events that led to World War II, and not only
the ones involving Germany.
3. Closer to us are two series of documents about the 2003
Anglo-American intervention in Iraq. The first
(http://www.whitehouse.gov/response/resources2.html) is about 55 speeches
made by President G. W. Bush between March 13, 2002 and March 17, 2003,
the day of the American entry into Iraq. Of the 172 speeches analyzed by Ho-
genraad (2005) over the period September 2001 - March 2003, March 13,
2002 is the point beyond which the risk of war increased regularly until the
day of the intervention. It is now public knowledge that President Bush was
already set on war when he met Prime Minister Blair on January 31, 2003
(Van Natta, 2006).
4. The other series continues with 72 speeches made by Prime Minister
Tony Blair, between September 11, 2001 and March 20, 2003
5. Then follow two documents about the solution of a conflict. One is the
tale (Walder, 1958), in French, of the negotiations that led to the “Paix de
Saint-Germain” settling the French “Wars of Religion” between Catholics
and Huguenots in 1570.
6. The other is about the taped conversations held by President Kennedy’s
Executive Committee (excomm) during the 1962 Cuban missile crisis
Perversion of words / 167
between October 16, 11:50 am, and October 29, 10:10 am (May & Zelikow,
168 / Aesthetics and Innovation
Table 9-1. The corpus
Perversion of words / 169

Equipped with proper dictionaries, content analysis packs an astonishing

amount of information by filtering out the redundancies of speech. Filtering
out allows us to keep only what one is looking for. In these dictionaries, any
word assigned to one category cannot be present in another one. We format-
ted the dictionaries to work within the Protan program of computer-aided
content analysis (Hogenraad, Daubies, Bestgen, & Mahau, 1995). A diction-
ary, in content analysis, is a list of words organized into categories, that is,
words with a role in a hierarchy. When one applies a dictionary to a text, one
looks for matches between a word in a dictionary and a word in a text. One
then shoves the text words into the categories, counts the number of word
matches in each category and takes the percentage of the number of word
In the English version, the Motive dictionary has 788 entries in affiliation,
and 1,422 in power. The regressive imagery dictionary (English version)
(Martindale, 1990) contains 1,815 entries in primary process thought con-
tents, and 668 in secondary process thought contents. The Dictionaries of
concreteness, imagery, and meaningfulness (Paivio and others, 1968) each
contain 1,508 entries (words rated by judges on 7-point intensity scales). The
French version of the Motive dictionary has 627 entries in affiliation and
1,391 in power. The French version of the Regressive imagery dictionary has
2,019 and 745 entries respectively in primary process and secondary process
thought contents. For the French, we used a Dictionary of imagery of 1,130
entries (Hogenraad & Orianne, 1981).
In support of the Motive dictionary, we analyzed texts that contained fa-
cets of a conflict (Hogenraad, 2003, 2005). For example, texts about the out-
break of World War II show an increasing gap between power and affiliation
(Hogenraad, 2003). By contrast, in the analysis of Robert Kennedy's memoirs
(1969) of the October 1962 Caribbean missile crisis, the gap decreases over
the 16 days of the crisis as a solution comes into view. Martindale (1975,
1979) and West, Martindale, and Sutton-Smith (1985) had previously proved
the dictionaries of Regressive imagery, concreteness, imagery, and meaning-
fulness to be valid indicators of creativity.
170 / Aesthetics and Innovation

The tide of war

“Vive la guerre” (long live the war) cried the crowd as President Ray-
mond Poincaré and Premier René Viviani alighted from the train on their re-
turn from St-Petersburg on July 29, 1914 in Paris. Five short days later, the
Great War broke out. Meanwhile, on the other side of the Channel, Virginia
Woolf complained that, because of war, “all creative power is cut off”
(Woolf, 1985, p. 235). What is proper of artists like Virginia Woolf is not of
political leaders. The general design of the mosaic of results in Table 2 brings
no support to Virginia Woolf. When the risk of war increases, imagery (and
its variants) increases too. Secondary process thought contents are the oppos-
ite of the variants of imagery; risk of war and secondary process thought con-
tents correlate logically in the opposite direction (Figs. 1 and 2). This general
design needs qualification. The pattern holds for the first four documents ana-
lyzed (WW I and II, and Bush and Blair), in which a conflict did break out. In
passing, the 95% bootstrapped confidence interval of two measures on the
Bush speeches includes zero. Bootstrap statistics are calculated from the dis-
tribution of values after randomly resampling the data (Péladeau, 1996). In
other words, we may be 95% confident that, say for risk of war and imagery,
the true value of the correlation (.21) lies somewhere between -.05 and .43 –
which is not good enough because the sign of the association changes radic-
ally (Hogenraad & McKenzie, 1999). In the main, the risk of war in the St-
Germain novel and the Kennedy tapes displays no statistically significant as-
sociation with any indicator of creativity. It is no coincidence, as academics
say, that these two documents point up to conflicts that were settled peace-
We further ran separate analyses on the Kennedy tapes corpus by splitting
it into “hawks”, “doves”, and “President Kennedy”. The “doves” included
Dean Rusk, Robert S. McNamara, George W. Ball, Theodore C. Sorensen,
and Adlai E. Stevenson (Blight & Welch, 1989, pp. 9-15). The “hawks” in-
cluded Paul H. Nitze, C. Douglas Dillon, General Maxwell D. Taylor, John
McCone, Dean G. Acheson, and General Curtis E. LeMay. We could not
catalogue Robert Kennedy or Special Assistant McGeorge Bundy in either
group. We thought the more a group, including the President, wanted to reach
a peaceful solution to the conflict, the more it would express complex
thought contents (lower imagery and concreteness, but higher levels of sec-
ondary process thought contents).
Table 9-2. Correlations between the risk of war and indicators of novelty

Perversion of words / 171

Note: a Confidence interval

172 / Aesthetics and Innovation

Figure 9-1. Risk of war and secondary process thought contents in diplomatic docu-
ments before and during World War II.

Figure 9-2. Risk of war and secondary process thought contents over 72 speeches
made by PM Blair between September 11, 2001 and March 20, 2003.
2001: speeches 1 to 40
2002: speeches 41 to 57
2003: speeches 58 to 72.

We were wrong on that account, except in one—perhaps remarkable—

case. President Kennedy stands out, with a correlation between risk of war
Perversion of words / 173
(decreasing) and secondary process thought contents (increasing) of r (18) =
-.37, p < .05, 95% confidence interval: .77/.16). The equation of the trend for
the risk of war in the interventions of President Kennedy is R² = .19, F(1, 18)
= 4.2, p < .10. For secondary process thought contents, the equation is R² =
.17, F(1, 18) = 3.7, p < .10. As the risk of a conflict between the USSR and
the USA decreases, the language of the president becomes more abstract and
complex (Fig. 3). Not by much, the correlation is barely significant, its
confidence interval includes zero, but perhaps we have picked up a tip here.
It is as if the president was summarizing the opinions of his advisers to
identify the grounds on which he could patch up an agreement with Chairman
Nikita Khrushchev. It would have been comfortable for readers to learn that
both the “doves” and the president displayed this pattern, and in a stronger
way at that. But readers would not pardon us their disappointment if future
confounding facts decisively discarded this finding.
Using the integrative complexity of messages, Guttieri and others (1995)
also analyzed documents related to the Cuban crisis of October 1962. The de-
gree to which one distinguishes and integrates different perspectives when
processing information makes their measure of integrative complexity effi-
cient to evaluate complex circumstances expressed through verbal messages.
Little wonder then if integrative complexity and secondary process thought
contents yield similar results. The Kennedy tapes database used here includes
most of the Guttieri data. Guttieri and others (pp. 614-616) showed the pres-
ence of moderate levels of integrative complexity through the crisis. They
noted no significant differences between “hawks” and “doves” on that meas-
ure. And President Kennedy’s levels of complexity were lower at later stages
of the crisis. Also, Suedfeld and Tetlock (1977) analyzed diplomatic docu-
ments related to the days before the outbreak of World War I. Their measure
was again the integrative complexity of messages. Their data differ from
ours, but the result of their analyses is consistent with ours. Complexity de-
creases significantly between the two periods before the outbreak of WW I
they dub preliminary and climax. Their result matches the significant correla-
tions of Table 2 (column “WW I”) between the risk of war (increasing) and
indicators of creativity.

Closing Remarks: Perversion and Creativity

The documents we analyzed are time data. For the historian, such docu-
ments suffer from the intrinsic weakness of being out of their historic context
174 / Aesthetics and Innovation

Figure 9-3. Risk of war and secondary process thought contents in the interventions of
President Kennedy during the 20 EXCOMM meetings held between October 16 and
October 29, 1962 during the Caribbean missile crisis

(Blight & Welch, 1989, p. 5).For the non-historian, that same weakness
causes the results to be all the more robust –despite the limited number of
cases. The association between risk of war and creativity, in the cases where
the war breaks out, reflects a management of simplification that brushes aside
opportunities of change. When a conflict is in the offing, political elites com-
municate to the public with an excess of simple concrete items that every-
body can understand. A denatured language indeed. A better combination of
circumstances exists when a potential conflict, once unearthed, remains un-
disclosed, as during the early stages of the Cuban crisis. Political elites are
Perversion of words / 175
then free to explore choices and rationalize them with their peers in as com-
plex terms as necessary. And keep open opportunities of change.

“And what is the purpose of a prophet except to find meaning in trouble”

(Marilynne Robinson, Gilead, p. 267)

Author note
For their inspiring comments, I thank Andrew Wilson, Linguistics, Lan-
caster University, UK, Dmitry Ushakov, Psychology, Moscow State Uni-
versity, and Vassilis Saroglou, Psychology, Université Catholique de Louv-
ain, Louvain-la-Neuve, Belgium.

Auden, W. H. (1989). The dyer's hand and other essays. New York: Vintage
Bacevich, A. J. (2005). The new American militarism. How Americans are
seduced by war. New York: Oxford University Press.
Blight, J. G., & Welch, D. A. (1989). On the brink: Americans and Soviets
reexamine the Cuban Missile Crisis. New York: Hill and Wang.
Camus, A. (1956). La chute. Paris: Gallimard.
Cattaneo, C. (1863). Dell’antitesi come metodo di psicologia sociale. Ac-
cessed May 13, 2006:
Chasseguet-Smirgel, J. (1996). Creativity and perversion. London: Free As-
sociation Books.
Cohen, J. (1966). Structure du langage poétique. Paris: Flammarion.
Green, H. (2000). Loving. London: Vintage. (Original work published 1945).
Greene, G. (2001). The quiet American. London: Vintage. (Original work
published 1955).
Guttieri, K., Wallace, M. D., & Suedfeld, P. (1995). The integrative complex-
ity of American decision makers in the Cuban missile crisis. The Journal
of Conflict Resolution, 39, 595–621.
Hayes, D. P. (1992). The growing inaccessibility of science. Nature, 356,
Headlam, J. W. (1915). The history of twelve days, July 24th to August 4th,
1914: Being an account of the negotiations preceding the outbreak of war
based on the official publications. London: T. Fisher Unwin.
176 / Aesthetics and Innovation
Hogenraad, R. (2003). The words that predict the outbreak of wars. Empiric-
al Studies of the Arts, 21, 5–20.
Hogenraad, R. (2005). What the words of war can tell us about the risk of
war. Peace and Conflict: Journal of Peace Psychology, 11, 137–151.
Hogenraad, R., Daubies, C., Bestgen, Y., & Mahau, P. (1995). Une théorie et
une méthode générale d'analyse textuelle assistée par ordinateur. Le sys-
tème PROTAN (PROTocol ANalyzer) (Version 32-bits du 10 novembre
2003 par Pierre Mahau, service informatique de la Faculté de Psycholo-
gie). Louvain-la-Neuve (Belgium), Psychology Department, Catholic
University of Louvain.
Hogenraad, R., & McKenzie, D. P. (1999). Replicating text: The cumulation
of knowledge in social science. Quality & Quantity, 33, 97–116.
Hogenraad, R., McKenzie, D. P., Morval, J., & Ducharme, F. A. (1995). Pa-
per trails of psychology: The words that made applied behavioral sci-
ences. Journal of Social Behavior and Personality, 10, 491–516.
Hogenraad, R., & Orianne, E. (1981). Valences d'imagerie de 1.130 noms de
la langue française parlée. Psychological Belgica, 21, 21–30.
Hussey, M. (Ed.). (1992). Virginia Woolf and war: Fiction, reality and myth.
Syracuse, NY: Syracuse University Press.
Kennedy, R. F. (1969). Thirteen days: A memoir of the Cuban missile crisis.
New York: W. W. Norton.
Kipling, R. (1899). “The white man’s burden”. Accessed April 26, 2006:
Kris, E. (1952). Psychoanalytic explorations in art. New York: International
Universities Press.
McClelland, D. C. (1975). Power: The inner experience. New York: Irving-
ton Publishers.
Machiavelli, N. (1999). The prince. London: Penguin Books. (Original work
published posthumously in 1531—1532).
Martindale, C. (1975). Romantic progression: The psychology of literary his-
tory. Washington, DC: Hemisphere.
Martindale, C. (1979). The night journey: Trends in the content of narratives
symbolizing alteration of consciousness. Journal of Altered States of
Consciousness, 4, 321–343.
Martindale, C. (1990). The clockwork muse: The predictability of artistic
change. New York: Basic Books.
May, E. R., & Zelikow, P. D. (Eds.). (1997). The Kennedy tapes: Inside the
White House during the Cuban missile crisis. Cambridge, MA: Harvard
University Press.
Perversion of words / 177
Merton, T. (1976). War and the crisis of language. In T. Merton (Ed.), On
peace (pp. 138–151). Oxford, UK: Mowbrays.
Orwell, G. (1990). Nineteen eighty-four. London: Penguin. (Original work
published 1949).
Paivio, A., Yuille, J. C., & Madigan, S. A. (1968). Concreteness, imagery, and
meaningfulness values for 925 nouns. Journal of Experimental Psycholo-
gy Monograph Supplement, 76, 1–125.
Parain, B. (1934). Essai sur la misère humaine. Paris: Grasset.
Péladeau, N. (1996). Simstat for Windows. User's guide (Version 2.0, 10 Oc-
tober 2000). Montreal: Provalis Research. ().
Perelman, C., & Olbrechts-Tyteca, L. (1969). The new rhetoric. South Bend:
University of Notre Dame Press.
Perret-Clermont, A.N., Mugny, G., & Doise, W. (1976). Une approche
psychosociologique du développement cognitif. Archives de Psychologie,
54, 135–144.
Robinson, M. (2005). Gilead: A novel. London: Virago.
Saperstein, A. M. (1995). War and chaos. American Scientist, 83, 548–557.
Sorokin, P. (1985). Social and cultural dynamics. A study of change in major
systems of art, truth, ethics, law, and social relationships. London: Trans-
action Publishers. (Original work published 1937).
Strachan, H. (2001). The First World War—Volume I: To arms. Oxford: Ox-
ford University Press.
Suedfeld, P., & Tetlock , P. E. (1977). Integrative complexity of communica-
tions in international crises. Journal of Conflict Resolution, 21, 169–184.
Thorngate, W. (1990). The economy of attention and the development of psy-
chology. Canadian Psychology/ Psychologie Canadienne, 31, 262–271.
Tuchman, B. W. (1994). The guns of August. London: Ballantine Books.
United States Government Printing Office (1944). Events leading up to
World War II. Chronological history of certain major international events
leading up to and during World War II with the ostensible reasons ad-
vanced for their occurrence. 1931—1944. Washington, D.C.: United
States Government Printing Office. Accessed March 1, 2006: .
Ushakov, D. (December 28, 2005). Personal communication.
Van Natta, D. (March 27, 2006). Bush was set on path to war, British memo
says. Accessed May 18, 2006:
Walder, F. (1958). Saint-Germain ou la négociation. Paris: Gallimard.
178 / Aesthetics and Innovation
West, A. N., Martindale, C., & Sutton-Smith, B. (1985). Age trends in the
content of children's spontaneous fantasy narratives. Genetic, Social, and
General Psychology Monographs, 111, 391–405.
Woolf, V. (1985). The diary of Virginia Woolf: 1936-1941, Vol. 5. A. O. Bell
(Ed.). New York: Harcourt Brace Jovanovich.
Woolf, V. (2000). Jacob's room. London: Vintage. (Original work published



Art Constructs as Generators of the Meaning of

the Work of Art

Viktor F. Petrenko and Olga N. Sapsoleva

Moderate nationalist and patriot, and one of the founders of the Russian rul-
ing party The Unified Russia, tries to express the Russian mentality (a “mys-
terious Russian soul”) as opposed to the pragmatic West. This study was con-
ducted to examine Mikhalkov's feature film ‘The Siberian Barber’ taken into
account its implicit ideology.

Kant (1787/1929) stressed that phenomenality requires intentionality to
be classified as consciousness. Husserl’s (1939/1954, 1913/1962) existential
phenomenology examined the life-world as apprehended by individuals
through their own perspectives.


Likewise, Kelly (1955/1991) developed the philosophy he called con-

structive alternativism. It comprises the idea that reality is always experi-
enced from one or another perspective, or alternative construction. Kelly’s
constructionism highlights a person as anticipating events by construing their
replications. This is the step from construction system (knowledge, under-
standing) to anticipation. A person's construction system varies as he or she
successively construes the replication of events, other people, or oneself.
Kelly’s constructionism contrasts with the Marxist reflection principle. The
latter underlies realism and naive materialism, or socialist realism in artistic
creative work. Whereas the reflection principle assumes there is only one true

Art constructs / 181
reality, the constructionism emphasizes an infinite number of alternative con-
structions one may take towards the world. Instead of "objective reality" con-
taining no subjective intentionality Rubinstein (2001) followed Heidegger
(1962) in that the "world of existence as the world of human suffering…"
(Rubinstein, 2001, p. 19) can be the subject of psychological consideration
and realization.
For Kelly (1955/1991) a construct is an individual form of categorization
of the world, other people, or oneself. Operationally, the construct serves as
an element gluing together a series of attributes in an individual cognitive
standard. If a child asserts that "a dirty shirt is warmer" (Chukovsky, 2005),
or a woman visiting a family consultant is of the opinion that "all men are
swine", these are their specific life constructs under consideration from a
teacher’s or psychotherapist’s perspective. Social stereotypes, fragments of
canonical texts, aphorisms by great thinkers, sayings, or even fragments of
advertising texts replacing the system of philosophical or religious world
view in the worldly consciousness can be specific social constructs adopted
by the individual and becoming his personal constructs. Deleuze and Guattar
(2000) suppose that the function of world cognition is creating concepts as
"stable clusters of meaning". Constructs are used as building materials for the
Bakhtin (1979a, 1979b), Lotman (1999), and Vygotsky (1930/1978) em-
phasized the dialogue-based origin of consciousness. It was considered
through interiorization of social interaction and human dialogue with signi-
ficant others. Bakhtin (1979b) defined the works of Dostoevsky (1846/1985)
as 'polyphonic' novels. Every character is a competent voice of full value in a
polylog to find and prove a truth of his/her own life. In terms of physical sci-
ence, we could say that every character of a polyphonic novel plays the refer-
ence role. When an absolute system of coordinates is absent (it is claimed by
an author position), a polyphonic novel describes the relativism of world
views passionately seeking to be understood and heard. Both judgments and
actions by characters can be considered as replicas in such a dialogue.

The concept of art construct

Based on Bakhtin's (1979b) idea, we introduce the concept of art con-

struct as an opposition to characters' life positions. In turn, the language of
characters conceives of oppositions that generate the art construct. The author
of the work expresses an idea with no wording in the language. This idea has
so many aspects that it cannot be expressed verbally. Oppositions of some
symbols are required for characters to raise their part. The character opposi-
182 / Aesthetics and Innovation
tions gains a simultaneous structure of concept displayed then in the text. In
their study, Petrenko and Pronin (1990) found that a reader's world view
changes through art constructs. It is operationally expressed through trans-
formation of the reader's semantic space. When the artwork effects the read-
er's world, new meaning dimensions (consciousness constructs) can appear as
art constructs.

The psychosemantic approach

We developed the psychosemantic approach to study artworks. A "person-

oriented" approach was used to understand the other. It consists of the work
of art as captured indirectly that is in how a person experiences (sees, hears,
perceives, understands) the work of art by construing their replications from
their own perspective. Then the work of art acquires another existence in the
reader’s consciousness. It can be examined through construction of subjective
semantic spaces.
A person needs some knowledge to live and to participate in social inter-
actions. As a rule, world views of some spheres are poorly structured and
poorly realized if the person received no special training (Kelly, 1955). Bru-
ner and Tagiuri (1954) called this knowledge ordinary consciousness and
linked it with personality implicit models. Consciousness is categorized on
the most general meanings such as time, space, causality, value, etc.
(Gurevich, 1972; Stepin, 2000): They proliferate implicit models which un-
derlie perception and understanding of world substantial spheres. Categories
are tools by which a person discerns the world. Usually, he or she is not
aware of it, though. In a psychosemantic study, a participant is asked not to
introspect or reflect on the categories. Rather, the participant is asked to pro-
duce another kind of activity. He or she should sort objects such as works of
art. In doing so, the participant evaluates them on some scales, considers po-
tential motives of characters, puts one art work as compared to another works
of art, etc. Categorizations and implicit models are used in a "consumption
mode". Participants’ judgments are recorded in the form of data matrices.
Then mathematical statistical procedures are used to bring to light categories
that the person used implicitly. In geometric presentations of semantic spaces,
categories are seen as categorical space axes. In turn, the objects appear as
coordinate points within such semantic spaces. Individual parameters of se-
mantic spaces are operational equivalents of various parameters of the per-
son's consciousness (Petrenko, 2005). Substantive use of semantic spaces
have put forward the approach according to which one’s world view or world
view of other persons can be measured and evaluated.
Art constructs / 183
General procedure

To study artwork’s perception and understanding we use, in particular, the

procedure of triadic choice developed by Kelly (1955; Francella & Bannister,
1987) that involves a comparison of characters. Based on this procedure it
was found that viewers use constructs when they estimate characters from the
feature film The Stalker (Petrenko, 2005). Another procedure is the "poly-
identification" developed by Petrenko (1987, 2005). It is widely used within
the psychosemantic approach. Depiction of tentative conduct in proposed cir-
cumstances of literary or film characters are used to frame semantic spaces.
Significant others, oneself, contemporaries, historical of literary characters
can be configured in a semantic space. In what follows the goal is to discover
a person’s world perception, to understand his or her values, settings, and
personal meanings. Our other psychosemantic procedure is the "motive attri-
bution". Viewers are asked to assign motives of characters’ conduct (Pet-
renko, 1987; Petrenko et al, 1988; Petrenko & Sapsoleva, 2002). Then view-
er's understanding of a character's inner world is reconstructed. We arrived at
our main conclusions in this way. When a viewer accepts the character and
self-identifies with it, the more multidimensional and complicated motiva-
tional palette of character’s conduct appears. Given this, viewer's perception
of the character is more subjective rather than object-oriented.
Examining the work of art can be considered as a search for the personal
meaning of a text. The creator expresses them in the text and through lan-
guage and emotion translates meanings to other people. This is a mental
product derived from original understanding of how the world is structured.
The emergent new categorizations differ from the stereotypical categories.
In recent years, a series of monographs appeared in Russian psychology.
They consistently develop the theory of psychology of art (Allakhverdov,
2001; Dorfman, 1997; Leontiev, 1998).

Art constructs of the feature film ‘The Siberian Barber’

We will briefly retell the story for those who did not see the film. It is
Russia, in the nineteenth century during the reign of Alexander III. An Amer-
ican adventurer, Jane, comes to Russia in order to "push" a technical project
by engineer McCrecken through the Russian military department. To achieve
that, she needs to gain the sympathy of general Radlov, head of the junker
school. On the way to Moscow, she gets acquainted with a young junker, Tol-
stoi, and love feelings arise between them. In carrying out the purpose her
coming to Russia and seeking sympathy from general Radlov, Jane provokes
a feeling of jealousy in Tolstoi and a dislike for Tolstoi in Radlov. A conflict
184 / Aesthetics and Innovation
arises between the men, and Tolstoi is condemned to penal servitude under a
faked accusation of terrorism. In our opinion, director Mikhalkov conceived
the film so as to show the unique features of the uncontrollable Russian open
and emotional soul in opposition to the rational and pragmatic West. The film
is interesting not only as a work of art, but also as a work of ideology.
Mikhalkov is a well known public figure, one of those who supports a unified
Russia. His view of Russian identity as expressed in the film brings a deeper
understanding of Russian search for its cultural originality and its part in the
modern world.
Mikhalkov's feature film ‘The Siberian Barber’ is undoubtedly very im-
portant for Russian cultural life. A large-scale advertising campaign, high fin-
ancial expenditure for the film production, amd famous actors yielded a great
cultural and artistic effect in Russia. Mikhalkov himself is one of the most in-
teresting directors of Russian cinema. Public opinion expects masterpieces
from him. Not surprisingly, the Mikhalkov's feature film was first presented
at the festival in Cannes in the hope of a prize.
On the contrary, Mikhalkov's feature film generated broad discussion in
the press, on TV and on the Internet. Judgments ranged from excited refer-
ences about tne glorious Russian army and homesickness about "the Russia
we lost" to ironic statements about popular presentation of Russian life as
"matrushka", "vodka", and "balalaika" (Sokolov, 2000) and presentation of
cadets as a "crowd of exalted chaps”. Likewise, judgments ranged from posit-
ive to negative ones in the Western press. Irina de Chicoff (‘Figaro’) wrote
that "Nikita Mikhalkov managed to express the spirit of ancient traditions and
charm of Russian life during the reign of Alexander III to avoid "unmoderate
patriotism". In contrast, Didier Peron (‘Liberation’) defined the film as "…
about wearisome three-hour long advertising of a product named "Russia".
In our opinion, it is an excellent director's work. One can see beautiful
plays by Menshikov, Ormond, Petrenko, Il'in. Operator Lebeshev created a
dynamic picture similar to the art of Surikov, Borisov-Musatov, Somov, or
Kustodiev. However, so broad a range of film judgments was conditioned not
so much by its artistic advantages. Rather, some historical and household in-
accuracies and the ideological implications led to impetuous discussions. In
an interview, Mikhalkov himself says that this is a film about "inner human
dignity". It is most probably so. However, the leitmotif of the film, in our
opinion, is the phrase "He is a Russian, and it explains a lot". Of course, this
is a film about the Russian idea so much necessary and disputable at the bor-
der of centuries. The uncertainty of values and lack of widely shared ideology
can be seen as a sort of point of bifurcation. The script writer Ibragimbekov
Art constructs / 185
and the film director Mikhalkov looked to the past in their search for way of
nationhood, cultural, and moral perspectives.
The compositional peculiarity of this film is shown this way. Russia is
seen as if through a foreigner's eyes (Jane). Jane’s cultural tradition and the
Russian one differ. Jane extrapolates some native attitude to Russian life,
likely of taking a fresh but somewhat surprised look at Russia. An offscreen
voice (by Mikhalkov himself) gives appropriate explanatory comments.
These meta-insertions fix key ideas and sets, in our opinion, of the author's
art constructs. They look as a sort of prompt to the viewer in his understand-
ing of the film.
This study was conducted to examine Mikhalkov's feature film ‘The
Siberian Barber’ taken into account its implicit ideology expressed in art con-
structs. In contrast to critics relying on their own taste, values and ideological
settings, our purpose was to assess film art constructs from the perspective of

Study 1. Character discrimination



Participants were 100 students from various higher schools of Moscow,

50 women and 50 men. Age ranged from 18 to 30.

Stimulus material

The following characters were administered to participants to their estim-

ations: Tolstoi, Jane 1 (upon arrival in Russia), Jane 2 (upon falling in love
with Tolstoi), Radlov, Mokin, Tolstoi's mother, McCrecken, Duniasha, Alex-
ander III, Grand Duke, Polievsky, Terrorist, Sergeant, Andrew. Participants
were suggested to make their estimations departing from two positions: as if
they were Russian or American persons.


After a collective viewing of the film, participants were asked to compare

film characters in pairs. The participants were to mark some bipolar features
of suggested characters. For example, the pair Andrew ↔ Sergeant: "Adhered
186 / Aesthetics and Innovation
to some value-laden domains of human living ↔ Absorbed in everyday in-
strumental behavior".
Participants made 1,038 constructs in total. The number of individual con-
structs ranged from 5 to 21, 10 constructs per one participant, approximately.
By means of cluster analysis 275 primary constructs were specified. Then
participants were asked to refer each character to the left or right pole of
these primary constructs. Next, participants rated the characters on seven-
point scale.

Data analysis

Individual data matrices were combined into a common data matrix. Prin-
cipal Components Analysis was performed to test differences among charac-
ters. A correlation input matrix and varimax rotation procedures were used
(Mitina & Mikhailovskaya, 2001).


A Principal Components Analysis resulted in the extraction of six factors.

Factor 1 (27.1% of variance explained) included the following constructs.
191. He lives by other people's instructions ↔ He lives as he wishes
without reference to others.
138. He does not manifest himself as a personality, as something unique
in any way; an ordinary person ↔ He differs from other people in his views,
he is not afraid of demonstrating his personality.
259. He lives by rules ↔ He lives as he wishes.
100. He tames to what happens without trying to change anything ↔ He
interferes in the course of events and uses methods of pressure and struggle.
196. He has no goal in life, he wants nothing ↔ He has a clear goal to be
achieved using any means.
102. He is more of an observer or contemplator in life, not its active parti-
cipant ↔ He is an active person in life, he needs everything, he interferes in
everything, he likes to live, not to merely observer life.
121. He is satisfied with what he has, he seeks nothing, he changes noth-
ing ↔ He strives to have more and better than is available.
28. He is incapable of defending his point of view ↔ He actively defends
his point of view
76. Natural conservatism, outstanding patience, non-aggressive life posi-
tion ↔ Strong instinct of struggle for rights.
33. Fear of openly expressing a protest ↔ Ability to openly protest.
170. Traditionalist ↔ Modernist.
Art constructs / 187
151. He submissively accepts his destiny ↔ He struggles for a better
place in life, he is capable of changing his destiny.
186. He lives as if on his last legs, as if he were dying already ↔ He lives
a full life.
34. He follows traditions and supports monarchy ↔ He wants to change
the world and its order.
111. He always expects the worst combination of circumstances ↔ He
believes that a better solution can be found.
236. He is quite happy and satisfied with himself ↔ He seeks changes.
164. He lives in his little world ↔ He has a broad range of interests and
248. He lives a simple and understandable life and knows in advance how
and what will happen to him ↔ He lives a life of his own, which is im-
possible to understand.
56. General inertia in life, he lives as instructed, he feels it is better to
obey and to submit to circumstances ↔ Desire to get to the top, to new posi-
tions and opportunities, even if using not very nice methods.
200. He is full of prejudices ↔ He has no prejudices.
168. He is dedicated to monarchy ↔ He is a democrat nuts-and-bolts.
203. He more speculates than does ↔ He is a man of deed.
180. He is wacky ↔ One cannot mock him without serious consequences.
235. He is a led person and thinks that his fate is predetermined, and noth-
ing can be changed ↔ His fate is in his hands.
The first factor was interpreted as "Passive submission ↔ Autonomy, ac-
tion, and struggle".
Loadings of Factor 1 and their poles (see Fig. 1) evidenced that Duniasha,
Great Duke, Tolstoi’s Mother, Mokin, Polievsky, and Tolstoi can be referred
to the pole of passivity and submission. Terrorist, Jane 1, Jane 2, Andrew, and
McCrecken can be referred to the opposite pole of those people who are able
to fight for justice and own interests.
Factor 2 (22.1% of variance explained) included the following constructs.
27. Inability to use other people for achieving one's own goals, desire to
support honest and open relations ↔ Using other people for selfish purposes.
256. He values honesty and integrity in people ↔ He values what people
can give him (power, wealth).
22. He possesses the felling of inner dignity, does not want to get even
with other people ↔ He is revengeful and inclined to get even with people
and to calculate possible revenge options.
188 / Aesthetics and Innovation

Figure 10-1. The semantic sphere organized by Factor 1 and Factor 2

269. He is capable of compassion ↔ He does not spend feelings.

230. He is sincere and cannot lie ↔ If necessary, he won't hesitate to lie.
221. Romantic ↔ Pragmatic.
237. He is a good and true friend ↔ One won't wish a friend like he to the
240. He can share other person's problems ↔ He does not care what other
people feel or think.
169. He dissolves himself in another person and lives only his life ↔ He
never forgets about himself.
19. He can feel guilt for what happened ↔ He deletes people from his life
and abandons them as unnecessary obstacles.
250. He can sacrifice anything for the sake of love ↔ Sacrifice is not typ-
ical of him.
226. He thinks not only about himself, but also about other people ↔ He
thinks only about himself.
127. He is capable of deep feelings, both pain and joy ↔ His feelings are
superficial, his soul is never touched.
199. He respects himself and does not go down to platitude ↔ He humili-
ates himself through indecent behavior.
Art constructs / 189
158. He is oriented toward abstract spiritual values (love, brotherhood,
honor, duty) ↔ He is oriented toward concrete tangible values (wealth,
prestige, power).
144. He strictly judges his own behavior ↔ He always find an excuse for
161. Humanist ↔ He has no human feelings for people.
231. He cannot do harm to people ↔ He can deliberately cause pain to
176. He follows the feeling of responsibility and duty ↔ When duty and
responsibility are an obstacle, he moves them to the background.
218. He has stable values, he does not change them ↔ He thinks so as it
is profitable to him.
23. He trusts people ↔ He is suspicious and expects deception.
59. He is too naive and does not understand that words are just words,
that there situations in life where people lie and deceive each other ↔ He
doubts everyone and everything.
163. Word and deed coincide ↔ Words and actions contradict each other.
215. He is not suspicious ↔ He is as if on alert and always expects to be
deceived or offended.
142. He acts without calculation ↔ He is a sample of adventurous style
and rationalism.
Factor 2 poles are Tolstoi, Andrew, Jane 2, Duniasha, Mokin, Russian role
position as people "capable of compassion, sincere feelings and direct artless
behavior" in opposition to Radlov, Great Duke, Tolstoi’s Mother, McCrecken,
American and Jane 1 characterized as "people using other people as a means
for achieving their selfish goals".
Factor 2 was interpreted as "Dignity ↔ Lack of principles".
Factor 3 (13.0% of variance explained) included the following constructs.
258. Reckless ↔ Governed by his head.
66. He never thinks over his actions, is impulsive, and, the main thing, he
never thinks afterwards either ↔ He thinks about the future and con-
sequences of what is happening now, he never lives in illusions, he is a real-
120. Dreamer ↔ Realist.
54. He is not used to discipline, cannot control emotions ↔ He is a dis-
ciplined person, i.e., strictly respects the division into the senior and junior.
113. His entire behavior and feelings are irrational, he always makes
choice not for the benefit of consciousness ↔ He always thinks in any cir-
cumstances, and expresses his emotions only after thinking.
190 / Aesthetics and Innovation
98. The entire life and its variety are emotions to him, he is very sensitive
and jumps from laughter to tears ↔ He is more oriented to rational arguments
than to emotions In principle, he is quite predictable in his reactions, one
does not expect hysteria or emotional explosions of him.
140. It is more important for him to express his feelings than to think over
the situation and to do something ↔ It is important for him to understand
how to behave and what to do.
247. He lives in the world of illusions ↔ He builds no illusions and lives
a real life.
119. He lets down his friends, family and loved woman without any reas-
on ↔ He prefers first to get all information, and then to make a decision.
114. He has completely lost the feeling of reality, he lives n an invented
world with his rules, and this mismatch between dream and reality brings him
into a stupor and as if suppresses the voice of reason ↔ He is well aware of
the real situation, and there is no place for empty dreams in his life. Maybe
he is too rational, but instead he has no "lost illusions".
246. Any trifle, unexpected event or unpleasant thing can throw him out
of joint ↔ Has firm character.
68. He has no contact with reality and lives in an invented world ↔ He
thinks about the future and consequences of what is happening now, he does
not live in illusions, he is a realist.
166. He has no core, he is unstable ↔ A man with a core.
266. He is a child in his soul, who steel needs care ↔ He is an independ-
ent adult person
67. He is a brinksman and has no inclination to trade-off ↔ He thinks
about the future and consequences of what is happening now, he does not live
in illusions, he is a realist.
275. If he loves, he does not keep silence, and is ready to cry about his
love ↔ He loves silently and suffers secretly.
162. He ignores facts and sees what he wants, not what is taking place in
reality ↔ He gathers facts and analyzes them before doing anything.
271. He is unreasonable ↔ He never loses ability to think and prudence.
Loadings of Factor 3 and their poles (see Fig. 2) evidenced Tolstoi is with
a large gap with other characters. Tolstoi's Mother, Terrorist, and Great Duke
are characters who cannot keep themselves in hand and are immersed in an
illusory world. Conversely, Mokin, Duniasha, Jane 1, Jane2, Andrew, and
McCrecken showed the rational and common sense. Participants perceived
them as calculating, pragmatic people who clearly know their own interests
and pursue them. The third factor was interpreted as "Cool pragmatic calcula-
tion ↔ Action under emotional impulse, attachment or one’s beliefs".
Art constructs / 191

Figure 10-2. The semantic sphere organized by Factor 3 and Factor 4

Factor 4 (9.1% of variance explained) included the following constructs.

147. Person without roots, nothing holds him, parents are left in the dis-
tant past ↔ He knows who he father was, and can be proud of him.
43. He has been taking care of himself since childhood, nobody has ever
made any gifts to him ↔ He enjoys support from the family. In principle, he
lives all found.
64. He neglects traditions, never shows his national traditions, although is
interested in other people's traditions ↔ He feel national pride and is raised
in the spirit of patriotism, "love for fathers' coffins", instinctive respect for
the Tsar and God, all traditions are absorbed with mother's milk.
198. He can step over his pride ↔ He never forgives people who touched
his pride.
32. He expects material support ↔ He can provide material support.
143. Is in pleading for help and dependent ↔ Russian pride on no visible
175. The family and parents were left long ago in the past ↔ He is proud
of his family and motherland.
189. He is not worldly and loses confidence in himself in the society to
some extent ↔ Society lion.
192 / Aesthetics and Innovation
244. Nobody respects him ↔ His opinion has a certain weight in the soci-
268. He feels he is nobody, as if he were furniture for other people ↔ He
feels his superiority and actively demonstrates it.
224. He is not respected ↔ He enjoys respect and is an authority for
31. He feels insignificant ↔ He feels he is an influential person, wants to
patronize other people and to be significant.
168. Democrat nuts-and-bolts ↔ Monarchy follower.
95. Other people notice or not, the main thing for him is that he is aware
of what he is ↔ It i s important for him that other people notice his achieve-
219. He does not care if he is better or worse than someone ↔ He wants
to become better than others very much.
One pole of Factor 4 occupies Duniasha, Tolstoi's Mother, McCrecken,
and Terrorist. They are people of low social status. The other pole holds Al-
exander III, Radlov, Polievsky, and Great Duke. They are people of power
with superiority over other people.
Factor 4 was interpreted as “High social status (proud superiority) ↔
Low social status ("poor people")”.
Factor 5 (6.0% of variance explained) included the following constructs.
42. He easily adapts himself to foreign environment ↔ He does not un-
derstand foreign culture and does not see cultural differences.
210. He creates a nice atmosphere about himself ↔ He creates tension
around himself.
148. His social position was so that he was well raised and received a
good education ↔ He is a dull soldier who knows nothing but commands.
206. He accepts life in its full variety ↔ Many things are alien to him, he
does not accept them and even does not try to understand.
192. In his opinion, every person has the right to decent existence and
good attitude irrespective of anything ↔ People are nothing like cockroaches
for him.
136. He plays many various roles in life without limiting himself to a
single one ↔ He is limited to a single role in life.
One pole of Factor 5 occupies Jane 1 and Jane 2 (taking a considerable
gap with other characters), Tolstoi, Andrew, Polievsky, and McCrecken. They
accept life in its all variety. The other pole holds Terrorist, Sergeant, and Rad-
lov. They fail to understand people or to put themselves in their places (see
Fig. 3).
Art constructs / 193

Figure 10-3. The semantic sphere organized by Factor 5 and Factor 6

Factor 5 was interpreted as "Open to variety ↔ Narrow-mindedness".

Factor 6 (4.6% of variance explained) included the following constructs.
159. He does intellectual work ↔ Martinet.
122. He is more of a theoretical than practical person ↔ He understands
only facts, and any ideas, theories, or abstract things are not for him.
267. He creates a reality of his own ↔ He is a utilitarian realist.
171. He is a fanatic of his work ↔ He is indifferent to his work.
45. He perceives life as something difficult, as a chain of obstacles ↔ He
reduced all complicated things of life to simple ones.
261. Leader in his relations and initiatives ↔ It is easier for him to follow
someone than to lead other people.
153. He does not care of his family but has remote and global goals ↔ He
is anxious for himself and his family.
One pole of Factor 6 occupies McCrecken (taking a considerable gap with
other characters), Alexander III, and Terrorist. They are given to remote and
global ideas. The other pole holds the rest of the characters. They are ab-
sorbed in current real life.
Factor 6 was interpreted as “Given to abstract idea ↔ Absorbed in every-
day life”.
194 / Aesthetics and Innovation
Once characters were discriminated as appeared in different factors the
later can be considered as the latent art constructs of the film. According to
our theory based on Bakhtin’s idea of he polyphonic novel, the author of a
work of art uses a language and life attitude oppositions (“truths” in Dosto-
evsky’s (1846/1985) terms) held by characters to express his or her ideas. An
art construct (or “artistic meaning of the work” used more traditionally) is be-
hind characters conduct. In a scenic action, actors should display the conduct
contingent on art constructs.

Study 2. Motive attribution

Once the basic lines of characters’ opposition were assessed, film con-
structs can be designated, as well. Of particular relevance to understand the
film’s gist departing from viewers’ perspective is the motives viewers attrib-
ute to characters’ conduct. Next, the motives viewers attributed to characters’
conduct were examined.



Participants were 100 students from various higher schools of Moscow,

50 women and 50 men. Age ranged from 18 to 30.

Stimulus material

The characters’ conduct was taken into consideration.


After a group viewing of the film 54 motives as tentative latent factors of

characters’ conduct were administered to participants. A list of motives in-
cluded such ones as ‘like feeling’, ‘get approval or admiration of other
people’, ‘material interest’, ‘need for risk and acute feelings’, ‘curiosity’,
‘imitation’, ‘strive to be like everybody’, ‘conformism’, ‘of the feeling of
duty and honor’, ‘proceeding from spiritual or religious requirements’, ‘need
for self-realization’, ‘strive to be creative’, ‘spontaneous behavior’, ‘fear to
get disapproval of other people’ (losing face, looking stupid, etc.), ‘fear of ad-
ministrative sanctions’, ‘threat of punishment’, etc. Each motive was estim-
ated on a 6-point scale ranged from 0 to 5.
Art constructs / 195

A psychosemantic procedure “motive attribution” was used to assess the

motives attributed tp film characters. Salient conducts yielded by each among
selected characters were specified. For example, Tolstoi's conducts were con-
sidered this way: ‘Fooling about with friends, he breaks into Jane's compart-
ment’, ‘Once Tolstoi's matchmaking failed, and Radlov accused him of an
unseemly conduct with respect to a lady, Tolstoi is going to have a duel’,
‘Acknowledges himself guilty of an attempt at the Great Duke’, ‘Marries the
housemaid Duniasha’, etc.
We departed from the viewpoint that motives are multiple (e.g., Vignoles
et al., 2006) and a conduct is polymotivated, that is it can be explained by a
number of motives rather than by single one (e.g., Beersma & De Dreu,
2005). Based on this concern participants were asked to fill in a full data mat-
rix ‘motive attribution x conducts’.

Data analysis

Individual raw data were grouped and obtained for each character separ-
ately. Motives were combined in motivation blocks. The number of motives
varied in different motivation blocks according to participants’ estimation.
Some motives were not appropriate with respect to a character. For example,
‘material interest’, ‘career goals’ or ‘striving for cognition’ motives did not
apply to Tolstoi. Raw data were processed using exploratory factor analysis.

Results and Discussion

Because of space limitation factor structures behind viewers’ motives at-

tributed to film characters are not given. Only leading character motives are
described. Those findings are under consideration that would seem of import-
ance to reconstruct viewer's understanding of film characters.
The motives attributed to film characters by critics and students greatly
differed. Our study was conducted in 2002/2003, while the film ‘The Siberian
Barber’ appeared in 1997. A lot has been written about it since that time. It
was very plausible that the student audience would project critics' interpreta-
tion on to characters. However it did not happen. For example, Jane was de-
scribed by film critics and art critics as a "subtly realistic American adven-
turer" (Sokolov, 2000), "wanton girl" (Moskvina, 1999) or "prostitute ordered
from abroad to seduce Russian authorities" (Stepnina, 2001) who is too real-
istic, manipulates people, and faols to understand the mysterious Russian
soul. The students, however, discerned Jane trying to overcome the injustice
196 / Aesthetics and Innovation
with respect to Tolstoi. Indeed, in contrast to other film characters, Jane is in
search for justice. After Tolstoi's arrest she visits Radlov, asks him to be toler-
ant, as well as about Tolstoi's conduct to the court. She applies also to Tol-
stoi's mother to write a solicitation to the Tsar. At the same time, participants
saw Jane as guilty of what happened.
Although critics evaluated Jane "completely free of moral restrictions or
affections" (Stepnina, 2001) students attributed to Jane such motives as love,
compassion, altruism, and feeling of guilt. Jane's ‘manipulatory’ intentions
that are of priority for critics were not of particular importance for the stu-
dents. The latter took Jane into account for cultural oppositions not only
"their pragmatism" ↔ "our spirituality", but also "feeling of a free person be-
ing indignant at obvious injustice" ↔ "slavish obedience”.
Critics described Tolstoi as a "person defenseless against fate strikes"
(Stepnina, 2001). He feels "fatal inclination for fate breaks, burden of inexor-
able fatum, intolerance to offence of honor, and ontological unhappiness" (Ir-
atov, 2001), or as a "psychopathic person" (Sokolov, 2000). On the contrary,
students attributed sincere feelings, continuity and emotionality to Tolstoi. He
is exalted like a "knight of ultimate line". The exaltation is fraught with frus-
trations. His young purity is opposed to pragmatism and greed. As a result,
Tolstoi found himself condemned to penal servitude. Students had clearly
preferred the impulsiveness of Tolstoi ("he did it without thinking"). The
motives of jealousy, competition, and to take revenge over a reasonable solu-
tion to the conflict were attributed to him.
Tolstoi's "lack of core" and "uncontrolled passions" were repeatedly
stressed by critics but students explained that as his "boyishness". Tolstoi is
almost an officer and he is only eighteen. "He is just a boy in love", Jane says
about him. His emotionality is not suppressed by the responsibility of a Rus-
sian army officer yet.
Students discerned a "conformist" line in Tolstoi's conduct too. It is not
blind obedience to conduct as socially required, but his own wish to meet
clear norms. The benchmarks established by the society are the hidden
motives that led Tolstoi to his duel with Polievsky. Even after undeserved
penal servitude and seeing enough human suffering and injustice there, Tol-
stoi keeps the Emperor's portrait in his home. Obviously, it is conceived by
script writer and director to evidence his patriotism and faith to military oath.
Students interpreted this as a manifestation of Tolstoi’s relatedness and con-
Critics described Radlov as “a charming drunkard and small ascal" (Ir-
atov, 2001), "despot and habitual drunkard" (Stepnina, 2001). Radlov is gov-
erned not only by “despot” motives (i.e., hatred, revenged, capitulated to jeal-
Art constructs / 197
ousy, envied or competed), but also he is revolted at deceit and self-protect
against outside intrusion when he is manipulated. In students’ view, of partic-
ular relevance are "family" values. Radlov takes Jane's coquetry in all good
faith, starts to pay his addresses to her, trying to get rid of solitude and to
bring an order in his life. Radlov is ready to make a serious step—introduce
Jane to his mother. Like critics, students saw Radlov as a "habitual drunkard".
However his hard drinking is more complicated than just alcoholism. Radlov
likes his profession, service, and military school. Herein, he makes a career
holding a perspective on those people at power who gained a high social
status due to the right of birth rather than own giftedness or achievements (re-
member the weak-willed replication of the Great Duke).
Radlov's broad nature, probably battle-tried officer in the past, and his in-
clination to gain promotion leads to a deep conflict in his motives. Two seem-
ingly opposite tendencies appear. First is obedience to social norms and
second to go beyond the social norms. This is like Freud (1910/1957) called
"an effect of conflicting Super Ego and Id". We dp not pretend to do a psy-
choanalytical interpretation of Radlov's personality but just emphasize that
the social norms for him clearly prevails. Herein the wide Russian nature
(well expressed by actor Petrenko) remains and a tension appears. Later this
leads to hard drinking.
Radlov's injustice is stressed by critics but is not a salient feature for stu-
dents. Motives such as to strength, self-esteem, and social status are of partic-
ular importance for Radlov too. From the students’ viewpoint altruism and
patronizing other people are inherent in Radlov, as well as his career goals,
material interest, and possessing power in order to improve his social repres-
entation. Still, students do not like Radlov as he lacks motives in the vein of
comradeship, friendship, brotherhood, pity, compassion, moral or religious
values. The lack of these motives in Radlov's conduct is conditioned by the
film authors. We see no friends of his and are unaware whether he has some.
Radlov functions in the coordinate system "supervisor―subordinates". He
feels no need for freedom or independence.
Students saw Radlov not only as a person living in accordance with the
principle "divide and rule" (Iratov, 2001), but also as a sensitive person who
is not confident in himself. He tries not only to achieve a certain social posi-
tion through his dishonest conduct but also to respond to an offence.

General Discussion
What is this film about, after all? What is the set of its art constructs? We
can answer these questions, at least from viewers' position, as soon as we
198 / Aesthetics and Innovation
have character oppositions semantically analyzed and film art constructs
highlighted. This is a film about human dignity and self-sufficient value of
human feelings. Mikhalkov is right about it. The art construct "Sincerity of
feelings ↔ Using other people" (factor 2) obtained in the study is evidence to
this statement. The poles of this factor are characters marked with "Russian
mentality" (Tolstoi, Russian role position, Duniasha, Andrew (as "origin Rus-
sians"), and Jane 2 (with Russified soul because of love for Tolstoi) in oppos-
ition to foreigners (Jane 1, McCrecken, and American role position). In turn,
Radlov, a Russified German, occupies the extreme position among "manipu-
Film authors have obviously suggested an ethnopsychological construct
relied on opposition between an open, uncontrolled and sincere "Russian
soul" and the mercantile and rational West. This is what the offscreen voice
says a peculiar prompt for the viewer: "Your mother tried to guess the mys-
tery of this vast country, which does not yield to common sense. One can de-
ceive, steal or rob in Russia, but a word given by someone can sometimes be
valued more than any official paper, while a carelessly said word can lead to
bloodshed". "I have the honor" is a phrase repeated by various characters
many times during the film. The Orient guided by Confucius’ ideas considers
the "loss of face" (i.e., loss of social status and dignity in the eyes of other
people) as a person’s drama. If a Russian (a nobleman is in the first place) vi-
olates the inner code of honor, he suffers bad losses for him- or herself.
However, is it the film about dignity solely?

Pomerants (1994) wrote: "Development stresses various extremes

balancing each other in each historical nation. One extreme pulls an-
other, the opposite one, after itself. For example, great value is put on
spirituality in Indian culture, up to negation of the world. It means that
one should look for exaggerated sensuality, and it will be found in tan-
trism. If respect for parents is brought to the top religious duty in
China, then one should look for the opposite extreme, and find it in
Buddhism chain. If increased tenderness strikes the eye in the Russian
people and literature, then what is that it is linked to? Obviously in-
creased ability to commit crime, unstable moral samples, and inclina-
tion to chasms that can be tracked in life and in literature…" (Pomer-
ants, 1994, p. 270).
"I remember Leontiev write that it is easier to meet a saint than an
ordinary honest man in Russia", and further: "I even think that tender-
ness is somehow directly linked to business unfairness. Sometimes,
they are combined in a person"
Art constructs / 199

(Pomerants, 1994, p. 271).

We think that Pomerants' deep observation about dominant constructs in-

herent in some cultures is a key to cross-cultural and ethnopsychological
studies, in particular to examine national character. One may only disagree
with the last statement by Pomerants about presence of both poles in a per-
son. Culture sets up extreme points of moral fluctuations that are similar to
the pendulum. If a person chooses the pole of spirituality, then he/she can do
spiritual work to maintain human being in himself and to avoid falling into
the abyss. Some people manage to do it.
The Russian character also relies on other art constructs highlighted in our
study. We already wrote about dignity. Other art construct comprises the op-
position "Sober pragmatic calculation ↔ Action governed by emotional im-
pulse". The emotionality, in the opinion of both film authors and viewers, is a
typical Russian feature. "It seems that nothing can happen "slightly" in Rus-
sia". No less than in Russians though, emotional freedom is probably inherent
in native Americans, descendants of pioneers. This suggestion would mean
that minds of Russian and American people are closer to each other as com-
pared to critics’ estimation.
Increased emotionality, inclination to "give a horse the reins" is a feature
inherent not only in positive film characters. General Radlov conducts out-
rageously during hard drinking, devoutly crosses himself and bathes in an
ice-hole, is a grotesque person. His "Russian nature" affects the Russian audi-
ence. Peasants and merchants, young educated aristocrat count Polievsky, and
a harmonist from ordinary people splash out emotions in common outburst.
"Generally, everything is extremes in this surprising country: semi–naked
peasants beat each other within an inch of their lives on the river ice, and then
beg pardon from each other, they go to war with songs, and go to wedding in
tears, and everything they do is done seriously, everything is done to the end"
(an offscreen voice).
The next art construct is the opposition "Ability to sacrifice oneself ↔
Calculation pursing one's own interests". As was mentioned, this construct
comprises the opposition of Tolstoi, Jane 2, and Duniasha on the one hand,
and McCrecken, Radlov, and partially Tolstoi's Mother on the other hand. In
our opinion, the sacrifice as a feature of Russian national character is not very
articulated in the film. Departing from Russian bipolarity sacredness would
oppose impulse and violence of feeling. This subject is weakly expressed in
the film. The sacrifice looks like a worldly rule ("one may not defile a lady")
rather than taking compassion itself. This is why the Orthodox line is not
200 / Aesthetics and Innovation
manifested, although for Mikhalkov it is of very importance (at least, this
subject appears many times in the film). By the way, McCrecken is shown as
an antipode of Russian spirituality. He is maybe the only person who is keen
on a creative work. In the framework of Protestant ethics (Weber, 1930), he
can be seen as very religious and to sacrifice oneself. However, this perspect-
ive is not for the present. It would be referred to the XX and XXI centuries.
The most powerful construct (factor 1) "Ability to actively resist injustice
↔ Fatalism" is very important to capture viewers' understanding of the film.
In viewers' opinion, an active resistance is inherent in Jane, Terrorist, and An-
drew. It would seem the Russian characters are located at the opposite pole or
in neutral positions in this factor. True, in case of obvious injustice with re-
spect to Tolstoi, only Jane took courage to express indignation and to try to
do something. Cadets and Mokin's indignation in the form of throwing news-
papers with a version of the attempt at the Great Duke can hardly be called an
active protest. The scene of cadets saying goodbye to Tolstoi, Mokin in civil-
ian cloths saluting the prisoner, and a crowd of prisoners' relatives rudely
pushed by the police only stress weakness and resignation of the power.
Tolstoi served out his sentence in a prison and likely saw people's troubles
and suffering there. Notwithstanding the Emperor's pictures cover the walls
of his village home. Viewers discerned the Emperor’s role brilliantly played
by Mikhalkov. Except self-presentation they spotted the Emperor as held no
ideological context. No pole of the constructs comprised him. In a polyphon-
ic play of character’s voices (in vein of Bakhtin, 1979b; see also: Hermans,
1996) the voice of the Emperor is quite muted.
Unlikely, Mikhalkov sympathizes with the Terrorist character. Mikhalkov
invited an actor with obviously a non-Russian appearance to play this role.
On the other side, Mikhalkov as a politician and ideologist and Mikhalkov as
a director and artist are in dissonance here. The monarchic Russia praised by
Mikhalkov is apparently ill. It suffers from authoritarianism and lack of feed-
back through the people. This is the power where cadets, young ladies from
the institute of noble ladies, and tsar's family indulge in a fake world of the
idyll of Mozart's vaudeville. The threatening rumble of forthcoming tragedy
is felt in the extremely emotionally presented scene of convict movement
through patriarchial Moscow. It will blow up Russia "that we lost" from in-
Some critics and Mikhalkov's colleagues graciously appraised the feature
film ‘The Siberian Barber’ as a well-made vaudeville, a cheap popular film.
We remind the reader without discussing the peculiarities of the genre al-
though Dostoevsky's (1865-1866/1991) "Crime and Punishment" can be
called a detective story too. ‘The Siberian Barber’ would seem a complicated
Art constructs / 201
and highly artistic work. It is resonant to nowadays, relies on world views,
and presupposes interpretations yielded by the audience itself.

Allakhverdov, V. M. (2001). Art psychology. Saint Petersburg: DNK. (in Rus-
Bakhtin, M. M. (1979a). Aesthetics of verbal creative work. Moscow: Sovet-
skaya Russia. (in Russian).
Bakhtin, M. M. (1979b). Problems in Dostoyevsky's poetics. Moscow: Sovet-
skaya Russia. (in Russian).
Beersma, B., & De Dreu, C. K. W. (2005). Conflict's consequences: Effects
of social motives on postnegotiation creative and convergent group func-
tioning and performance. Journal of Personality and Social Psychology,
89, 358–374.
Bruner, J. S., & Tagiuri, R. (1954). The perception of the people. In G.
Lindzey (Ed.), Handbook of Social Psychology (Vol. 2, pp. 20–41). Ad-
dison Wesley.
Chukovsky, K. I. (2005). From Two to Five. Moscow: KDU. (in Russian).
Deleuze, G., & Guattar, F. (2000). What is philosophy? Saint Petersburg:
Aleteya. (in Russian).
Dorfman, L. Ya. (1997). Emotions in art: Theoretical approaches and empir-
ical studies. Moscow: Smysl. (in Russian).
Dostoevsky, F. (1865-1866/1991). Crime and punishment. (D. McDuff,
Trans.). London: Penguin.
Dostoevsky, F. (1846/1985). The double (E. Harden, Trans.). Ann Arbor, MI:
Francella, F., & Bannister, D. (1987). New method of personality study. Mo-
scow: Progress. (in Russian).
Freud, S. (1910/1957). Five lectures on psycho-analysis. In J. Strachey (Ed.
and Trans.), The standard edition of the complete psychological works of
Sigmund Freud (pp. 3–55). London: Hogarth Press.
Gurevich, A. Ya. (1972). Categories of middle age culture. Moscow:
Iskusstvo. (In Russian).
Heidegger, M. (1962). Being and time (M. Macquarrie & E. Robinson,
Trans.). New York: Harper & Row.
Hermans, H. J. M. (1996). Voicing the self: From information processing to
dialogical interchange. Psychological Bulletin, 119, 31–50.
Husserl, E. (1913/1962). Ideas: General introduction to pure phenomenology
(W. R. B. Gibson, Trans.). New York: Collier Books.
202 / Aesthetics and Innovation
Husserl, E. (1939/1954). The crisis of European sciences and transcendental
phenomenology (D. Carr, Trans.). Evanston, IL: Northwestern University
Iratov, A. A. (2001). Story of Siberian land 2 (www.russ.ru/2001/3/4.html).
(in Russian).
Kant, I. (1787/1929). Critique of pure reason (N. K. Smith, Trans.). New
York: Macmillan.
Kelly, G. A. (1955/1991). The psychology of personal constructs: A theory of
personality (Vol. 1). London: Routledge.
Leontiev, D. A. (1998). Introduction to art psychology. Moscow: Moscow
State University Press. (in Russian).
Lotman, Yu. M. (1999). Inside thinking worlds. Man—Text—Semiosphere—
History. Moscow: The Languages of Russian Culture. (in Russian).
Mitina, O. V., & Mikhailovskaya, I. B. (2001). Factor analysis for psycholo-
gists. Moscow: Psykhologiya. (in Russian).
Moskvina, T. (1999). Don't say that youth destroyed you
(www.kinoart.ru/1999/6/4.html). (in Russian).
Petrenko, V. F. (1987). Psychosemantic approach towards etnopsychological
studies. Sovetskaya Etnographiya, 3, 9–18. (in Russian).
Petrenko, V. F. (2005). The fundamentals of psychosemantics. Saint Peters-
burg: Piter. (in Russian).
Petrenko, V. F., & Pronin, A. A. (1990). Art construct as a form of semantic
organization of literary text. In R. Kotov (Ed.), Optimization of voice ef-
fect (pp. 31–35). Moscow: Nauka. (in Russian).
Petrenko, V. F., Sobkin, V. S., Nistratov, A. A., & Gracheva, A. V. (1988).
Psychosemantic analysis of perception of film characters by the audience.
Voprosy Psikhologii, 4, 22–34. (in Russian).
Petrenko, V. F., & Sapsoleva, O. N. (2002). Psychosemantic approach to-
wards analysis of perception of work of art. In G. Akopov (Ed.), Art psy-
chology (Vol. 2, Part 1, pp. 3–11). Samara: Samara Teachers Training
University Press. (in Russian).
Pomerants, G. S. (1994). Craving for goodness. Russkoye Bogatstvo, 2, 254–
289. (in Russian).
Rubinstein, S. L. (2001). Man and the world. Saint Petersburg: Piter. (in Rus-
Sokolov, M. (2000). Nothing happens slightly in Russia
(www.russ.ru/2000/ms.html). (in Russian).
Stepnina, N. V. (2001). Ideals and interests
(www.kinoart.ru/2001/7/16.html). (in Russian).
Stepin, V. S. (2000). Theoretical knowledge. Moscow: Progress. (in Russian).
Art constructs / 203
Vignoles, V. L., Regalia, C., Manzi, C., Golledge, J., & Scabini, E. (2006).
Beyond self-esteem: Influence of multiple motives on identity construc-
tion. Journal of Personality and Social Psychology, 90, 308–333.
Vygotsky, L. S. (1930/1978). Mind in society: The development of higher
mental processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman,
Eds. & Trans.). Cambridge, MA: Harvard University Press.
Weber, M. (1930). The Protestant ethic and the spirit of capitalism. New
York: Scribner.

Portrayal of Women and Jungian Anima Figures

in Literature: Quantitative: Content Analytic

Anne E. Martindale and Colin Martindale

Women as they appear in the fantasies, dreams, and literary works of men are
often quite different than the women we encounter in real life. In part, Jung
has argued, that is because these imaginary women symbolize the uncon-
scious or undeveloped part of a man’s psyche. This is especially true of what
Jung called the anima archetype. Thus women and the anima often occur in a
context of evil and of primordial content indicative of the type of thought
found in regressive or unconscious states of mind. Based upon what Jung and
his followers have written on the anima archetype, a computerized content
analytic dictionary was developed and applied to a sample of texts that Jungi-
ans have said concern anima figures. As predicted, the anima is portrayed as
numinous or awe-inspiring, pure, timeless, mysterious, wise, and powerful.

Women as they are portrayed in literature are often quite unlike the wo-
men of everyday life. In part this is merely due to the fact that a literary de-
piction of a perfectly ordinary woman would be of no interest. However, at
least for male authors, it may be due to the fact that authors are not really
writing about real women but about what Jung called anima archetypes.
Jungian theorists have postulated that female characters in general may sym-
bolize unconscious or undeveloped aspects of the ego or self of the author
(Harding, 1975; Jung, 1916; Neumann, 1955). Of course, men portrayed in
literature tend to be unlike ordinary men. Jungian theorists have argued that
the male protagonist of a narrative may symbolize what they called the hero
archetype, which represents an idealized version of the conscious ego of the
author. Many but certainly not all works of literature may profitably be read

Women and Animas / 205
as symbolizing the spirit and soul, as personified by the hero and heroine (an-
ima), of the author. In this chapter, we discuss some quantitative research on
this idea. We begin with a description of Jungian psychology that sets the
stage for this work. Because all of the narratives we studied were written by
men, we focus on the Jungian description of the male psyche.

Jungian Theory
Jung divided the psyche into several parts. The ego or self consists of the
traits we are conscious of or can easily become conscious of. It operates ac-
cording to what he called the principle of Logos. By this he means that it is
rational, reality oriented, purposeful, and so on. Logos is essentially identical
with what Freud (1900) called secondary process thinking or Martindale
(1990) called conceptual thinking.
In contrast, the unconscious operates according to the principle of Eros,
which is irrational, free-associative, intuitive, and dedifferentiated. Eros is
about the same as what Freud called primary process thinking and Martindale
called primordial cognition. The unconscious is characterized by often femin-
ine traits that a man has suppressed or not developed. What Jung calls the
collective unconsciousness is populated by what he called archetypes. Arche-
types may be defined as a priori, inborn propensities that humans have to
imagine certain universal symbols or motifs that are not the result of the indi-
vidual's personal experience (Jung, 1943). Jung argued that certain symbols
have recurred throughout the history of humankind and across all cultures.
The actual archetype (called the “archetype-as-such”) is not a symbol or im-
age, but a predisposition to form certain symbols or to perceive in particular
ways. According to Jung, these predispositions rather than the specific sym-
bols that represent them are inherited (Jung, 1956). He held that our brains
have evolved with certain “psychic aptitudes” (Jung, 1935, p. 190) that func-
tion much like instincts.
Instinct is defined by William James (1890) as “the faculty of acting in
such a way as to produce certain ends without foresight of the ends and
without previous education in the performance" (p. 383). These psychic
aptitudes are adaptive because they provide us with appropriate reactions to
typical, fundamental experiences that have confronted humans throughout the
evolutionary history of the species. Jung stressed that the idea of archetype is
not meant to denote an inherited idea but rather an inherited mode of psychic
functioning (Jung, 1949, p. 518).
That our minds are built to work in specific ways is not an idea originated
by Jung. In his definition of archetypes, Jung (1921, 1954) cited Kant's use of
206 / Aesthetics and Innovation
the term, "Urbild" (which could be translated as archetype). In his Critique of
Pure Reason Kant (1781) argued that there are a priori categories of know-
ledge that are independent of experience and that are universal. Psychologists
have described human perceptual and cognitive processes in ways similar to
archetypes. Wertheimer's (1938) contention that there are "ideal types" of
stimuli that act as anchoring points for perception was the basis for Rosch's
(1975) work on prototypes that are highly typical "best examples" (Rosch,
1975, p. 544) of a category in terms of which other members of the category
are defined. Rosch and Mervis (1975) hold that some categories such as col-
ors, forms, and facial expressions have a physiological basis and that the pro-
totypes of these categories are present before the category is formed. Rosch
(Heider, 1972; Heider & Oliver, 1972) has found that the prototypes of colors
are the same cross-culturally. Cantor and Mischel and others (Cantor & Mis-
chel, 1979; Deaux & Kite, 1993; Isen, Niedenthal, & Cantor, 1992) take
Rosch's work one step further in their research on person prototypes. They
have provided evidence that people have standard expectations about the
traits composing different personality types and that new information ob-
tained about an individual is compared to these standard person prototypes.
Although Cantor and Mischel do not theorize about the origin of these proto-
types, their description of how the prototypes work is consistent with Jung's
description of the way that archetypes work. Cantor and Mischel essentially
rediscovered aspects of what Jung called archetypes.
One of the characteristics of an archetype that is not mentioned by other
writers who have proposed similar concepts is that archetypes are accompan-
ied by a highly charged emotional state (Jung, 1964). The archetype-as-such
is comparable to a Kantian category of knowing connected with energy or
emotion that is experienced as a particular image or symbol. There are a
number of archetypes, all of which produce emotional reactions. Each arche-
type can be described as a universal prototype (Hall & Nordby, 1973). Al-
though all individuals have the same set of archetypes, the conscious image
or symbol that is invested with emotional energy will differ for each indi-
vidual. Von Franz (1981) argues that both instincts and archetypes are similar
across all human beings. However whereas instincts involve physical behavi-
ors, archetypes are mental inclinations. She also maintains that each arche-
type has an instinct as its counterpart. Thus, archetypes can be described as
the instincts of emotional life. Corresponding to the sexual instinct, for ex-
ample, is the experience of falling in love. It is universal (Buss, 1994) or at
least relatively common for people fall in love or become infatuated;
however, the love object or the image of the ideal beloved differs greatly
from one individual to another. The experience of failing in love and the
Women and Animas / 207
emotion that is felt when in this state are produced by the archetype-as-such.
The image or love object (designated as the actualized archetypes or ar-
chetypal symbol) has been influenced by the individual's experience and

The Anima Archetype

Jung described a number of archetypes, including the mother, trickster,
shadow, divine child, hero, wise old man or woman, the anima, and the anim-
us. Of these, Jung wrote most prolifically about the anima archetype. The an-
ima is defined as the feminine element in a man that compensates for or mir-
rors his masculine consciousness (Jung, 1935). In order fully to understand
the meaning of this definition, we must look briefly at Jung's ideas about per-
sonality development and the unconscious.

Personality Development

According to Jung, psychological development involves differentiation.

During the first half of life, it centers on the formation of a stable ego. This
involves the development of characteristics that the individual consistently
accepts and identifies as part of himself. Usually, these are traits that are also
accepted by the person's culture. Although anyone can be described by a cer-
tain set of traits, according to analytical theory, each individual has some ca-
pacity to develop all possible traits (Jung, 1921). Everyone is assumed to be
born with potentials that are incompatible and opposite of the developed,
ego-syntonic (acceptable to or in consonance with the ego) characteristics.
During the first half of life people develop traits that are associated with their
biological sex (Jung, 1935). Yet there is in each individual some potential to
develop contrasexual traits (Jung, 1921).
According to analytical theory, the ego-dystonic (alien to or unacceptable
to the ego), undeveloped potentials are still present, although they remain un-
conscious. Furthermore, because it is difficult for people to see themselves as
possessing paradoxical qualities, the more developed any characteristic be-
comes, the more unconscious and undeveloped will be its opposite (Jacobi,
Because the anima and, at least in literature, women in general, symbolize
aspects of a male’s personality that he finds alien, they are often portrayed as
evil. In a content analytic study of poetry by 42 eminent French and British
18th and 19th century poets, Martindale (1975) found that women in general
were accompanied by words with evil connotations, whereas male characters
208 / Aesthetics and Innovation
were accompanied by words with connotations of goodness. Women are not
necessarily portrayed as being evil. Rather, they travel in bad company as it
were. That is, they tend to be mentioned in sentences that contain words with
evil connotations.

Components of the Anima Archetype

Fordham (1966) delineated the three components that contribute to the

anima image created by any one man.
1. The basis of a man's anima image is his own latent, undeveloped traits.
This includes the ego-dystonic, unconscious, feminine potentials not actual-
ized by the man. Not all undeveloped potentials are specifically masculine or
2. Analytical psychologists maintain that an important component in the
formation of the individual's anima image is an innate aptitude that all men
have for perceiving women in certain ways. Jung believed that this aptitude is
a result of the evolutionary heritage of the race. Harding (1975, p. 8) de-
scribed it as the “universalized picture of woman as she has appeared through
the centuries of human experience in relation to man." This complicates the
anima in that this universalized picture will not coincide exactly with the un-
developed traits of a man.
3. The third component that contributes to an individual's anima image is
the personal experience that any particular man has had with women. How a
man perceives women will be influenced by the impressions he has gained in
his dealings with real women throughout his life. This may well be so, but the
women who have influenced a man are not going to have the same traits as
the undeveloped traits of a man or the inherited universal picture of women.
If Jungians are right about the three determinants of the anima, it is a very
contradictory and ambiguous psychic entity. This makes it difficult for the
conscious mind to grasp or understand.
Despite the differences between individuals and their cultures, there are
certain characteristics of the anima image that are hypothetically universal.
The first is what Jung (1935) has called the *historical aspect" of the anima,
which may be experienced in several ways. The man may perceive his anima
image to be "timeless” or “immortal." Often the anima is described as seem-
ing paradoxically young yet very experienced.
The second characteristic attributed to many anima figures is great wis-
dom; the man believes that the anima is far more knowledgeable than he is.
This wisdom is often connected to the attribute of immense power. The anima
and her power may be interpreted as either positive and good or negative and
Women and Animas / 209
evil (Jacobi, 1942). Whether she is perceived as good or evil depends in part
on the man's experiences with women and in part on how adamantly he de-
fends his conscious attitude. The more dogmatically one-sided a man is, the
more likely it is that the anima will appear negative and threatening (Jung,
1935); this is because she embodies parts of the self that have been left un-
developed because they were perceived as unworthy or undesirable.
Third, Fordham (1966) notes that anima figures are often connected with
earth or water; presumably this is because earth and water symbolize the un-
conscious. Other analytic writers have noted that animas are often associated
with plants or animals (Jung, 1951; Neumann, 1955) or air and other natural
elements (Jung, 1951). Thus, the anima is often associated with nature.
Another function of the anima archetypal is adaptive. The anima arche-
type compensates for the one-sided development of the ego, thereby provid-
ing psychic balance (Jung, 1928). In his work with his patients, Jung (1935)
came to the conclusion that the archetypes are compensatory; he found that
traits undeveloped or absent in the person's ego are well represented in the ar-
chetypal symbols making up the person's dreams and fantasies. Theoretically,
it is the archetypes that help to bring an individual into touch with uncon-
scious potentials. With the anima archetype, the fascinating and attractive
power of the archetype draws the man into relationships with women whom
he believes to have the attributes of his anima image. If the women to whom
he is attracted happen to possess some of the characteristics of his anima,
they will be able to bring him into touch with potentials that he has not yet

Dynamics of the Anima Archetype

As mentioned, one function of the anima archetype is to provide psychic

balance by compensating for the one-sidedness of the ego. This one-sided-
ness is a by-product of the normal development of the first half of life. Jung
maintained that the psyche is a relatively closed system in which the principle
of conservation of energy is operative (Jung, 1928). Jung's definition is con-
sistent with definitions of the principle of conservation of energy given by
Helmholtz (1891). If any psychic component loses energy, that energy can be
found in one or more other components that are either conscious or uncon-
scious. The law of conservation of energy applies to the anima archetype as
well as to the ego because both are parts of the psyche. Thus, the more energy
the conscious ego has, the less energy the archetype will have. Conversely,
the more powerful the archetype is, the weaker is the ego (Jung, 1951).
However, the ego may expend energy in suppressing or repressing uncon-
210 / Aesthetics and Innovation
scious archetypes, so that the latter end up with more energy than the ego.
This trend has been found in the relation between male characters (represent-
ing the ego) and their anima figures in Edgar Allan Poe's short stories (Mar-
tindale, 1973, 1978). As the hero gains energy in these stories, the anima fig-
ure weakens, or as the anima gains energy, the hero weakens.

Types of Anima Figures

Jungians have described a number of universal anima types which are fre-
quently found in art, myth, and literature as well as in case material from pa-
tients. Because of space limitations, only Neumann’s (1949, 1955) descrip-
tion of types of anima will be discussed.
According to Neumann, the first archetype to emerge is the *primordial
archetype.” The primordial archetype is the term Neumann (1955, p. 7) uses
for the “great complex mass” of archetypes before differentiation or frag-
mentation into the individual, separate archetypes has occurred. This arche-
type is described as ambivalent because it embodies a union of contradictory
characteristics. In this respect it is similar to the pre-ego state in which it
arises. All characteristics are equally present; none have been differentiated
or developed yet. There is a paradoxical, simultaneous presence of male and
female attributes and of negative and positive attributes. Neumann mentions
sphinxes, harpies, and bearded mother goddesses as examples of the primor-
dial archetype. These types of mythological creatures embody the confusion
and lack of differentiation characteristic of the pre-ego state because they are
composed of a combination of different animals.
As ego development begins, the elemental character takes precedence
over the primordial archetype. The elemental anima is represented mainly by
mother goddesses. Neumann analyzed the dual aspects of the archetype. He
describes the positive form of the mother like anima as nurturing, containing,
and preserving. Frequently there are connections between the elemental char-
acter and vegetation, flowers, or fruit. The positive elemental character is
most often represented by earth-mothers or fertility goddesses, such as De-
meter or Isis.
The next stage in anima development is the negative elemental character.
Neumann describes this type of character as terrible and devouring. In the
negative elemental stage, the sheltering, possessive attributes of the positive
elemental character have become stifling to the developing ego, thus inhibit-
ing its further development. Rather than being supportive of the ego, the un-
conscious is now dangerous and capable of destroying it. Symbolically,
mother figures of this type pose a threat of captivity combined with sickness,
Women and Animas / 211
dismemberment, or death. Dangerous, devouring goddesses such as Kali or
Hecate are examples of the negative elemental character.
The transformative animas follow the elemental characters in the develop-
mental sequence. It is given its name because it drives the ego toward,
growth, change, and transformation. The positive transformative character is
embodied in anima figures that fascinate the male and lure or drive him to de-
velop further. In myth this encouragement is often represented by the motif of
the night journey in which the hero travels to some distant and dangerous do-
main, rescues a “persecuted maiden” from a dangerous creature, and returns
to the world with both the maiden and some treasure; having returned, the
hero is in some sense transformed for the better. To use analytical termino-
logy, this mythological theme symbolizes the need of the ego (as represented
by the hero) to confront the unconscious (as represented by the dangerous
task) in order to get in touch with his undeveloped feminine potentials (maid-
en in distress). Neumann speaks metaphorically of the ego descending into
the unconscious during the night journey. What he seems to mean is that
there is a regression of consciousness toward a primordial or Eros-dominated
state. In content analytic studies, it has been found that primordial content
does increase as the hero “descends” into the unconsciousness and a decrease
as he returns to the world (Martindale, 1979, Martindale & West, 2002). An
example of the night journey theme that shows this pattern is Book VI of Vir-
gil’s Aeneid, in which Aeneas descends into hell, receives guidance, and re-
turns inspired to found Rome. The theme is ubiquitous in adventure stories.
However, the pattern of increasing and decreasing primordial content is only
found when the hero is in some sense transformed. For example, Joseph Con-
rad’s Heart of Darkness seems to be a perfect example of the night journey,
but does not show the expected trend in primordial content (Martindale &
West, 2002). Presumably this is because the hero merely carries out an inter-
esting job, but his personality is not transformed in any way. As well as
providing a symbolic roadmap for personality development, the night journey
may also be seen as symbolizing the creative process, which also involves a
regression and return cycle (Kris, 1952).
Positive transformative characters are often accompanied by allusions to
religion, whiteness, and virginity. They also tend to be isolated from men and
from earthy female figures. Examples include the Indian Tara, the Greek
muses, and the Virgin Mary.
The negative transformative anima is embodied by a provocative and hos-
tile female figure—but more importantly, there is still pressure for transform-
ation and change. Neumann (1955, p. 80) describes these figures as the “al-
luring and seductive figures of fatal enchantment." These destructive femmes
212 / Aesthetics and Innovation
fatales are frequently accompanied by dangerous animals, and associated
with madness, intoxication, and sexuality. Sexuality is often used as the
means to enslave or destroy the hero. Examples include Lilith, Circe, and the
There are individual differences in the sequence that a man or culture will
experience the positive and negative transformative animas. Some men may
experience both while others may only experience one type. Although they
may occur in either sequence, in most cases the negative transformative char-
acter precedes the positive because, theoretically, a rejection of one's uncon-
scious feminine potentials results in the emergence of the negative character.
Most males reject the feminine potentials more strenuously when they first
begin to surface. When there is acceptance of the feminine side of the self,
the unconscious femininity is symbolized more positively. This change in at-
titude is usually accompanied by a change from the negative to the positive
transformative anima in fantasies or myths. Often, what appeared to be a de-
structive anima is seen as a positive one after transformation as occurs in the
story of Odysseus and Circe.

Quantitative Studies of the Anima Archetype

Analytical psychologists have provided many interpretations of written

works that focus on the anima (e.g., Bruner, 1963; Neumann, 1952, 1978;
von Franz, 1978). The works of Jungians tend to be intuitive, impressionist,
qualitative interpretations. The aim in this chapter is to describe quantitative
computerized content analytic measures of the concept of the anima. We have
just seen that the anima takes many often contradictory forms. The goal is to
see if all anima figures share some common traits. The first task was to find
texts that Jungian theorists had said were about anima figures.


Anima Figures

Texts were selected from works cited as describing prime examples of the
anima archetype by Jung and his first-generation students. The citations were
obtained through a systematic search of Jung's entire collected works. The
major works of the theorists who worked directly with Jung were also
Twenty-one of the 37 texts mentioned by Jung and his followers were
studied. They are listed in Table 1.
Women and Animas / 213
Table 11-1. Citations, Authors, and Anima Characters

Literary Work Author Character

Wisdom's Daughter Haggard Ayesha

Siegfried (opera) Wagner Brunhilde
Aurelia Nerval Aurelia
L’Atlantide Benoit Atlantida
To Walk the Night Sloane Selena
Private Life of Helen of Erskine Helen
Faust Goethe Gretchen,
Helen, Mary
Song of Solomon Bible Shulamite
Orphée (film) Cocteau death demon
Magic flute (opera) Mozart Queen of the night
Paradisio Dante Beatrice
Grail Legend von Eschenbach Conduit Amour
Der goldene Topf Hoffmann Veronika
Green Mansions Hudson Rima
David Copperfield Dickens Dora
Carmen (opera) Bizet Carmen
Hamlet Shakespeare Ophelia
Forsyte Saga Galsworthy Irene
Great God Brown O'Neill Sibyl
Amor and Psyche Apuleius Psvche

All 37 texts could not be studied for a variety of reasons. For example,
several were by the same author. Works by the same author would not be in-
dependent samples and would thus compromise the validity of statistical ana-
lyses. In other cases, mention of the anima figure was far too brief for the
work to be used or there was no English translation of the work.

Female Control Characters

The control sample of non-anima female characters were characters who

had not been identified as anima figures and who appeared in the same works
as the identified anima characters. If there were a number of female charac-
ters in a text, selection of the control character was based upon two criteria.
First, she must interact with the hero enough to provide sufficient text for
214 / Aesthetics and Innovation
analysis. Second, she should be as little like an anima character as possible.
Drawing the control female sample from the same work as the anima figures
allowed maximal matching by literary genre, publication date, and so on.

Hero and Male Control Characters

In all cases it was obvious who the hero was. Selection of male control
characters was done in a way completely analogous to selection of female
control characters.


It was important to discriminate whether a character was the subject or

object of an action. According to Jung's theory, it is not so important that the
hero attract the anima (he could have absolutely no impact on her); but, it is
almost imperative that she attract him. A method had to be devised to indicate
who did what to whom. Separate documents were set up for each character
alone, and separate documents were created for each character's interactions
with every other relevant character defined by who the subject was and who
the object of an action was. Thus, for every work there could be up to 20 doc-
uments as follows:
1. Anima
2. Anima as object
3. Hero
4. Hero as object
5. Control Female
6. Control Female as object
7. Control Male
8. Control Male as object
9. Anima as subject acting on hero as object
10. Anima as subject acting on control female as object
11. Anima as subject acting on control male as object
12. Hero as subject acting on anima as object
13. Hero as subject acting on control female as object
14. Hero as subject acting on control male as object
15. Control Female as subject acting on anima as object
16. Control Female as subject acting on hero as object
17. Control Female as subject acting on control male as object
18. Control Male as subject acting on anima as object
19. Control Male as subject acting on hero as object
20. Control Male as subject acting on control female as object
Women and Animas / 215
Not every work had all 20 documents. This arrangement of data is com-
plex but can accommodate a variety of analyses by combining documents in
various ways. A considerable amount of text in each work was not relevant to
any character in particular. The clearest example would be a paragraph de-
scribing the setting without mention of any characters. Words that were not
directly relevant to any focal character and irrelevant words (e.g., articles,
pronouns) were not put into any of the documents.

Application of Content Analysis to Anima Measurement

Regressive Imagery Dictionary

The first question of interest is whether archetypal figures and female

characters in general symbolize the unconsciousness or Eros-dominated
states of consciousness. The Regressive Imagery Dictionary (Martindale,
1975, 1990) is a well validated measure of primordial content. We would ex-
pect female and archetypal characters to occur in the context of such content.
The dictionary was applied to the texts. Results were as predicted. In a 2
(Gender) x 2 (Numinosity) x 2 (Interaction Mode) ANOVA, main effects
were found for Gender and Numinosity on primordial content, which essen-
tially measures what Jung called Eros. The Gender main effect F(1,19) =
20.12, p < .0005, indicated that text describing female characters (M = 11.68)
had more primordial content than text describing male characters (M = 3.37).
The Numinosity main effect F(1,18) = 13.75, p < .005, indicated that the nu-
minous characters (hero and anima) were described with more primordial
content (M = 11.36) than were the control characters (M = 3.27).
When just the female characters were the focus of study, a similar effect
was found. In a Female Type (Anima versus Control) x Interaction Mode
(Character as Subject versus Character as Object) analysis of variance, main
effects were found for female type on primordial content. The animas were
associated with higher use of words denoting primordial content. The primor-
dial content mean for anima figures was 4.23; for control females it was
-1.66. Primordial content is measured by subtracting the percentage of words
indicating conceptual content from the percentage of words directly measur-
ing primordial content, so negative scores are possible.
The results of the primordial content scores from the Regressive Imagery
Dictionary showed that, of the characters in these works, women were associ-
ated with more primordial content than were men. This is interesting in that it
supports the beliefs of analytic writers that the feminine is more closely
216 / Aesthetics and Innovation
linked than is the masculine with the unconscious (Harding, 1975; Jung,
1916, 1954; Neumann, 1949, 1955).
The anima and animus characters had more primordial content when
compared to the control characters. This indicates that, as expected, the ar-
chetypal figures are closer to the unconscious.

Creation of an Anima Dictionary

The next step was to construct a content analytic dictionary to see if Jung
and his followers were correct about the specific traits that they attributed to
the anima. The first task in dictionary construction is the selection of categor-
ies that define the constructs one wishes to measure and the compilation of
lists of words that indicate the categories. The categories contain independent
word lists; for statistical reasons, no word should appear in more than one
category. Based upon what Jung and his followers said about the anima a
number of categories were constructed that should be relevant to the anima in
one way or another. The categories, along with sample words for each are
shown in Table 2 and are described below.

Table 11-2. Anima Dictionary Categories and Sample Words

Category Title Sample Words

Energy glow, luminescent, lustrous, magnetic, starry

Positive Numinosity desirable, fairy, nimbus, serene, transform
Negative Numinosity ghostly, repellent, odious, seductive, volup-
Wisdom foresight, guide, intelligent, prophetess,
Timelessness ancient, enduring, fated, undying, universal

Table 11-2. Continuation

Mysteriousness bewilder, illusive, strange, unfathomable,

Perfection/Purity undefiled, integrity,, symmetry, trustworthy,
Touch brush, caress, cling, hold, kiss
Body Parts arm, bosom, flesh, lip, shoulder
Blood blood, vein
Women and Animas / 217
Blood Adjective boil, hot, pulsate, sped, surge
Heart heart, heartbeat, heartrate, pulse
Heart Adjective broke, leap, pulsate, rend, stir
Breathe breath, breathe
Breathe Adjective difficult, hard, labor, short
Breathless gasp, heave, heavy, stammer
Chill cold, goosebump, shake, shiver
Faint collapse, dizzy, giddy, stuporous, swoon
Flush blush, fevered, flush, fluster, redden
Pole ashen, blanch, pole, sallow, wan
Sweat febrile, perspire, sudorous, sweaty
Tremble fidget, flutter, quiver, squirm, stagger
Positive Animal dolphin, gazelle, kitten, peacock, rabbit
Neutral Animal elephant, goat, groundhog, monkey, pony
Negative Animal crocodile, gargoyle, locust, lynx, pig
Gems agate, brass, crystal, jewel, ruby
Spices aromatic, cinnamon, clove, perfume, thyme
Flowers bloom, bouquet, daisy, magnolia, primrose
Gaze behold, gape, leer, linger, stare
Lover attentive, languish, love, spellbound, woo
Beloved darling, disquieted, fascinate, impassion, rile
Musicality euphonic, harmony, lyrical, sing, trill
Miscellaneous Anima curious, dainty, intimate, ivory, silky
Anti-Anima boring, eyesore, hag, shriveled, skinny
Caring/Gentle gracious, gratify, sentimental, succor
Strength/Heroic competent, noble, overcome, unflinching
Control/Cruelty allow, bode, capricious, embarrass, mali-
Object of Control abhor, kneel, powerless, suffer, timid

One group of categories was intended to categorize the core characterist-

ics of the anima; they have category titles such as Wisdom, References to An-
imals, Numinosity, Mysteriousness, Energy, Historical References/Timeless-
ness, and Perfection/Purity. Additional categories were created that classified
references to Gems, Spices, and Flowers—which Jungian writers have more
marginally discussed as related to anima figures.
Another group of categories defined different aspects of physiological re-
sponses. These categories were meant to function as measures of affect. Jung
stated that the intensity of an archetype such as the anima is reflected by the
degree of affect experienced in relation to ft. In many cases the author of a
218 / Aesthetics and Innovation
story does not have the hero directly state the emotion hat he is feeling. In-
stead, physiological symptoms are reported to describe the type and amount
of emotion that he is experiencing. Therefore, ft is appropriate to include cat-
egories that code references of this type. The physiological categories meas-
ure Breathlessness, Chills, Faintness, Flushing, Paleness, Sweat, and Trem-
Several themes more generically related to love relationships were con-
sidered to be sufficiently relevant to the anima construct to warrant inclusion.
Words used to denote behavior or feelings common to lovers (e.g., "spell-
bound” were included in the Lover category. Similarly, words indicating how
the beloved might impact the lover or terms of endearment for the beloved
(e.g., "darling," "impassion”) comprised the Beloved (object of love) cat-
egory. A category was designed to code words specific to the behavior of
Gazing. Additionally, several categories were included that, although not dis-
cussed as particular characteristics of animas or their lovers, seemed poten-
tially fruitful for exploratory analysis. These were: Control/Cruelty, Object of
Control, Strength/Heroism, Caring/Gentleness, Evil, and Good. The Good
and Evil categories are the only categories that are not independent of the
other categories. These summary categories were designed to include as com-
prehensive a list as possible of words consistent with good and evil. The Evil
category includes the Control/Cruelty category, the Negative Animal cat-
egory, and the Negative Numinosity category. The Good category includes
the Caring/Gentleness category, the Positive Animal category, the Positive
Numinosity category, and the Musicality category.
Two categories were designed to classify words that appeared in relation
to anima characters. One of these categories tallied miscellaneous words used
to describe anima characters. It also included examples of the colors that
Jungian writers had linked to anima description. This was the Miscellaneous
Anima category. Second, a Musicality category was constructed. Last, a cat-
egory was created that was parallel to the Miscellaneous Anima category.
This included words that appeared in conjunction with characters that were
clearly not anima characters (Anti-Anima category).


A number of analyses were done to determine what categories were able

to discriminate anima figures from non-anima figures and, correspondingly,
other characters' reactions to anima versus non-anima figures. The means
given below refer to the proportion of total words in a document falling into
the category under discussion.
Women and Animas / 219
ANOVAs Relevant to Discrimination of Characters

Gender by Numinosity Anova. The first ANOVA included the factors of

gender and numinosity. The hero and the anima were defined as numinous;
control male and female as non-numinous. This analysis was done on the
characters-as-subjects and the characters-as-objects data sets.
When characters were the object of an action a significant Gender x Nu-
minosity interaction was found for the Mysteriousness category, F(1, 16) =
5.13, p < .05. Examination of the means indicated that the heroes (M = .22)
are less subjected to mysteriousness than are the control males (M = .38).
More importantly, animas (M = .40) are more the object of mysteriousness
than are control females (M = .23). This is an unexpected finding, since the
theory is that the anima is mysterious. However, results show that the anima
is subjected more to mystery than any of the characters. This may be due to
the fact that some of the animas were priestesses or otherwise in touch with
otherworldly powers.
When characters are the objects of actions, a significant main effect is
also found for Numinosity on the Paleness category, F(1,18) = 5.58, p < .05.
These results indicate that the hero and anima characters (M = .02) were more
often described as drawing a response of paleness (other characters pale in
the presence of the hero and anima characters) than were the control charac-
ters (M = .00).
When characters were the subjects, a Gender x Numinosity interaction
was found for the Gaze category, F(1,8) = 5.71, p <.05. Although control
males (M = .89) gaze more than heroes (M = .58), the greatest difference was
between the animas (M = .73) and the control females (M =.03). A number of
the animas use gaze as a “weapon” or means of control of others. The Gaze
category was also significant for Gender when characters were objects of ac-
tions, F(1, 19) = 7.04, p < .05. Male characters (M = .55) were gazed at more
often than were female characters (M = .29). When characters were subjects,
a significant main effect for Numinosity was found for mention of Body
Parts, F(1, 15) = 7.57, p < .05. Numinous characters had a mean of 2.11,
whereas controls had a mean of 1.05. An author is more likely to discuss the
physical characteristics of the hero and anima figures than of the control
characters. A significant main effect was also found for the category com-
posed of adjectives describing the movement of Blood, F(1,15) = 6.46, p <
.05. Numinous characters had a mean of .27, and control characters had a
mean of .08. For words having to do with Musicality, an interaction of
Gender x Numinosity was found, F(1,8) = 25.39, p < .001. Anima figures had
a mean of .31; all other cells had a mean of 0.00.
220 / Aesthetics and Innovation
Female Type by Hero Interaction Mode ANOVAs. Two 2 x 2 ANOVAs
were done to compare the anima characters to the control female characters.
The first was a Female Type (Anima versus Control) x Hero Interaction
Mode (Hero as Subject versus as Object). In this ANOVA, it was possible to
determine whether the animas and control females differed in how they re-
lated to the hero and also whether they were treated differently by the hero.
An interaction was found for the Caring category, F(1,13) = 7.72, p < .05.
Control females relating to heroes are described with relatively few caring
words (M = .05) while anima figures are quite caring to heroes (M = 1.85).
Rather surprisingly, heroes are more caring toward the control females (M =
1.78) than they are to anima figures (M = .47). It is not clear why heroes
would express more caring to the control females; one possibility is that the
control females are more approachable.
The same type of interaction was significant for the Goodness category,
F(1, 13) = 7.90, p < .05. Anima figures were more associated with goodness
in relation to the hero, (M = 5.23) than were the control females (M = .36).
The heroes' goodness scores when interacting with animas were not vastly
different from the heroes' goodness scores when interacting with controls, but
the difference was weighted more toward the hero exhibiting goodness in re-
lation to control females. With the controls, M = 3.64; with the animas M =
2.91. These results are surprising because, based on theory, the heroes should
be more attentive to anima figures than to control characters.
A third interaction was found for the Energy category, F(1,13) = 6.68, p <
.05. Control females are virtually unenergetic when interacting with heroes
(M = 0.00), though heroes have some energy in that relationship (M = .13).
When they are with anima figures, heroes have more energy (M = .22) but the
most energetic are the anima figures when relating to the heroes (M = .55).
An identical pattern of results was found for the Positive Numinosity cat-
egory, F(1,13) = 6.19, p < .05. The Positive Numinosity means were 0.00,
1.70, 2.29, and 3.38, respectively.
In this analysis, main effects were found for Negation and for Trembling.
The Trembling category was significant, F(1,14) = 14.23, p < .01, with anima
figures trembling more (M = .31) than the control females (M = 0.00). The
Negation category was highly significant, F(1,14) = 11.58, p <.005, with an-
ima figures being described by negated words more frequently (M = 1.31)
than control females (M =.35). The anima is more often described indirectly
by what she is not rather than by what she is. It might be said that words can-
not describe the anima, so authors must content themselves with negating
available words.
Women and Animas / 221
Female Type by Female Interaction Mode ANOVA. The second 2 x 2 AN-
OVA was more general than the first. It was a Female Type (Anima versus
Control female) x Female Interaction Mode (Subject versus Object) ANOVA
that combined all documents in which the anima was the subject (acting on
any characters including the hero) versus the combination of all documents in
which the anima was the object. Analogous cells were created for the control
The ANOVA produced results consistent with the second 2 x 2 ANOVA
on the Goodness, Negation, and Positive Numinosity categories. In this more
global analysis, the Goodness category produced a significant interaction,
F(1,16) = 10.96, p < .005. There are more good words used when anima fig-
ures are the subjects (M = 3.76) than when the control females are the sub-
jects (M = 1.86), but the control females are the objects of goodness (M =
2.99) more than the anima figures (M = 1.83). A main effect of Negation was
significant in this analysis, F(1, 19) = 15.70, p < .001. The anima figures
were more often described by negating characteristics (M = 1.15) than were
the control females (M = .48). Positive Numinosity was again significant,
F(1,16) = 8.11, p < .05; the anima figures are clearly the most numinous (M =
2.59) and the least likely to be exposed to numinous others (M = 1.17). The
control females received numinosity scores yielding a mean of 1.43 and as
objects of numinosity had a mean of 1.77. This indicates that they are more
often the objects of others' numinosity than they are numinous themselves.
In this analysis an interaction was found for the Negative Animal cat-
egory, F(1, 16) = 5.65, p < .05. Anima figures as subjects (M = .09) are more
often associated with negative animals than are the control females as sub-
jects (M = .04). Animas are rarely in contact with others associated with neg-
ative animals (M = .01) whereas the control females are much more fre-
quently involved with others related to negative animals (M = .08) than they
themselves are involved with such animals (M =.04).

Composite Measures.

Because the significant results did not include many of the elements that
are, according to analytical theory (Fordham, 1966; Jacobi, 1942; Jung
1935,1951, 1954; Neumann, 1955), the essence of the anima (e.g., Purity,
Timelessness, and Wisdom), a formula based on the theoretical considera-
tions was created. The formula included scores on categories measuring: 1)
Interactions between female and male characters: male characters gazing at
the female, mention of the female's body, contact of the female's body with
the hero, female responses to the male involving breathing, heart, or blood,
and male responses to the female involving breathing, heart, or blood; and 2)
222 / Aesthetics and Innovation
theoretical traits of the anima: Negative or Positive Numinosfty, Wisdom,
Timelessness, Mysteriousness, Perfection/purity, and the score on the Nega-
tion category.
For purposes of comparison, an empirical composite formula was also
computed. To create an empirical measure of the anima, only the categories
which yielded significant differences between the two types of characters
were used. Categories that describe only the anima character herself were
combined with categories that involved other characters' reactions to her
when these categories yielded significant results.
Categories describing the anima character herself were Negation, Body
Part, Blood Adjectives, Positive Numinosity, Caring, Energy, Trembling,
Mysteriousness, Blood reaction (with the anima as object), Paleness (with the
anima as object), Musicality, Negative animals, and Goodness. Additional
categories added into the score were: male characters' Mysteriousness, En-
ergy, and Paleness in relation to the females, and male characters' scores on
the evaluation category.
Analyses of variance were done on the sample to test the efficacy of the
two formulas discriminating between anima figures and control females. In
an Anima Level (Anima versus Control) x Text (Literary Work) ANOVA the
theoretical anima formula and the empirical anima formula both produced
highly statistically significant results; the theoretical anima produced F(1, 17)
= 27.98, p <.0001, with an anima-figure mean of 26.67 and a control female
mean of 15.08. The empirical anima formula produced F(1,17) = 30.29, p
<.0001 with an anima figure mean of 24.88 and a control female mean of
13.87. As can be seen, the theoretical and empirical formulas produced very
similar results. The success of the theoretical formula, even though many of
its components were not statistically significant on their own, suggests that
the traits of the anima may combine in a multiplicative way. If they are all
present, the degree to which they are present is not of consequence. Taken to-
gether, they indicate the presence of an anima. There is no need for an author
to emphasize or overstate them.

Bruner, C. (1963). Anima as fate. Dallas, TX: Spring Publications, 1986.
Buss, D. M. (1994). The evolution of desire: Strategies of human mating.
New York: Basic Books.
Cantor, N. & Mischel, W. (1979). Prototypicality and personality: Effects on
free recall and personality impressions. Journal of Research in Personal-
ity, 9, 187–205.
Women and Animas / 223
Deaux, K., & Kite, M. (1993). Gender stereotypes. In F. Denmark & M.
Paludi (Eds.), Psychology of women: A handbook of issues and theories.
Westport, CT: Greenwood Press.
Fordham, F. (1966). An introduction to Jung's psychology. Baltimore, MD:
Penguin Books.
Franz, M. L. von (1978). The interpretation of fairytales. Dallas, TX: Spring
Franz, M. L. von (1981). Puer aeternus. Santa Monica, CA: Sign Press.
Freud, S. (1900). The interpretation of dreams. New York: Avon Books,1964.
Hall, C. S., & Nordby, V. J. (1973). A primer of Jungian psychology. New
York: Mentor.
Harding, M. E. (1975). The way of all women. New York: Harper.
Heider, E. R. (1972). Universals in color naming and memory. Journal of Ex-
perimental Psychology, 23, 10–20.
Heider, E. R., & Oliver, D. C. (1972). The structure of color space in naming
and memory for two languages. Cognitive Psychology, 3, 337–354.
Helmholtz, H. von (1891). Popular lectures on scientific subjects. New York:
D. Appleton.
Isen, A., Niedenthal, P. M., & Cantor, N. (1992). An influence of positive af-
fect on social categorization. Motivation and Emotion, 16, 65–78.
Jacobi, J. (1942). The psychology of C. G. Jung. New Haven, CT: Yale Uni-
versity Press, 1962.
Jacobi, J. (1965). The way of individuation. New York: Harcourt, Brace, &
World, 1967.
James, W. (1890). Principles of psychology. New York: Henry Holt.
Jung, C. G. (1916). Symbols of transformation. In H. Read, M. Fordham, &
G. Adler (Eds.), The collected works of C. G. Jung (Vol. 5). Princeton, NJ:
Princeton University Press, 1956.
Jung, C. G. (1921). Psychological types. In H. Read, M. Fordham, &.G.
Adler (Eds.), The collected works of C. G. Jung (Vol. 6.) Princeton, NJ:
Princeton University Press, 1976.
Jung, C. G. (1928). On psychic energy. In H. Read, M. Fordham, & G. Adler
(Eds.), The collected works of C. G. Jung (Vol. 8, pp. 3–66). London:
Routledge & Kegan Paul, 1960.
Jung, C. G. (1935). The relations between the ego and the unconscious. In H.
Read, M. Fordham, & G. Adler (Eds.), The collected works of C. G. Jung
(Vol. 7, pp. 123–241). Princeton, NJ: Princeton University Press, 1972.
Jung, C. G. (1943). On the psychology of the unconscious. In H. Read, M.
Fordham, & G. Adler (Eds.), The collected works of C. G. Jung (Vol. 7,
pp. 3–119). Princeton, NJ: Princeton University Press, 1972.
224 / Aesthetics and Innovation
Jung, C. G. (1949). Foreword to Harding: Wornan's mysteries. In H. Read,
M. Fordham, & G. Adler (Eds.), The collected works of C. G. Jung, (Vol.
18, pp. 518–520). Princeton, NJ: Princeton University Press, 1976.
Jung, C. G. (1951). Aion. In H. Read, M. Fordham, & G. Adler (Eds.), The
collected works of C. G. Jung (Vol. 9). Princeton, NJ: Princeton Uni-
versity Press,1959a.
Jung, C. G. (1954). Archetypes of the collective unconscious. In H. Read, M.
Fordham, & G. Adler (Eds.), The collected works of C. G. Jung (Vol. 9,
pp. 3–41). New York: Pantheon, 1959.
Jung, C. G. (1956). Jung and religious belief. In H. Read, M. Fordham, & G.
Adler (Eds.), The collected works of C. G. Jung (Vol. 18. pp. 702–744).
Princeton, NJ: Princeton University Press, 1976.
Jung, C. G. (1964). Approaching the unconscious. In C. G. Jung (Ed.). Man
and his symbols (pp. 1–94). New York: Dell.
Kant, I. (1781). Critique of pure reason. New York: The Modem Library,
Kris, E. (1952). Psychoanalytic explorations in art. New York: International
Universities Press.
Martindale, C. (1973). Transformation and transfusion of vitality in the narra-
tives of Poe. Semiotica, 7, 46–59.
Martindale, C. (1975). Romantic progression: The psychology of literary his-
tory. Washington, DC: Hemisphere.
Martindale, C. (1978). A quantitative analysis of diachronic patterns in some
narratives of Poe. Semiotica, 22, 287–308.
Martindale, C. (1979). The night journey: Trends in the content of narratives
describing transformation of consciousness. Journal of Altered States of
Consciousness, 4, 321–343.
Martindale, C. (1990). The clockwork muse: The predictability of artistic
change. New York: Basic Books.
Martindale, C., & West, A. N. (2002). Quantitative hermeneutics: Inferring
the meanings of narratives from trends in content. In W. van Peer & M.
Louwerse (Eds.), Thematics: Interdisciplinary studies (pp. 377–396). Am-
sterdam: John Benjamins.
Neumann, E. (1949). The origins and history of consciousness. Princeton,
NJ: Princeton University Press, 1970.
Neumann, E. (1952). Amor and Psyche: The psychic development of the fem-
inine. Princeton, NJ: Princeton University Press, 1971.
Neumann, E. (1955). The great mother: An analysis of the archetype. Prin-
ceton, NJ: Princeton University Press, 1972.
Neumann, E. (1978). The magic flute. Quadrant, 11, 5–32.
Women and Animas / 225
Rosch, E. (1975). Cognitive reference points. Cognitive Psychology, 7, 532–
Rosch, E., & Mervis, C. B. (1975). Family resemblances in the internal struc-
ture of categories. Cognitive Psychology, 7, 573–605.
Wertheimer, M. (1938). Numbers and numerical concepts in primitive
peoples. In W. D. Ellis (Ed.), A sourcebook of Gestalt psychology. New
York: Harcourt-Brace.

Translation of Values through Art:

Non-Classical Value Approach

Dmitry A. Leontiev

The chapter describes the theoretical foundation and empirical realization of

the non-classical value approach to art. The non-classical value approach is
based on the presumption that creation and perception of artworks are the
processes of the translation of values from individuals to embodied artworks
and from embodied artworks to individuals. Value Spectrum Technique elab-
orated by the author makes it possible to prove the reality of value translation
processes and opens a perspective of studying value content of artworks and
its impact upon the recipients’ personality.

The chapter is intended to summarize a number of results of the non-clas-

sical value approach—a theoretical and methodological research approach
elaborated by the author a decade ago. The approach far exceeds the field of
psychology of art and empirical aesthetics; however, it is this field where the
approach has found its most numerous and convincing applications to date.
Separate research findings have been published elsewhere (e.g., Leontiev,
Delskaya, & Nazarova, 1996; Leontiev & Yemelyanov, 1996); however, they
have not yet received a systematic overview. In this paper I shall present the
theoretical framework, research method, and empirical results obtained in the
studies of perception of painting, music, theatre, poetry, and animation. The
non-classical value approach applications to the fields of psychology of ad-
vertising, personality development and transformations, and delinquent beha-
vior are still waiting for detailed analysis.

Non-classical value approach / 227
Protean Nature of Values
The value concept belongs to the interdisciplinary concepts of human sci-
ences; it is considered one of the most important concepts in philosophy, so-
ciology, psychology, economics, anthropology, etc. However, it is understood
not only among different disciplines from this list, but also within each of
them in quite different, often contradictory manners, both at the level of
definitions and at the level of colloquial word use. We cannot present here the
complete overview (see Leontiev, 1996a); instead, we summarize the existing
contradictions in six dichotomies regarding the value concept.
A. Value as an attribute that something has versus value as an object that
something is. In the first case the word is used as a synonym of concepts like
significance, valence, or personal meaning, hardly adding anything new to
them; in the second case it seems more capable to play a heuristic role in em-
pirical research.
B. Value as a real single significant or useful object versus value as ideal
(abstract) object that has a special status to be considered further. The first
version we meet in economics and utilitarian philosophy where value-object
appears as a secondary concept still to be explained through superordinate es-
sences. The second version is compatible with the intuitive understanding of
value as something superordinate. The point is to throw some light on its
C. Value as individual reality versus value as superindividual reality. In
the first case value can be treated either as a synonym of individual signific-
ance, revealed by the individual mind. In the second case values are treated
as existing beyond and before individual mind. Basically, there seem to be no
contradiction: there are enough evidences to recognize that values belong
both to the realm of individual psyche and to some broader realm. The real
problem, central for the non-classical value approach, consists in explaining
the relations between the two aspects.
D. The recognition of existence of values outside and beyond individual
mind leads to the following alternative: either the values are to be treated,
like in some philosophical approaches, as transcendent essences subduing to
special laws different from the laws of the earthly world, or as the products of
social communities that appear as the primary subjects of values. In the first
case values are to be excluded from the field of scientific analysis, including
human sciences. In the second case we are dealing with a supraindividual, but
not suprahuman phenomenon; the concept of value becomes the key concept
for understanding the complicated relationships between individual and so-
cial realities.
228 / Aesthetics and Innovation
E. Within the understanding of value as a phenomenon of individual psy-
chology one more opposition can be seen: whether the values are treated as
mere mental representations or as deeper structures linked to motivational dy-
namics. Here again both variants are not mutually exclusive and real prob-
lems are those of differentiation between the value representations and the
value dynamics and of understanding the relationships between the both
F. The last opposition found in the multitude of different approaches to
values is that of a value as a standard versus a value as an ideal. In the first
case value is treated as a norm, something precisely articulated to be pre-
cisely followed; in the second case as a meaningful purpose or ideal, giving
general direction to individual activity rather than exact guidelines.
In line with the idea of dimensional ontology (Frankl, 1985) we treat all
the inconsistent and seemingly incompatible views on value as various two—
dimensional projections of a complicated 3—(or more) dimensional object.
Our task is thus to reconstruct the multidimensional object in such a way that
makes all the existing projections intelligible. To solve this task, we have to
assume that value is a borderline phenomenon that bridges individual to so-
cial reality; it has several forms of existence, being in constant transitions and
transformations from one form to another. The nature of these transforma-
tions may be understood through the idea of converted form, introduced by
the Russian philosopher Merab Mamardashvili (1970), to denote the pro-
cesses of transition of some content from one substrate to another substrate.
The features of the content do change in course of this transition according to
the properties of the substrate. An illustration can be borrowed from the psy-
chology of art: when you try to transform a novel into a movie script, even if
you plan to maintain the content as close as possible to the original work of
literature, you can’t do it without some important changes. Indeed, the sub-
strate (the film) imposes some limitations and offers some new possibilities;
the properties of the new substrate make some conversion inescapable. And
in every content we may distinguish between its elements depending on the
original form and those depending on the actual substrate.
The primary form of value is a social ideal, elaborated by a social unit
that may range from a family to the humanity at large. This or that social unit
is the primary subject of every value; every value is a condensed expression
of group experience in form of an idea of the desirable, the required, the per-
fect in this or that field of social activity.
The second form is value objectified in human deeds and artifacts. In-
deed, we get the idea of beauty not from the abstract concept of beauty but
rather while enjoying artworks, landscapes, or humans embodying this value.
Non-classical value approach / 229
The idea of justice is acquired when we know the examples of just and unjust
deeds. These embodied values, however, are not self-sufficient phenomena; it
is social ideals that give artifacts and deeds their value status.
The third form is personal values, which make a substantial part of per-
sonality structure. Indeed, the embodiment of value in deeds and artifacts is
possible only in human activity motivated by the corresponding values. Abra-
ham Maslow (1976) in his metamotivation theory presented numerous evid-
ences of dynamic effects of values, though he restricted these effects to the
highest level of personality development, the level of Being. In fact, the dy-
namic effects of values can be seen also at the lower level of Becoming,
where basic needs still maintain all their motivating power.
From this viewpoint values may be treated as the key mechanism of hu-
man socialization, bridging individuals to social communities within which
they live and develop. Every value finds itself in constant transitions between
the three forms of its existence: the form of group social representation; the
form of artifact or deed, and the form of dynamic personality structure. This
Protean view on value is consistent with both the majority of theories of
value in philosophy, social sciences, psychology, and common sense, helping
to put aside many contradictions in definitions. We call it non-classical to
locate it within the general framework of Vygotskian methodology.
The term “non-classical psychology” was coined not long ago by Daniel
Elkonin (1904—1984), a Russian developmental psychologist and a close
friend and disciple of Lev Vygotsky. In a paper given shortly before his death
in 1984 and published posthumously (Elkonin, 1989), he called Vygotsky the
founder of non-classical psychology. Non-classical psychology was defined
as “the science of the way the subjective world of a single person emerges
from the objective world of art, the world of production tools, the world of
the entire industry” (p. 478). Unlike all the “classical” psychology, in Vygot-
skian theoretical perspective mental states and processes are viewed as loc-
ated not only within individual’s mind. In opposition to the Cartesian view
which assumed the principal borderline between internal and external (the
whole mental life being definitely located “within”) Vygotsky stated that
mental contents and processes do exist in extracerebral and extracorporeal
forms, outside individual mind, in the world of human artifacts, cultural sign
structures, human-made environment, and interpersonal communication, pri-
or to their intraindividual functioning. There are the transitions from one
form to another, from the objectified form of existence of human mental phe-
nomena to the subjectified forms of human mental processes. One of the ex-
amples is Vygotsky’s “Psychology of Art” (Vygotsky, 1971), where he clearly
stated that human mental processes and contents are objectified in the struc-
230 / Aesthetics and Innovation
ture of artworks. For Vygotsky, the process of artistic creation is a transform-
ation of human feelings and emotions from one form of existence, a subject-
ive one, to another, objective one. The reverse movement takes place in the
process of art perception (for a detailed explication of Vygotsky’s approach to
art see Leontiev, 1997; Sobkin & Leontiev, 1992). Elkonin noted that it was
essentially the preliminary formulation of the general developmental prin-
ciple, later elaborated by Vygotsky in a broader context: from interpsychic to
intrapsychic (Vygotsky, 1983).
In line with non—classical views, we assume that due to their threefold
nature values serve as mediators and ties between the realms of collective
representations, of human—made artifacts and individual dynamics (behavior
regulation). Humans in the course of socialization share the collective activ-
ity of a social group, being engaged in a joint, collectively distributed activity
(Leontiev, 1992). In this process the group value ideals may be gradually per-
sonalized and internalized by the individual and become incorporated into
one’s motivational system (Leontiev, 1996b; 1997). It is due to these indi-
vidual values that the individual strives to externalize them again in the form
of artifacts or deeds exemplifying this or that value, thus bringing them again
into the social, collective, public realm. What seems the most important is
that we are speaking of one value changing it’s forms of existence (social,
personal, embodied) in the course of such circulation in line with the prin-
ciple of the converted form (Mamardashvili, 1970). We speak thus of a sort
of value metamorphosis. A certain value, say, charity, may emerge in a person
as a dynamic entity only if the person identifies oneself with a social group
oriented toward this value as a group ideal be it a real group, like one’s fam-
ily, or virtual group, like Christians, meets the examples of the embodied
value (acts of charity) and starts implementing it in his/her activity.
Different social artifacts or events may embody various values to a great-
er or smaller degree. In fact, all the social symbols, artifacts, organizations,
etc. bear the traces of a number of values; that is why they resonate so much
to individual motivational structure. Though people are very rarely aware of
these value attributes, the latter are somehow perceived and influence indi-
vidual attitudes toward these artifacts, symbols, etc. However, these value
components can be revealed through the procedure of value attribution that is
asking the individuals to define which values of the given list they feel inher-
ent in the given objects or symbols.
Art seems to be the best illustration for this process. We suppose that in
the course of creating works of art an artist invests personal values of his/her
own into the body of the work, usually without being very aware of this pro-
cess. These values can be perceived by the audience as the values embodied
Non-classical value approach / 231
in the artwork. This process also goes usually without awareness but can be
brought into awareness through using special research techniques, such as the
Value Spectrum Technique described below. This hypothetical description
has received a solid empirical justification, which is presented in the follow-
ing pages.

Value Spectrum Technique

The starting point of the methodological development in line with the
above ideas was the theory of metamotivation and Being-values by Abraham
Maslow (1970, 1976). In his latest writings Maslow has come to the conclu-
sion that individuals who had reached self-actualization and passed from the
stage of Becoming to the stage of full-range Being, have now different motiv-
ation described in other terms than basic needs. Maslow claimed that they are
metamotivated by ultimate Being-values like Truth, Good, Justice, Process,
Completion, Play, etc., which are simultaneously parts of the world and ele-
ments of the motivational structure of individuals who reached this level (cf.
non-classical value approach). He has also composed the list of 17 B-values.
My colleague Evelyne Ulyanova (personal communication, 1987) proposed
to use this list as the basis of a ranking procedure, similar to that of the
Rokeach Value Survey (Rokeach, 1973); however, it turned out that even
highly intelligent subjects failed to rank B-values, unlike earthly values from
Rokeach lists. We changed the initial idea and made the technique of reper-
tory grid type (Fransella & Bannister, 1977) which was called Value Spec-
trum Technique (VS). The subjects had to solve the task of linking the given
values to the given objects. They were given the table, its rows made by 18
invariant B-values, slightly adapted from Maslow’s list, and columns by a
number of objects (usually between 6 and 15). Objects were different in dif-
ferent studies; the objects presented could be quite general (life, art, death,
past, future, music, painting), more special (rock-music, pop-music, classic
music, this composer), and individual (painting Nr. 1, 2 etc., music piece Nr.
1, 2, etc.)
The guidelines were as follows: “Here you see the list of values. Please
put marks in the column with the word Life at the top, indicating the values
which you consider to be inherent in Life; in the next column with the word
Music please mark the values inherent in Music, etc. Do not think too long;
no responses can be right or wrong. You may put any number of marks or no
marks in any column, depending on your feeling”.
The summarized results of a sample have the form of a group matrix in
the same form as individual sheets; the cells are filled with figures meaning
232 / Aesthetics and Innovation
the number of subjects who marked this cell value x object. The distribution
of the figures in matrix is far from being random, revealing the unique value
profile of every object. The data provide the rich opportunities of different
ways of analysis: qualitative phenomenological description; comparison of
row sums (frequency of a given value), comparison of column totals (total
number of values ascribed to different objects), different measures of similar-
ity/distances between whole value profiles of different objects (city-block
method, Euclidean distance, chi-square), factor and cluster analysis. A lot of
data has been collected in the VS studies of perception of painting, music,
and theatre, and some data on poetry and animation with the help of our stu-
dents Halaa Abdel-Fattah, Elena Belonogova, Milena Budsiak, Tatyana Del-
skaya, Galina Ivanchenko, Larissa Lagoutina, Mary Nazarova, Alexandra
Nikitina, Pavel Sabadosh, Natalia Serebryakova, Julia Volkova, Georgy
Yemelyanov, and Eugenie Zhukova.
VS technique makes it possible to compare the VS of different objects in
one sample; to compare the VS of the same objects in different samples; to
compare the location and distances between various target objects in the com-
mon “value space”; also in different samples; to assess the value potential of
artworks. The following is an overview of the main findings obtained in em-
pirical studies using VS technique, some of them yet unpublished.

Works of art carry embodied painter’s values

The crucial assumption of the approach says that B-values are being “in-
vested” into the artwork by the author and then detected by the recipients,
both processes typically going on without awareness. We have attempted to
check the possibility of intentional embodiment of values in the artworks in a
joint study made with Halaa Abdel-Fattah, both professional psychologist and
professional painter (Leontiev & Abdel-Fattah, 1997). Ms. Abdel-Fattah cre-
ated eight sketches in water-colors and pastel, each of them dedicated to
some value from A. Maslow’s list of B-values: Beauty, Completion, Dicho-
tomy-transcendence, Effortlessness, Justice, Meaningfulness, Order, Whole-
Twenty six participants aged 16 to 40 were asked to evaluate all the eight
sketches by VS technique. If our assumption was correct, the “title value” of
every sketch (the one the painter aimed to express) would have been ascribed
to the sketch significantly more often than other 17 values. The factors that
were expected to level the results to some degree were eventual imperfection
of value embodiment in a given sketch and the interconnectedness of differ-
Non-classical value approach / 233
ent values, hardly strictly distinguishable from one another (see Maslow,
Despite the influence of these factors, the results turned out to be close to
perfect confirmation of the hypothesis. In all the eight cases “title values”
were ascribed to corresponding sketches much more often than other ones:
three times they have been ascribed more frequently than all the others, two
times they had the second frequency rank, once the third rank, once the rank
4.5 and only once being ranked as low as 7 (from 18 items). In fact, it was as
a rule a group of interconnected values, rather than a single value, that stood
out in the value spectrum of every sketch. The results in general allow us to
state that works of art (paintings) can really serve as means of intended and
effective communication of the author’s values. It is easy to infer that this can
occur also without awareness.
Different sketches were not equally value loaded: the sum total of
ascribed values varied from 50 (Order) to 179 (Completion), the rest of the
sketches falling into the range 82 to 111. This can be explained by unequal
success in creating different sketches, as well as by varying value width. In-
deed, as our results show, quite a number of values are ascribed to the sketch
“Completion” with nearly equal frequency, among them: Goodness, Truth,
Meaningfulness, Completion, Aliveness, Justice. This sketch seems to absorb
all of them.
The total frequency of use of each value category also varies notably:
Beauty 83, Aliveness 79…..Necessity 33, Self-sufficiency 32, Justice 25 (the-
oretical maximum is 26 x 8 = 208). This might be explained as general values
of painting, special values inherent in the works of the given painter, values
characteristic of our sample, or the combination of all the three biases.

Naïve and professionally trained participants are equally successful in identi-

fying values inherent in artworks

These findings revealed themselves in some studies with musical pieces.

Pavel Sabadosh (2000) presented a number of relatively popular classic mu-
sical pieces (Beethoven, Piano sonate # 14 (“Moonlight”), part 1;
Tchaikovsky, Dance of little swans from the “Swan lake”; Albinoni, Adagio;
Khachaturian, Sword Dance from “Haiane”; Beethoven, Symphony # 5, part
1; Biset, Toreador’s couplets from “Carmen”) to a group of 26 participants
without musical training and another group of 26 professional musicians. The
results showed only minor differences between the groups, both in value
loadings of different pieces and in value spectrum of individual pieces. In a
recent unpublished study with avant-garde academic music of the 2nd half of
234 / Aesthetics and Innovation
XXth century (Webern, Schnitke, Penderetzki, Gubaidullina, etc.) Leontiev
and Nikitina (2006) have found significant differences between naïve and
musically educated listeners in the evaluation of pieces of mixed style, and
no differences in the evaluation of the purely avant-garde pieces.
However, naïve and expert groups differ in their value profiles of general
concepts like art, special kinds of art, and especially life. In an unpublished
study by Leontiev and Lagoutina (1996) natural science students of Moscow
State University have been compared to students of State Institute of Theater
Art. Both groups ascribed similar values to Theater, but different ones to Life.
The former attribute more values to both target categories than the latter; they
also see Life as richer in value spectrum than Theater, while Theater Art stu-
dents, on the reverse, attribute more values to Theater than to Life. In the
above mentioned study of Sabadosh (2000) the total value loading (sum-total
of all ascribed values) of Life was 1.5 times greater in naïve than in profes-
sional participants. These and some other data suggest that art experts and
professionals tend to appreciate art to some degree at the expense of life that
becomes less value loaded.
Similar differences have been found in the comparison of different groups
of spectators who watched the play “Two on a Swing” (Moscow Maly Theat-
er, small stage) and evaluated the concepts Life, Theater Art and The Actual
Play by VS (Leontiev, Delskaya, & Nazarova, 1995). The Actual Play was
evaluated as closer to Life than to Theater Art by its value spectrum. Then the
subjects have been divided into the groups of occasional visitors and non-oc-
casional spectators by the set of criteria. For occasional visitors, the VS of all
the three target objects was very similar, for non-occasional spectators fairly

Different objects, including individual artworks, abstract concepts and mental

images, have distinguishable value profiles

Table 1 below presents the summarized data on value spectrum of 11 gen-

eral concepts, including Life, Art, some special kinds of art or mass culture,
Dream, Myth and Fairy Tale provided by a random sample of 63 adult parti-
cipants of both sexes.
The most value-loaded concept is Life, closely followed by Fairy Tale.
Life is characterized first of all by Uniqueness, Goodness, Dichotomy-tran-
scendence, Playfulness and Meaningfulness, least of all by Completion and
Truth. Goodness especially stands out for Fairy Tale, much more than play-
fulness and meaningfulness. What differentiates Fairy Tale from Life is lack
of Uniqueness and Dichotomy-transcendence. Music and Poetry with pretty
Non-classical value approach / 235
similar value profiles follow next by their value loading. Top values for Mu-
sic are Beauty and Uniqueness, and for Poetry—Beauty, Meaningfulness and
Effortlessness. Art, Animation, and Painting make the next group, Art being
distinguished by the top value of beauty and (much less) Playfulness and
Uniqueness; Animation by Playfulness, Goodness and Effortlessness, and
Painting by Beauty and Uniqueness. The last and very special form of art is
Documental Cinema with top values of Aliveness, Truth and Meaningfulness.
Still less value loaded are Myth with the top values Beauty and Meaningful-
ness and Dream with the top value Necessity. TV series has the lowest value
loading with the only distinguished value being Playfulness.
Many interesting results have been obtained in the study of VS of musical
genres (Leontiev, Ivanchenko, Avraamova, & Kravtsova, 2001). We hypo-
thesized that different musical genres carry different values for different audi-
ences; indeed, all the musical (and not only musical) genres serve as object of
admiration for some group of devoted fans. Musical preferences seem to be
quite a stable aspect of the person’s and group’s identity. Treating music, like
art in general, as an existential phenomenon, we considered it rather import-
ant to study it in real life contexts, including social group context and histor-
ical (temporal) context. The target objects were six rather abstract concepts:
Life, Music, Pop-Music, Rock Music, Classical Music, Chanson. The last
genre was characteristic for late Soviet underground in 1970s—1980s; char-
acteristically the composer and the performer (often also the poet) is the same
person. Data were collected among the listeners of concerts of different
genres; we roughly labeled the person met at a special musical event as a fan
of the corresponding musical genre. The study covered two periods with a
decade gap between them: 1986—1990 with the samples of rock-fans (R1),
classical music fans (C1), chanson fans (A1) and 1998—1999 with the
samples of rock-fans (R2), chanson fans (A2), pop-music fans (P2), 198 re-
spondents in total. In this way we could compare synchronic and diachronic
aspects of perception of music through the prism of values.
The results show that we were right to treat musical genres as existential,
rather than purely aesthetic phenomena, especially in the corresponding audi-
ence. The values of Life turn out to be as relevant for many genres as the val-
ues of Music. Value spectrum of Rock-Music reveals significant correlations
to VS of Life, rather than to the VS of Music, in R1, R2, A2. VS of Chanson
also correlates to the VS of Life rather than to the VS of Music in A1, A2, K1
and R2. In pop-fans (P2) VS of Life revealed no significant
Table 12-1. Value spectrum of general categories, N = 63 (Leontiev & Serebryakova, 2004, unpublished)

236 / Aesthetics and Innovation

Non-classical value approach / 237
correlations to any musical genres. VS of Music significantly correlates to
Classical Music in A1, R1, and C1; to Сhanson in A1 and C1; to Pop-Music
in R1 and P2. Only Rock-Music reveals no significant correlations to Music.
All the investigated musical genres can be classified in two dimensions,
in terms of their closeness to Life and to Music; Life and Music make 2 nearly
orthogonal axes of the classification space for musical genres (see Table 2).

Table 12-2. Value similarity of musical genres to Life and Music

Genre Value similarity

to Life to Music

Classical music Neutral High

Pop-music Reverse High
Rock-music High Reverse
Chanson High Neutral

Despite the apparent differences, in various samples we find significant

positive correlations between VS of hardly compatible genres: Pop-Music
and Rock Music (in A1 and C1), Pop-Music and Chanson (in C1, P2), Rock-
Music and Chanson (in C1, A2, R2). Significant negative correlations are
rarer: Classical Music and Rock-Music (A2). It is amazing that all the fans,
except for pop-music fans, perceive their preferred genre as close to Life in
value dimension.
We had the possibility to notice also some diachronic trends, comparing
the results of the first and the second section. The general trend seems to be
convergence, especially that of Rock-Music and Chanson: if in late 1980s the
fans of both genres saw more differences than similarities, by the late 1990s
the perception of their similarity prevails over difficulties.

The idea of connection between art and values does not seem too radical:
our common sense easily accepts it. Nevertheless, no attempt has been made
so far to conceptualize the psychological mechanisms of this connection, nor
to investigate the functioning of values in art creation and perception. The
238 / Aesthetics and Innovation
above proposed approach, based on Lev Vygotsky’s insights, may open a new
perspective. The experimental studies, described in the last section, make it
rather evident that not only artworks, but also many general categories do
possess a distinctive value potential (spectrum) of their own, that can be re-
vealed by recipients. Similar studies made with advertisements have proved
that the amount of values attributed to an object account for its attractiveness.
Humans live in the realm of values, and it is the dynamics of these values that
underlie the particularly humane mechanisms of aesthetic activity along with
a number of other advanced forms of activity as well. From this viewpoint art
creation and perception is not an exotic activity, but rather a form of activity,
perfectly corresponding to the evolutionary task of developing and realizing
our value potential.

Elkonin, D. B. (1989). Ob istochnikakh neklassicheskoi psikhologii (On the
sources of non-classical psychology). In Elkonin D. B. Izbrannye psikho-
logicheskie trudy (Selected psychological writings) (pp. 475–478). Mo-
scow: Pedagogika. (In Russian).
Frankl, V. (1985). Logos, paradox and the search for meaning. In M. J. Ma-
honey & A. Freeman (Eds.), Cognition and psychotherapy (pp. 259–275).
New York: Plenum.
Fransella, F., & Bannister, D. (1977). A manual for Repertory Grid Tech-
nique. London: Academic Press.
Leontiev, D. A. (1992). Joint activity, communication, and interaction. Journ-
al of Russian and East European Psychology, 30, 43–58.
Leontiev, D. A. (1996a). Tsennost kak mezhdistciplinarnoe ponyatie: opyt
mnogomernoi rekonstruktsii (Value as an interdisciplinary concept: an at-
tempt of multidimensional reconstruction). Voprosy filosofii, 4, 15–26.
Leontiev, D. A. (1996b) Ot sotsialnykh tsennostei k lichnostnym: sotsiogenez
i phenomenologiya tsennostnoi regulatsii deyatelnosti (statya pervaya)
(From social to personal values: Sociogenesis and phenomenology of the
value regulation of activity, paper 1) Vestnik Moskovskogo Universiteta,
Ser. 14. Psikhologiya, 4, 35–44.
Leontiev, D. A. (1997a) Ot sotsialnykh tsennostei k lichnostnym: sotsiogenez
i phenomenologiya tsennostnoi regulatsii deyatelnosti (statya vtoraya)
(From social to personal values: Sociogenesis and phenomenology of the
value regulation of activity, paper 2) Vestnik Moskovskogo Universiteta,
Ser. 14. Psikhologiya, 1, 20–27.
Non-classical value approach / 239
Leontiev, D. A. (1997b). Vygotsky’s psychology of art: Non-classical
paradigm for general psychology. In L. Dorfman et al. (Eds.), Emotions,
creativity, and art (Vol.1, pp. 105–115). Perm: Perm State Institute of Arts
and Culture. (In Russian).
Leontiev, D., & Abdel-Fattah, G. (2000). Translatsia tsennostei v zhivopisi
(Translation of values in art). In E. Malyanov, L. Shipitsina, L. Dorfman,
S. Kornienko, & C. Martindale (Eds.), Tvorchestvo v obrazovanii, kul-
ture, iskustve (Creativity in education, culture, and art). Proceedings of
International conference (pp. 108–111). Perm, Russia.
Leontiev, D., Delskaya, T., & Nazarova, M. (1994). Images of a theater play:
Expectations, impressions, and group differences. In C. Dreyer u.a. (Hg.),
Lebenswelt: Zeichenwelt—Life world: Sign world. (Bd. 2, S. 799–814).
Lueneburg: Jansen–Verlag.
Leontiev, D., Ivanchenko, G., Avraamova, M., & Kravtsova, O. (2001). Tsen-
nostnyi spektr muzykalnykh zhanrov: sinkhronicheskie I
diakhronicheskie aspekty (Value spectrum of musical genres: Synchronic
and diachronic aspects). In G. M. Balim, V. M. Petrov, & V. P. Ryzhov
(Eds.), Informatsionnyi podkhod v naukakh o cheloveke (pp. 174–198).
Taganrog: Taganrog State Radio Engineering University. (In Russian).
Leontiev, D., & Yemelyanov, G. (1994). Catching psychological effects of
poetic form: Experiments with parodies and translations. SPIEL, 13, 1,
Mamardashvili, M. K. (1970). Forma prevraschennaya (Converted form).
Filisofskaya entsiklopediya (Philosophical Encyclopedia) (Vol. 5, pp.
386–389). Moscow: Sovetskaya entsiklopediya. (In Russian).
Maslow, A. (1970). Religions, values, and peak-experiences. New York: Vik-
ing Press.
Maslow, A. H. (1976). The farther reaches of human nature. Harmondsworth:
Penguin, 1976.
Rokeach, M. (1973). The nature of human values. New York: The Free Press.
Sabadosh, P. (2000). Tsennostnyi spektr muzykalnykh proizvedenii (Value
spectrum of musical pieces). In V. M. Petrov & V. P. Ryzhov (Eds.), In-
formatsionnaya paradigma v naukakh o cheloveke (Information
paradigm in the human science) (pp. 213–219). Taganrog: Taganrog State
Radio Engineering University. (In Russian).
Sobkin, V. S., & Leontiev, D. A. (1992). The beginning of a new psychology:
Vygotsky's psychology of art. In G. C. Cupchik & J. Laszlo (Eds.), Emer-
ging visions of the aesthetic process (pp. 185–193). NY.: Cambridge Uni-
versity Press.
Vygotsky, L. S. (1971). The psychology of art. Cambridge: The MIT Press.
240 / Aesthetics and Innovation
Vygotsky, L. S. (1983). Istoriya razvitiya vysshikh psikhicheskikh funktsii
(History of development of higher mental functions). In Vygotsky L. S.
Sobranie sochineniy (Collected works) (Vol. 3, pp. 5–328). Moscow: Ped-
agogika. (In Russian).

Individual and Professional Differences in the

Perception of Dramatic Art

Dmitry A. Leontiev and Larissa Lagoutina

The chapter presents a study of the perception of dramatic art by profession-

ally competent and naïve spectators based on the free description method
elaborated by the first author. Nine strategies of free description of dramatic
theater performances have been identified, four of them proved to be stable
across time. The data presented describes the frequency of every strategy in
professional and lay samples, relation to other strategies and correlations with
other variables. Qualitative and quantitative analysis leads to emotionality vs.
rationality as the basic dimension structuring the perception of dramatic art in
lay audience. In professionally competent spectators this dichotomy becomes
less important than the dimension of competence.

G. B. Shaw noted in the introduction to "St. Joanne" that many playgoers
go to theater for the same reasons as many go to church: to expose one's ap-
parel, to keep up with fashion, to have something to talk about, to admire a
star, to spend the evening anywhere except at home; in other words, theater is
often motivated by anything except for the interest regarding the dramatic
artwork itself. Based on theoretical and empirical research in the field of art
consumption (we deliberately use the economic term here, because this word
is the only one that can embrace all the different forms of a person’s interac-
tion with art), we can propose some additions to this list: recreation, looking
for models of behavior, searching for the solution to personal problems,
search for meaning, search for some emotional states lacking in everyday

242 / Aesthetics and Innovation
life, getting support regarding one’s values, getting rid of one’s thoughts and
anxieties, confirmation of one’s world view, dating, etc.
In all these cases theater plays, like art in general, are perceived and com-
prehended in different ways. The spectrum of these ways seems, however, to
be limited. In a series of experimental studies the first author attempted to ap-
ply an individual approach to art consumption as opposed to the traditional
theoretical-normative one (Leontiev, 1992; 1998). The theoretical-normative
approach presumes that there is a standard, perfect, advanced type of true
aesthetic perception; all imperfect processes of art consumption we meet are
to be evaluated by to the closeness of these empirical processes to the theor-
etical-normative one. We consider such an approach misleading, because it
reduces all the qualitative peculiarities of person-art interaction to a single di-
mension of adequacy. In fact, perfect aesthetic perception is not so often the
case, even if we deal with advanced participants. At the same time “imper-
fect” empirical types of art consumption are not just failures of a perfect one;
they have qualitative peculiarities of their own and deserve special attention.
There may be a full-range contact between a person and a work of art even if
this contact has nothing to do with aesthetic communication. “Demands and
expectations of definite categories of the audience may transform the mean-
ingful stuff of a work of art. The latter thus fulfills its social functions, gains
popularity, effectively influences the audience also in case when it is not
evaluated by aesthetic criteria and is in fact outside the system of aesthetic
perception in the proper meaning of the term” (Dondurei, 1975, p. 28). We
must keep in mind the difference between art perception and aesthetic per-
ception; art perception is not always and not even as a rule aesthetic;
however, we should not exclude from our focus of interest non-aesthetic
forms of art consumption.
The individual approach, alternative to the theoretical-normative one, is
based on the primacy of qualitative phenomenological analysis of empirical
processes of person—art interaction in all of its variety. A special research
technique of free descriptions has been elaborated for this purpose (Leontiev
& Kharchevin, 1994). The participants were asked to list four books (movies,
theater plays, paintings) they have recently read or watched. Upon complet-
ing the list they were asked to write something, roughly half a standard page,
about each of these four items so that the reader could get some idea of them.
Then we applied a type of content analysis to the collected descriptions. Us-
ing some categories of words and judgments as indicators of definite types of
information processing in the broadest meaning of the term, we described
sets of free description strategies. This approach has been applied earlier to
the study of fiction literature (Leontiev & Kharchevin, 1994) and (painting
Dramatic art / 243
Leontiev & Belonogova, 2002). In the present paper the application of the
free description methodology to the perception of theater is described.

On the psychology of theater perception

In a paper published over a decade ago (Leontiev, Delskaya, & Nazarova,
1994), it was stated that theater had not yet become the object of more or less
systematic empirical studies. Since then the situation does not seem to have
changed much. What makes theater a complicated object of study, less at-
tractive than literature, painting, or music, is the following.
First, theater is a complex kind of art. The impression of a theater play is
determined by a complicated interplay of stimuli of quite different nature:
speech, its extralinguistic components, bodily expressions and movements,
visual stimuli, music and noise, etc., including such higher-order gestalts as
the composition of the staging, created by producer, and the special atmo-
sphere of the theater that contributes a lot to the whole impression. Delacroix
called theater evidence for human striving to experience the greatest amount
of various feelings at the same time.
Second, theater is a collective kind of art. Not only in the sense that a play
is created by many persons, but also in the sense that it can be perceived in a
special collective setting, in a more or less large audience. The role of the at-
mosphere of the theater, mentioned above, makes it impossible to provide a
full-range perception of a theater performance outside the theater setting, e.g.,
on a home video. Theater is a show; sociopsychological regularities of com-
munication and mass behavior influence the process and the results of theater
play perception at least as strongly as aesthetic regularities of art perception
do. This obstacle makes laboratory studies of theater perception completely
Third, theater is a process related, temporal kind of art: unlike most others
kinds of art, in this case the meaning and sense of the whole work develop
gradually in the course of the performance. The whole performance lasts usu-
ally two or three hours and, unlike movies and music pieces, it cannot be
presented in parts, interrupted, stopped, or changed in order to catch some
impression of the play at its intermediate stages.
Fourth, theater is not only art, but also the way of life for many engaged
persons, including a number of spectators. Theater imitates life; this is why it
acts as if creating a parallel life, a parallel world, a parallel reality that is in
some way separate from the real ordinary life. One of the relatives of the first
author, who was an actor in his youth, said that he has abandoned this profes-
sion after he realized one day that continuing to be an actor he would never
244 / Aesthetics and Innovation
live a real life. As an institution, theater developed into a kind of religion,
cult, or church, with its devoted believers and formal adherents, as well as oc-
casional visitors. The atmosphere of the theater, starting from the doors, plays
an extraordinary role in the whole impression; R. Demarcy (1973) revealed
symbolical meanings of different elements of theater interiors as expressing
wealth, power, and the sacral. "Entering a theater edifice, like a temple, one
comes from the amorphous world space to the meaningful universe, built
upon a world view. And, while entering the world temple, I reincarnate;
pierced by flashes of world order, the spiritual structure of my being changes;
the excitement emerges, that is the heightened tune of the soul. I feel like a
microcosm, the heart and the eye of the great world..." (Gachev 1968, p.
212). It is thus hardly possible to study the perception of theater play outside
the theater setting, as well as cut it into pieces in order to investigate the tem-
poral development of artistic meanings and senses in course of the perform-
In a special field study (Leontiev, Delskaya, & Nazarova, 1994) we stud-
ied the dynamics of the impression of the play "Two on a swing" after W.
Gibson, produced by A. Shuisky in the Moscow Maly Theater, Chamber
stage. The anticipations and the impressions were assessed before the begin-
ning and in the entr’actes. We have found in all the spectators some rather
concrete nonverbal anticipatory images of what the play would look like.
These were rather congruent in a sample of spectators; actual impressions
somewhat differed from anticipatory attitudes.
There was no possibility to consider individual differences of the spectat-
ors except for the most rough ones. Among the theater audience there are
competent skilled spectators who are able to decode rather complicated lan-
guages of stage action and to get aesthetic pleasure from pieces of high art.
There are also incompetent, unskilled spectators who fail to understand many
aspects of symbolic language of theater performance; what they need and like
most are kinds of performance less complicated by the expressive means.
And there are also spectators, called occasional by contemporary theater soci-
ologists—these, who came to the theater tonight by chance, looking for any
kind of relaxation and rest and meeting the theater on their way.
We tried to consider the type of spectators having asked four simple ques-
tions: 1. How often do you usually go to theater? 2. Why do you usually do
it? 3. Why are you at this performance tonight? 4. What do you know about it
(the play, its author, producer's name, actors, etc.). The answers to all these
questions turned out to be quite coherent in over 70% of the sample who fit
to one of the two categories: A. Those who go to theater to relax; have come
tonight for occasional reasons; don’t know anything about the play to see (oc-
Dramatic art / 245
casional spectators). B. Those who go to theater for different reasons; have
come tonight purposefully; know something about the play to see (non-occa-
sional spectators). Non-occasional spectators were more congruent in their
anticipatory images than occasional ones, while actual impressions of the
two-third of the play leveled these differences: the congruence of estimates
increased in occasional spectators and decreased in non-occasional ones.
The Value Spectrum technique has also been used (see Leontiev, this
volume): the participants had to assign any of the set of 18 ultimate Being-
values to three objects: Life, The art of theater, The current play. Both groups
perceived the values of the play as close to life in respect to its value spec-
trum. However, non-occasional spectators see less similarity between life and
theater and much greater distance between the play and the art of theater. We
suppose that they have a more complicated view of the The art of theater and
expect from the play more than just lifelikeness.

Goals, design, samples and methods of the study

The aim of the present study was to reveal more subtle individual differ-
ences in the perception of theater beyond the basic distinction of non-occa-
sional vs. occasional spectators. We have been solving three tasks: 1. To re-
veal and to describe various form of theater perception using the method of
free descriptions; 2. To compare the spectrum and distribution of strategies in
professionally competent and nonprofessional samples; 3. To investigate the
connections between the types of free description strategies and some person-
ality variables as well as the level of competence in dramatic art.
Since we wished to focus on stable individual differences, the study was
not linked to a specific play. Instead, we have used a quasi-experimental
design with a partial one-year retest.
The main set of methods included free description technique and the Dra-
matic Art Competence Survey. The additional set included the Emotionality
Express Test and a Russian version of the Personal Orientation Inventory.
1. The free description technique, elaborated by the first author has been
used for the study of individual peculiarities of perception of different arts
(Leontiev & Kharchevin, 1994). At least some of the individually preferred
free description types proved to be stable in time and reflecting some indi-
vidual characteristics of the participants. The instruction is as follows:
“Please recollect four dramatic theater performances you saw recently and
write down the titles, and the names of the theaters. (Pause). And now, please,
describe each of them in such a way so that someone reading this description
246 / Aesthetics and Innovation
could get some idea about the performance. Each description should take
roughly half of a page.”
2. Dramatic Art Competence Survey (DACS) has been elaborated by the
authors specially for this study. It is a sociological inventory with questions
referring to demographic variables, art preferences and interests. Two special
testing tasks are included to check the participant’s knowledge in the field of
dramatic art. The first task requires to recollect and write the authors of 11
plays, both classical and actual ones, and the second task requires to name the
Moscow theaters where each of the 10 given plays is performed.
3. Emotionality Express Test (EET), published in Petrova (1982) is sup-
posed to measure most general rationality/emotionality ratio. The task is to
classify four words—nightingale, cockroach, kiss, cough—into two groups
by any reason. Some classify by formal rational reasons, some by emotional,
grouping nightingale with kiss, and cockroach with cough. There are no data
on the validity of such a procedure; however, it seems relevant to our pur-
4. Personal Orientations Inventory (POI; Shostrom, 1966; Russian Ver-
sion: Gozman & Kroz, 1987) is a multiscale inventory supposed to measure a
number of traits relevant for the ideal of a self-actualizing person (Maslow,
Two groups participated in the study. Group L included 18 undergraduate
students of biology and economics from Moscow State University of both
sexes, age 20 to 23. Group T included 15 undergraduate students of the State
Institute of Theater Art (GITIS) in Moscow of both sexes, age 20 to 25.
Group L was assessed twice with an interval of one year; the first section
(L1) included only free description and DACS, the second section (L2) the
whole set of methods described above. Group T filled only free description
and DACS.
We expected (1) that free description types would be generally stable
across time; (2) that there would be both differences in competence and qual-
itative differences in free description types between the groups L and T; (3)
that the free description types would correlate to personality and motivation

Results and Discussion

Free description strategies

As in other studies, qualitative analysis of free descriptions of dramatic

performances allowed us to single out several types of theme, which we call
Dramatic art / 247
free description strategies. By strategy we mean an unconscious attitude
rather than conscious plan. Each strategy was singled out by definite words
and judgments used in the description. In total we distinguished nine
strategies; Table 1 presents their description and examples.

Table 13-1. Strategies of free descriptions, criteria for their identification and some

Strategy Criteria Examples

Narrative Description of the charac- Adventures of a young

ter’s actions, the succes- man, his meetings with
sion of events, the plot, the people of various profes-
narrative sions, social positions
Judgments expressing the Everything can be seen in
general idea of the play, its everything. Everything re-
intention or morale peats in the world, the
events are projected one
onto another
Judgments made as if from ... principles of the per-
the viewpoint of a theater formance: lingering,
critic, evaluation of sta- smooth gestures; calm,
ging, producer’s and act- gentle voices...
ors’ work, expressive
Words and judgments con- I have been watching it
taining general emotional many times and I still feel
evaluation of the play as a excitement. That’s great!
whole or of its aspects, and
immediate impression of
the play

Table 13-1. Continuation

Analysis of one’s own Is it sadness or not, I like
states and experiences this feeling. Even don’t
aroused by the play know why. Just feeling sin-
248 / Aesthetics and Innovation
Linking the play to the I dreamt of this for long—
context of one’s biography, to see “Junone and Avos“.
one’s personality Always have been fascin-
ated by the romantics and
the eternal love
Associations on everyday, ... This apocalyptic mental-
personal, philosophical is- ity of the new wave is
sues aroused by the play already losing its power
and truth
Judgments on the connec- It’s a kind of expressionist
tions of the play with cul- German tradition
tural, historical, or aesthet-
ic traditions, author’s in-
tent etc.
Judgments about the atmo- It suffices just to get into
Theater space
sphere, interior, events oc- this “temple of the
curring outside the stage in beauty”—that’s how I per-
the theater hall ceive our Opera. Even
without watching the per-
formance you would feel
happy with everything you
see there

Table 2 presents the frequency of all types of strategies across samples

and sections as well as the results of checking the stability of free description
strategies across time. For this purpose we compared individual data of L1
and L2 by the binomial Phi coefficient and Spearman rank correlation, rho.
Table 13-2. Distribution of free description strategies across samples and sections and their stability in time

Note: ** p < .01; * p < .05; “ p < .06

Dramatic art / 249
250 / Aesthetics and Innovation
It is evident from the table that there were very few notable changes in
free description strategies distribution in L2 as compared to L1; however, in
the professionally competent sample T there are remarkable differences from
the lay sample. The only change across time is that in L2 we see a higher dif-
ferentiation, more types of different strategies are combined in the same num-
ber of descriptions. Emotional and narrative strategies belong to the most fre-
quently used by random samples for all kinds of art; a relatively high em-
phasis on technological details is rather special for dramatic art. It is remark-
able that if in the lay group technological strategy, second by frequency, was
found in approximately every fifth description, their frequency in theater stu-
dents increased up to two of three. Cultural strategy is also very frequent in
the T sample, unlike the L one. The introspective strategy and the personal
one are practically not used in the group T.
By statistical criteria all the free description strategies can be divided to
those which reveal good stability across time (emotional, personal, cultural
and narrative) and those which reveal no notable stability (summarizing,
technological, associative and theater space). We have inconsistent results
with only the introspective strategy. It can be concluded that though the
strategies in sum reveal good consistency across time (rho = .56, p < .05), in
line with the previous findings on fiction literature (Leontiev & Kharchevin,
1994), only some of them are intraindividually stable and assumedly connec-
ted to personality variables, while others are used more ad hoc and probably
depend on the special artwork being described. It is amazing that the most
stable strategies are not always the more frequent ones. For example, a very
frequent technological strategy is absolutely unstable, and relatively rare per-
sonal and cultural strategies are as stable as emotional and narrative ones.
We have tried to apply factor analysis (main components) to our free de-
scription data; however, small samples make our data a little unreliable. The
factorial structure in L1 (5 factors covering 82% of the variance) was not rep-
licated in L2 (4 factors covering 79%). The only point worth noting is that the
factorial structure in T revealed only 3 factors covering 72% of variance, that
is in line with previous findings (Delskaya & Nazarova, 1991): the factorial
space of theater performance evaluation by semantic differential in profes-
sional spectators revealed less factors, i.e., was more structured than in lay
More consistent were the results of cluster analysis: in all the three
groups. The two largest clusters at the last step had to do with the dimension
rational vs. emotional perception. This was most evident in L2, as can be
seen in Table 3. It seems remarkable that in theater students unlike non-theat-
er ones, the rational pole embraces even emotional strategy; emotion prob-
Dramatic art / 251
ably appears for them as a technological element of theater performance
rather than an aspect of individual impression; the latter is expressed more in
personal and associative descriptions.

Table 13-3. The results of cluster analysis of free description strategies: strategies fall-
ing to two largest clusters in every group

Group Rational type Emotional type

L1 narrative, summarizing, in- personal, cultural, emotional,

trospective associative, technological
L2 narrative, summariz- personal, cultural, emo-
ing, technological, as- tional, introspective, theat-
sociative er space
T narrative, summariz- personal, associative,
ing, technological, theater space
emotional, cultural

The analysis of correlations between strategies is also consistent with

these findings, though significant correlations are few, that speaks in favor of
our treating different strategies as mutually independent. Four of 36 correla-
tions reached the .05 significance level. Emotional strategy negatively correl-
ated with the narrative one (rho = -.51); the other three correlations were pos-
itive: narrative and summarizing (.63), personal and introspective (.51), asso-
ciative and technological (.50). Emotional perception, exemplified by the
emotional free description strategy, and rational perception, exemplified by
the combination of narrative and summarizing ones, seem to be two basic
types of the perception of dramatic art.

Relation to other variables

The answers to DACS revealed huge differences between the groups. The
average frequency of visiting theater performances was 4.2 times a year in
lay spectators and about 60 in theater students. The task to write a name of
any contemporary play writer and any of his plays was successfully fulfilled
by 14 of 15 theater institute students and by only seven of 18 of non-theater
students. The average number of correct answers to the task to guess the au-
thors of the given plays was 2.11 of 11 in group L and 10 of 11 in group T.
252 / Aesthetics and Innovation
The average number of correct answers to the task to name the theaters where
the given plays go was 1.11 of 10 in group L and 7.46 of 10 in group T. It fol-
lows from these figures that the samples we were comparing do in fact rep-
resent strong differences in dramatic art competence.
We have found no significant correlations of different strategies with dra-
matic art competence because of small statistics and small variance in dra-
matic art competence in L group (correlational analysis was made only on the
L2 group). The only highly significant correlation (.82, p < .01) was that of
theater space strategy with recognition of theaters running the given plays;
however, this correlation in fact reflects the single case of using this strategy
in L2.
The answers to EET allowed to classification all the participants of the
group L to two subgroups: “rational” and “emotional”. Despite small sample
size, we found very notable differences in the frequency of the use of four

Table 13-4. The frequency of use of different free description strategies by “rational”
and “emotional” participants (per person)

Strategy Emotional Rational

(N = 12) (N = 6)

Narrative .50 1.83

Summarizing .33 1.16
Technological 1.08 1.33
Emotional 2.33 1.50
Introspective .50 .33
Personal .42 .16
Associative .50 .66
Cultural .58 .00
Theater space .03 .00

As can be seen in Table 4, the “rational” participants used narrative and

summarizing strategies more often and emotional, personal, and cultural
strategies less often than the “emotional” participants. The difference
Dramatic art / 253
between the two groups was, however, significant only for narrative strategy
(p < .003 by Fisher exact one-tailed test) and summarizing strategy (p < .03).
For the widespread emotional strategy the difference was near-significant (p
< .08); differences for personal and cultural strategies that have not been re-
gistered in the “rational” group more than once could not be tested statistic-
ally. Generally however this data reveals a good correspondence with both
cluster analysis data and stability across time data: four of five strategies that
seem to distinguish the “rational” from the “emotional” spectators belong to
the stable ones.
Significant correlations of free description strategies with POI scales are
also few, most of them referring to the basic strategies. Narrative strategy
positively correlates with Contacts scale (.68) and negatively with the Time
competence scale (-.57). Individuals paying attention mostly to the narrative
tend to be social, but perceive life discretely, the present, the past end the fu-
ture being weakly connected. Emotional strategy negatively correlates with
the Self-support scale (-.52) and Acceptance of aggression scale (-.54). Emo-
tional perception thus seems to be characteristic of rather impulsive and ex-
ternally oriented individuals. Cultural strategy positively correlates with the
Cognitive needs scale (.54) and Associative strategy with the Creativity scale
(.52) that hardly requires additional comments.

The first stage of a person’s interaction with a work of art is mental recon-
struction of this work, the mental shift from the real world of material objects
to the fictional world of artistic images, reconstructed by mental mechanisms
from printed letters on paper, or colored dots on canvas, or humans moving
on the theater stage and talking. Our approach through free descriptions
makes it evident that such a reconstruction proceeds in different subjects in
different ways and leads to different results. German authors introduced the
concept of Kommunikat to denote this result of mental reconstruction of the
world of a book that is the first immediate product of reading: further, reader-
’s responses and impressions refer to Kommunikat rather than to the original
book (Viehoff, 1992). There are different ways to build the Kommunikat
from the same book or other artwork; free description strategies seem to re-
flect these variations. Some reconstruct the narrative, some emotions, some
meaning, some cultural context, some their own impressions, some the au-
thor’s or director’s technical skills and discoveries. Free description strategies
reveal a qualitative variety of individual ways of interaction with different
forms of art. Some of them are apparently more advanced and some less ad-
254 / Aesthetics and Innovation
vanced, but they cannot be ordered along this quantitative dimension pre-
cisely enough.
The present study was aimed at investigating the possibilities of free de-
scription methodology for studying professional and individual differences in
the perception of dramatic art. Despite rather small sample size, that makes it
impossible to zoom into many details, the above results allow for some quite
definite conclusions.
1. There are different ways of describing dramatic performances (free de-
scription strategies) that, on all evidences, reflect different ways of percep-
tion. The strategies are mutually independent. Some of them, described above
as narrative, emotional, personal, and cultural strategies, are stable across
time and linked to some personality variables; some lack temporal stability
and seem to be used ad hoc.
2. Emotional vs. rational, narrative oriented perception seems to be the
basic dichotomy structuring individual differences in dramatic art perception.
This evidently follows from correlational and cluster analysis data. Spectators
of the emotional type describe performances in terms of immediate emotional
impressions, characters’ features, and cultural contexts; they tend to frame
their perception on an emotional basis and are more externally oriented and
impulsive. Spectators of the rational type describe performances in terms of
the narrative and the morale, they frame their perception in a rational way,
tend to be social and present-focused.
3. Professionally educated (competent) spectators reveal somewhat differ-
ent use of strategies than lay spectators. They focus more on the technologic-
al side of the performance, its cultural context, and the experience of the spe-
cial theater space, and less on emotional impressions and characters.
The strategies characteristic of the emotional type of perception seem to
covary negatively with dramatic art competence and the strategies character-
istic of the rational type seem to be generally unrelated to it. In competent
spectators the dichotomy emotionality vs. rationality thus seems to be of
much lesser importance, giving priority to the dimension of competence.
The methodology of free descriptions thus opens new perspective for the
study of the perception of dramatic art.

Delskaya, T. A., & Nazarova, M. D. (1991). Psikhologicheskii analiz vospriy-
atiya spektaklya (Psychological Study of Play Comprehension). In L.
Dorfman, D. Leontiev, V. Petrov, & V. Sozinov (Eds.), Iskusstvo i emotsii
(Art and emotions) (pp. 232–241). Perm, Russia: Perm State Institute for
Dramatic art / 255
Arts and Culture. (In Russian).
Demarcy, R. (1973). Element d'une sociologie du spectacles. Paris: UGE.
Gachev, G. (1968). Soderzhatelnost khudozhestvennykh form: Epos. Lirika.
Teatr (Sapidity of Artistic Forms: Epos. Lyrics. Theater). Moscow: Pros-
veshchenie. (In Russian).
Dondurei, D. B. (1975). Osobennosti vospriyatiya iskusstva shkolnikami (Pe-
culiarities of art perception in schoolchildren). Sovetskaya Pedagogika, 8,
Gozman, L. Ya., & Kroz, M. Ya. (1987). Izmerenie urovnya samoaktualizatsii
lichnosti (Measurement of Person’s Self-Actualization level). In Yu. E.
Aleshina, L. Ya. Gozman, & Ye. M. Dubovskaya (Eds.), Sotsialno-psik-
hologicheskie metody issledovaniya supruzheskikh otnoshenii (Social
psychological methods of study of family relationships) (pp. 91–114). Mo-
scow: Moscow University Press.
Leontiev, D. A. (1992). What's Hecuba to us? Basic propositions for a psy-
chological theory of art. In L. Dorfman, D. Leontiev, V. Petrov, & V.
Sozinov (Eds.), Emotions and art (pp. 45–64). Perm, Russia: Perm State
Institute for Arts and Culture.
Leontiev, D. A. (1998). Vvedeniye v psikhologiyu iskusstva (Introduction to
the psychology of art). Moscow: Moscow University Press. (in Russian).
Leontiev, D. A., & Belonogova, E. V. (2002) Individualnye strategii vospriy-
atiya zhivopisi (Individual strategies of the perception of paintings). In V.
A. Koptsik (Ed.), Sinergeticheskaya paradigma: nelineinoe myshlenie v
nauke i iskusstve (Synergetic paradigm: Non-linear thinking in science
and art) (pp. 356–377). Moscow: Progress-Traditsia. (In Russian).
Leontiev, D., Delskaya, T., & Nazarova, M. (1994). Images of a theater play:
Expectations, impressions, and group differences. In C. Dreyer et al.
(Ed.), Lebenswelt: Zeichenwelt—Life world: Sign world (Vol. 2, pp. 799–
814). Lueneburg: Jansen–Verlag.
Leontiev, D. A., & Kharchevin, S. P. (1994). Types and levels of fiction com-
prehension: Assessment through free descriptions. SPIEL, 13, 1, 89–100.
Maslow, A. H. (1970). Motivation and personality. New York: Harper &
Petrova, Ju. N. (Ed.) (1982). Problemy khudozhestvennogo razvitiya lichnosti
(Problems of aesthetic development of the person). Moscow: State Insti-
tute for Art Studies. (In Russian).
Shostrom, E. (1966). The Personal Orientation Inventory manual. San Diego
CA: Educational and Industrial Testing Service.
Viehoff, R. (1992). Literary understanding and emotion. In L. Dorfman, D.
Leontiev, V. Petrov, & V. Sozinov (Eds.), Emotions and art (pp. 123–
256 / Aesthetics and Innovation
142). Perm, Russia: Perm State Institute for Arts and Culture.

The Role of Attachment Patterns in Emotional

Processing of Literary Narratives

János László and Éva Fülöp

The present study aims at investigating the emotional processing of literary

narratives in relation to the emotional organization of a readers’ self. Experi-
mental results show that securely attached subjects in terms of Bowlby’s at-
tachment theory are likely to endeavour to perceive and to get involved in a
large scale of positive and negative affects when reading literary narratives
about interpersonal conflicts, and this may increase chances of affective vari-
ations and self-perceptual shifts. Insecure subjects seem to be sensitive only
to a particular array of emotions, whereas dismissing subjects exhibit a gener-
ally low sensibility.

Reading narrative literature is double mind reading in the sense that a

reader reads the minds of the characters and her own mind. Literary texts pro-
voke this activity (László, 1999) and, as a consequence, readers not only
comprehend the text by processing the semantic information coded in the
words and sentences, but construct meaning of the text by processing their
personal experiences as well. This personal relation between text and reader
was first studied by psychoanalysts. On the ground of the variability of the
responses of different readers to the same text, and the similarity of the re-
sponses of one reader to different texts, Holland (1975) concludes that neither
the author nor the text, but it is the reader, who determines the story. He pro-
posed a four-component scheme of how readers re-create literary texts. Ac-
cording to this scheme (DEFT) readers bring literary, as well as non-literary,
cultural expectations to the text based on their prior experiences with literat-
ure and other entities in the world. Mental processes are, however, filtered
through and shaped by the characteristic defense mechanisms (e.g., identifica-
tion, projection, repression) of the reader, and are imbued by the reader's

258 / Aesthetics and Innovation
characteristic fantasies. In this process the reader continuously transforms the
text and the invested fantasies into a morally, intellectually, socially, and aes-
thetically coherent experience. Based on the ever-present personal defenses
and fantasies Holland claims that the literary work is re-created by the reader-
's personal identity, and personal identity is re-created by the literary experi-
ence (1985, p. 9). Holland (1975, p. 115) also remarked that readers preferred
to read according to their actual identity needs. In an elegant series of experi-
ments, Charlton (2003) provided evidence for this assumption. As early as
preschool children can select media stories so that they allow reference to
their own life situation.
Cognitively oriented reception studies suggested that variability of reader
responses was partly due to the thematic social knowledge readers accumu-
lated in relation to the literary narrative (László, 1988) and the culturally ac-
quired personal experiences (Larsen & László, 1990; László & Larsen,
1991). Nencini (in press) provided evidence that perception of the characters
in a novel proceeded in terms of self-perception.
The present study aims at investigating the emotional processing of liter-
ary narratives in relation to the emotional organization of readers’ self. Emo-
tional organisation of the self is a research topic in psychology which has at-
tracted several scholars in the past decades (Stern, Bowlby, Stroufe). In order
to arrive at a literary experience, readers should not only make sense of the
plot of a narrative or pursue the “landscape of action”, but also construct a
“landscape of consciousness” (Bruner, 1986). That is, they should compre-
hend what the characters think and feel. Feelings such as empathy or sym-
pathy with an author, narrator, or narrative figure are involved in the inter-
pretive processes by which a representation of the fictional world is de-
veloped and engaged (Kneepens & Zwaan, 1994). However, serving an im-
portant mimetic role within text comprehension, these narrative feelings do
not derive from the distinctively literary aspects of reading. Similarly, feel-
ings such as enjoyment, pleasure, or the satisfaction of reading are reactions
to an already interpreted text. While providing an incentive to sustain read-
ing, these feelings play no significant role in the distinctively literary aspects
of text interpretation.
As Miall and Kuiken (2002) and Kuiken, Miall, and Sikora (2004) point
out, readers react to the formal components of literary texts (narrative, stylist-
ic, or generic) by fascination, interest, or intrigue. These are the initial mo-
ment in a readers’ literary response, which serve to capture and hold the read-
ers’ attention. These aesthetic reactions interact with narrative feelings to pro-
duce metaphors of personal identification that modify self-understanding.
This level of emotional processing is called self-modifying feelings.
The role of attachment patterns in emotional processing / 259
Thus, so as to arrive at self-modifying feelings, readers should have the
capacity to generate narrative feelings. Attachment theory (Bowlby, 1969)
suggests a typology of emotional self-organization based on early mother-
child relationships, which predicts social perception for the whole course of
life. We predict that attribution of emotions, particularly social emotions, to
the characters (i.e., empathy or sympathy) in a short story is dependent on the
attachment style of the reader. This aspect of emotional processing of literary
texts has implications not only to comprehension but also to the impact a nar-
rative text can exert on a reader.

Attachment and Emotions

Bowlby’s attachment theory has been used as a framework for under-
standing interpersonal and intrapsychic processes. According to this theory
early child-parent attachment experiences are internalised into expectations
about the availability of the attachment figure in stressful situations and about
the self as worthy of love. A central notion of the theory is the ‘internal work-
ing model’ which is a mental representation of the attachment relationships
involving the self, the significant other, and the interaction. These working
models organise cognition, affect and behaviour in close relationships
throughout the life span. Collins and Read suggest that the working models
of the attachment styles are knowledge structures through which social events
are filtered and understood. “Thus individuals with different working models
may be predisposed to interpret social events in ways that are consistent with
their existing expectations and beliefs” (Collins, 1996, p. 812).
Following Bowlby’s ideas, Ainsworth (1978) distinguished three types of
attachment: secure, anxious, and avoidant. Bartholomew and Horowitz
(1991) suggested for adult attachment styles a two-dimensional, four cat-
egory model (see Fig. 1).
This model divides the avoidant attachment style into two categories. Dis-
missing (avoidant individuals detach themselves from emotional situations to
avoid the emotional involvement in interpersonal relationships), whereas
fearful (avoidant individuals feel high levels of anxiety in those situations,
what they can’t regulate and therefore they become helpless)). The authors
assume that secure attachment entails positive self-image and positive other
image, preoccupied attachment entails a negative self-image and a positive
other image, dimissing-avoidant attachment enzail a positive self-image and a
negative other image whereas anxious (fearful)-avoidant attachment entails a
negative self-image and a negative other image.
260 / Aesthetics and Innovation
Using this classification Brennen, Clark and Shaver (1998) built the Ex-
periences of Close Relationships questionnaire. Studies, such as those of
Pereg (2001) and Mikulincer and Orbach (1995) show that there are differ-
ences between the four categories in the emotional responses. Securely at-
tached people have no problems with the expression of emotions; they can
handle both positive and negative affects. Anxious people are preoccupied
with negative emotions, while avoidant (dismissing) individuals are insens-
ible of the attachment-related emotional happenings, and people categorised
as fearful reveal a chaotic emotional life (Mikulincer, Shaver, & Pereg,

Figure 14-1. Attachment styles according to Barholomew & Horowitz (1991)

Social Emotions
Social emotions can be discussed in different ways. Whereas most emo-
tions appear in interpersonal situations, there are emotions which are spe-
cially related to social interactions. In this sense, social emotions are elicited
by real, imaginary, anticipated, or remembered interactions. Social emotions
can be divided to two types: social-evaluating and social-interpersonal emo-
tions (Leary & Baumeister, 2000). In the case of social- evaluating emotions
one feels some one-way emotion towards another such as love, contempt, an-
ger, gratitude, or respect. Our social-interpersonal emotions originate from
the perception of the others’ evaluations, emotions towards us, such as jeal-
ousy, social anxiety, shame, embarrassment, and pride.
The role of attachment patterns in emotional processing / 261
According to the theory of social emotions these emotions regulate inter-
personal situations. The cognitive pattern of the self, other, and the relation-
ship will determine the nature of the emotion in the particular case. The dif-
ferent emotions mobilise certain thoughts about the other’s mental states in-
volving us and the situation (e.g., shame: you can expect to the other’s de-
valuation because of your undesirable behaviour). Accordingly, Bowlby’s in-
ternal working model can be a general frame of the representations of the self
and the other, which works as a schema in attachment situations. So the early
attachment experiences can be very important in the evaluating of the social
situations and in the emotional reaction. Thus after the estimating of the so-
cial happenings appears an emotional answer, which can influence the inter-
pretation of the subsequent events. For example the thoughts related to the
evaluating processes and the negative feelings will preoccupy an intensively
anxious person’s mind, so he/she will be not able to think a in rational way.
(Leary & Baumeister, 2000)
The other categorisation which was applied in this survey, was Plutchik’s
basic-emotion theory. Plutchik described eight basic emotions with the re-
lated situations. The primary emotions are sorrow, fear, anger, joy, trust, dis-
gust, anticipation, and surprise. (Plutchik, 1980).


Thirty-two undergraduates participated in the experiment, males and fe-

males in equal number. They were pre-selected by the Adult Attachment
Questionnaire (Brennen, Clark, & Shaver, 1998). Each attachment style (se-
cure, fearful, preoccupied, dismissing) was represented by 8 subjects.


For the purpose of the study an American short story, The Mexican Girl,
by Jack Kerouac whose theme and style very likely harmonized with the lit-
erary experiences of our prospective readers (i.e., experimental subjects). The
story describes a romantic affair between partners belonging to different so-
cial classes. The partners meet by chance, form a relationship, and eventually
part from each other. Thus, there is a large scale of of both positive and neg-
ative emotions, which may occur in the characters.
262 / Aesthetics and Innovation
The method of self-probed retrospection

This method has originally been developed in order to catch readers re-
mindings of personal experiences that go beyond the content of the text as
these are evoked during natural reading (Larsen & Seilman, 1988; Larsen &
Laszlo, 1990). In the procedure that we slightly changed, there are two
phases. In the concurrent phase, subjects are reading a text silently and
simply asked to be attentive to those occasions when any of the characters
may feel some emotion. They are asked just to put a mark at that point in the
text where the emotion occurred. In the retrospective phase, immediately
after reading is finished, the reader is questioned about each emotion by us-
ing each of his marks in the text as a probe, one at a time. Emotions are thus
"caught on the fly", as in concurrent thinking aloud, but their verbalization is
not allowed to stop reading and possibly interfere with comprehension. Since
the original emotions are recovered with readers' self-selected text locations
as probes, the method is called self-probed retrospection. With (almost)
identical cues for detecting emotions in the reading and the retrospection situ-
ation, the reader's chances of recovering what he or she was feeling and
thinking of should be high, as confirmed by Larsen and Seilman (1988).


Subjects were tested individually in a non-classroom at their school dur-

ing school hours. They were asked to mark the location of occurrence of each
emotion that a character might feel with a pencil on their copy of the story,
either by underlining or by putting a mark in the margin next to the text loca-
tion. It was stressed that they should not exert any special effort to generate
emotions, but try to read the story as they normally would. Subjects were told
that they would be asked questions about the emotions afterwards. To avoid
self censoring of emotions, they were also ensured that they would be free to
leave any question unanswered.
While the story was read, the experimenter occupied herself at another
table. After reading, the marks in the subject's copy of the text were
numbered consecutively, and for each mark in turn the subject filled in a
questionnaire that asked for the description of the emotion. In addition, the
subject rated the intensity, positivity-negativity, endurance, adequacy of the
emotion, and the extent to which the subject herself experienced the emotion.
The role of attachment patterns in emotional processing / 263
Statistical analyses were carried out for overall emotions, overall basic
emotions (Plutchik, 1980), and separately for positive and negative emotions
in general and basic emotions. Data concerning social emotions were pro-
cessed similarly, i.e., first overall social emotions than separately positive and
negative social emotions were analysed.

Frequency of emotions

The number of the attributed emotions varied from 5 to 99. The attach-
ment style had no effect for the frequency of emotions.

Frequency of positive and negative emotions

These data were analysed by a 4 X 2 ANOVA with attachment style and

sex as between-subject variables. Attachment groups had been found very
heterogeneous in the frequency of positive and negative emotions. Thus the
effect of attachment style didn’t reach significance. However, chi-square tests
for positive and negative emotions according to attachment groups yielded
significant effects. For positive emotions the Preoccupied group produced the
lowest number of affects, whereas for negative emotions the Securely at-
tached group reported the lowest number of affects, Chi-square(3) = 14.97, p
< .001; chi-square(3) = 14.34, p < .001, respectively. Results are shown in
Table 1.

Frequency of basic emotions

According to Plutchik (1980) basic emotions involve sorrow, fear, anger,

joy, trust, disgust, anticipation, and surprise. No differences were obtained in
the frequency of basic emotions.

Frequency of social emotions

Attachment style yielded no significant effect for overall social emotions,

but when social emotions were split into social positive and social negative
emotions, there was a difference between the four attachment styles.
A MANOVA showed a tendency F(3,24) = 2.7, p < .07. With pairwise com-
parisons preoccupied subjects attributed less social positive emotions than
others, F(3,24) = 2.7 p = .01, whereas dismissing individuals gave more so-
264 / Aesthetics and Innovation
cial emotions, both positive and negative, F(3,24) = 2,7, p < .05 (see
Table 2).

Table 14-1. Frequency of positive and negative emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)

Positive 108 130 86 140

M:13.5 M:17.3 M: 10.7 M: 14.8
SD: 12.3 SD: 17.3 SD: 9.3 SD: 13.9

Negative 74 124 102 118

M: 9.2 M:16.8 M: 12.7 M: 14.8
SD: 5.7 SD: 15.0 SD: 9.0 SD: 9.6

Table 14-2. Frequency of social emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)


Social 1.13 0.9 1.88 1.8 0.63 0.7 2.38 2.8


Social 1.25 0.8 1.75 2.3 2.25 1.5 3.00 2.5


Phenomenal quality of emotions

As was mentioned earlier there were 5 phenomenal qualities in which the

participants had to evaluate the emotions they marked in the story. Data were
analysed by multiple ANOVAs. For social emotions, there were significant
The role of attachment patterns in emotional processing / 265
differences between groups in all dimensions, whereas for basic emotions we
did not get significant results.


Preoccupied people found negative social emotions more intensive, the

two aviodant groups found them less intensive and secure attached reported
moderate intensity. With positive social feelings, the Preoccupied group
showed the lowest intensity, then came the Fearful, Secure, and Dismissing
groups: F(3,24) = 2,29, p < .05. With pairwise comparisons preoccupied per-
sons differs from fearful individuals. p < .05 (see Table 3).

Table 14-3. Intensity of social emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)


Social 4.31 2.7 3.80 1.6 2.10 3.1 4.9 1.4


Social 4.88 2.3 2.73 2.5 5.17 1.7 3.19 2.0


Positivity of social emotions scored by participants showed no significant

effect for attachment style.
Positivity of social emotions scored by participants showed no significant
effect for attachment style.


With positive social emotions, the Preoccupied group estimated the

shortest duration. Than the order is Fearful, Secure, and Dismissing. On the
contrary, Preoccupied subjects found negative social emotions much more
sustained than the secure and the two aviodant attachment groups, F(3,24), p
< .01. Pairwise comparisons show significant differences between Preoccu-
pied individuals and Secures and between Preoccupied and Dismissing per-
sons, p < .05 (see Table 4).
266 / Aesthetics and Innovation

On this dimension there was a significant result, F(3,24) = 2,4, p < .01.
Preoccupied individuals find positive emotions less adequate in the situations
than others, while they feel negative emotions most adequate, then Secures
and the two Avoidants. With pairwise comparisons can be found a difference
between Preoccupied and Fearful groups, p < .01 (see Table 5).

Table 14-4. Duration of social emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)


Social 3.43 2.3 3.07 1.4 0.93 1.3 3.96 0.8


Social 2.57 1.4 2.11 1.7 3.99 2.0 2.41 1.5


Table 14-5. Adequacy of social emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)


Social 4.38 2.8 4.30 1.8 1.81 2.6 4.57 1.0


Social 3.6 1.8 2.41 2.1 5.26 1.5 3.25 2.2

The role of attachment patterns in emotional processing / 267
Experiencing social emotions

On the fifth dimension participants had to value to what extent they ex-
perienced the emotion they attributed to the character. In this dimension a
tendency was found, F(3,24) = 1.73, p < .10. With positive feelings Secures
showed the greatest experience, Preoccupied showed the less and the two
Avoidants are in the middle. With negative emotions preoccupied felt the
greatest empathy, then comes secures and avoidants (see Table 6). With pair-
wise comparisons Preoccupied individuals differ from Fearful persons, p <

Table 14-6. Experiencing social emotions

Secure Fearful Preoccupied Dismissing

(N=8) (N=8) (N=8) (N=8)


Social 3.41 2.3 2.64 1.5 1.62 2.6 2.95 1.4


Social 2.37 1.7 1.05 0.9 2.76 1.0 2.06 1.7


Overall results support our basic assumption that reader’s attachment
style, i.e., emotional self-organization influences how readers perceive social
interaction between characters in Kerouac’s short story, The Mexican Girl.
One of our salient results is that attachment style predicts not so much emo-
tional processing in general, but processing specifically social emotions. On
the other hand this result seems to be plausible, because adult attachment is
directly related to interpersonal relations.
Securely attached subjects who are able to openly express their positive as
well as negative feelings in their social relations and are able to control their
emotions showed a moderate emotional involvement in the text. They attrib-
uted the lowest number of negative emotions and showed the highest identi-
fication with positive emotions. They also took an intermediate position
268 / Aesthetics and Innovation
between the two insecure groups when judgeing emotional intensity, duration
and adequacy. Overall results show that securely attached subjects are open
to both positive and negative emotions, however emotions do not overwhelm
their aesthetic cognition.
Preoccupied subjects differ from the other groups on almost each dimen-
sion. They report on less positive emotions and this deficit is even more
marked with social emotions. They experience negative affects as being more
intensive and having longer duration. They evaluate the adequacy of negative
affects higher and experience themselves more the negative affects of the
characters. On the contrary, perceived positive emotions, whether social or
non-social, seem to be low in intensity, are perfunctory, and less adequate.
They show less empathy to positive feelings.
Character perception of the preoccupied subjects also reflects negative ex-
periences. They filter out positive feelings, whereas they show high sensitiv-
ity to negative emotions. Phenomenal quality of their perceived negative
emotions seems also to be different from that of the other groups. They tend
to absorb negative emotions, which, in turn, increases sensitivity to attribut-
ing negative emotions to the characters.
Dismissing subjects neglect negative emotions in as much as they per-
ceive them as less intensive, less adequate, and having shorter duration as
compared to the other groups’ perceptions. This under-evaluation can also be
observed with the frequently mentioned positive emotions, however to a less-
er extent. The number of the perceived positive emotions, particularly posit-
ive social emotions is relatively high in this group, but data with the phenom-
enal qualities of emotions show that they experience these emotions on a low
scale. Fearful subjects exhibited a similar response pattern to that of the dis-
missing group.
These findings show an interesting relation with Cupchik’s (2005) results.
He supposes that individuals who feel too much anxiety during the reading of
a literary work because of the emotional involvement, turn their attention to
other, stylistic features of the text (Cupchik, 2005). In our study preoccupied
persons could not defend against this anxiety—as their evaluations in phe-
nomenal qualities show—they experienced much more negative emotions
and less positives as the others. Contrary to them, secure and avoidant indi-
viduals who attributed emotions in about equal number, were not preoccupied
by these negative feelings.
Although it was not the primary target of our study, the relation between
emotional self-organization and generating narrative emotions has implica-
tions for the impact a literary narrative can exert on the reader, including the
influences on himself. Kuiken et al. (2004) studied self-modifying feelings
The role of attachment patterns in emotional processing / 269
during literary reading in relation to absorption, a personality trait which is
highly relevant for emotional involvement. Participants read a short story, de-
scribed their experience of three striking or evocative passages in the story,
and completed the Tellegen Absorption Scale (Tellegen, 1982). Compared to
readers with either low or moderate absorption scores, those high in absorp-
tion were more likely to report affective theme variations and self-perceptual
shifts, especially during an emotionally complicated portion of the story. At-
tachment styles work probably in a similar manner. Securely attached sub-
jects are likely to endeavour to perceive and to get involved in a large scale of
positive and negative affects when reading literary narratives about interper-
sonal conflicts, and this may increase chances of affective variations and self-
perceptual shifts. Insecure subjects seem to be sensitive only to a particular
array of emotions, whereas dismissing subjects exhibit a generally low sens-
ibility. It should not mean, however, that self-modifying feelings would not
occur either with insecure or with dismissing subjects when reading literary
narratives. It only means that emergence of this type of feelings would be
more likely in relation to emotionally negative aspects of the self for insecure
subjects, whereas the likelihood of self-involvement, because of the neglect
even for narrative emotions, would be lower for dismissing subjects. This re-
lationship has implications for bibliotherapy, whose aim is to modify the self
through literary experience.

Ainswoth, M., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of at-
tachment: A psychological study of the strange situation. Hillsdalr, NJ:
Bartholomew, K., & Horowitz, L.M. (1991). Attachment style among young
adults: A test of a four-category model. Journal of Personality and Social
Psychology, 61, 226-244.
Bowlby, J. (1969/1982). Attachment and loss, Vol. 1. Attachment. New York:
Basic Books.
Bowlby, J. (1973). Attachment and loss, Vol. 2. Separation, anxiety, and an-
ger. New York: Basic Books.
Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measure-
ment of adult
attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.),
Attachment theory and close relationships (pp. 46-76). New York: Guil-
ford Press.
270 / Aesthetics and Innovation
Charlton, M. (2003). How children appropriate themes they find in media
products. Empirical Studies of the Arts, 21, 81-96.
Collins, N. L. (1996). Working models of attachment: Implications for ex-
planation, emotion, and behaviour. Journal of Personality and Social
Psychology, 71, 810-832.
Cupchik, G. C. (2005). Emotion in aesthetics and the aesthetics of emotions.
In P. Locher, C. Martindale, & L. Dorfman (Eds.), New directions in aes-
thetics, creativity and the arts (pp. 209-224). Amityville, NY: Baywood.
Holland, N. (1975). 5 readers reading. New Haven: Yale University Press.
Kerouac, J. (1961). On the road. London: Pan Books.
Kuiken, D., Miall, D. S., & Sikora, S. (2004). Forms of self-implication in lit-
erary reading. Poetics Today, 25, 171-203.
Kuiken, D., Phillips, L., Gregus, M., Miall, D. S., Verbitsky, M., &
Tonkonogy, A. (2004). Locating self-modifying feelings within literary
reading. Discourse Processes, 38, 267-286.
Larsen, S. F., & László, J. (1990). Cultural-historical knowledge and personal
experience in appreciation of literature. European Journal of Social Psy-
chology, 20, 425-440.
Larsen, S. F., & Seilman, U. (1988). Personal remindings while reading liter-
ature. Text, 8, 411-429.
László, J. (1988). Readers’ historical-social knowledge and their interpreta-
tion and evaluation of a short story. Poetics, 17, 461-481.
László, J. (1999) Cognition and representation in literature. The psychology
of literary narratives. Budapest: Akadémiai Kiadó.
László, J., & Larsen, S. F. (1991). Cultural and text variables in processing
personal experiences while reading literature. Empirical Studies of the
Arts, 9, 23-34.
Miall, D. S., & Kuiken, D. (2002). A feeling for fiction: Becoming what we
behold. Poetics, 30, 221-241.
Leary, M. R., & Baumeister, R. F. (2000). The nature and function of self-es-
teem: Sociometer theory. In M. Zanna (Ed.), Advances in experimental
social psychology (Vol. 32, pp. 1-62). San Diego, CA: Academic Press.
Mikulincer, M., & Orbach, I. (1995). Attachment styles and repressive de-
fensiveness: The accessibility and architecture of affective memories.
Journal of Personality and Social Psychology, 68, 5, 917-925.
Mikulincer, M., Shaver, P. R., & Pereg, D. (2003). Attachment theory and af-
fect regulation: The dynamics, development, and the cognitive con-
sequences of attachment-related strategies. Motivation and Emotion, 27,
The role of attachment patterns in emotional processing / 271
Nencini, A. (2005). The reader at work. The role of the text and text-receiver
in the construction of the protagonist of a novel. Empirical Studies of the
Arts. (In press).
Plutchik, R., & Kellerman, H. (Eds.) (1980). Emotions: Theory, research, and
experience (Vol. 1). New York: Academic Press.
Tellegen, A. (1982) Content categories: Absorption items. Unpublished
manuscript. University of Minnesota.



Pictures in the Mind:

Symmetry and Projections in Drawings

Diane Humphrey and Dorothy Washburn

This study examines drawings from life, memory, fantasy, and diagrams cre-
ated by art students and students untrained in art. The use of seven one-di-
mensional symmetry patterns as well as of station points and metric, affine,
projective, divergent, and other projections was analyzed. Art students were
less likely to use symmetry in life drawings, but otherwise did not differ sig-
nificantly from untrained students. All participants made the least use of ver-
tical reflection with bifold rotation and glide reflection in their drawings. Art
students, particularly the males, made greater use of projective projections in
drawings, compared to other participants. For all participants considered to-
gether, the use of less conventional types of projections was negatively cor-
related with the use of symmetries. The use of divergent projections, on the
other hand, was positively correlated with the use of symmetries, especially
vertical reflection and glide reflection.

Drawings as structured representations of objects, surfaces and environ-

ments, vary in method and function. In increasing levels of abstractness we
can list life drawing, drawing from memory, fantasy drawing, and diagrams
as some examples of kinds of drawings. Each of these has a different purpose
and therefore should have different characteristics. Life drawings should have
the most perceptual resemblance to the objects portrayed while diagrams
should have the least with memory and fantasy drawings somewhere in
between. Life drawings should have the fewest geometric elements, while
diagrams should have the most geometric elements, again with memory and
fantasy drawings somewhere in between.
Drawings in general have been considered in two very different ways.
Arnheim (1974, 1988) has looked at visual art as consisting of visual dynam-

274 / Aesthetics and Innovation
ic structures. Following this approach leads to consideration of Gestalt group-
ing principles such as symmetry, (for example Humphrey, 1997; Kubovy &
Strother, 2004; Washburn & Crowe, 1988). Kennedy (1974, 1993) and Wil-
lats (1997), on the other hand, see drawings as representations of objects. In
this latter approach we might use projective systems, such as those described
in art by Hagen (1988) and consider station points (Gibson, 1966) in the rep-
resentation of objects in depth. Now, if we look at symmetries not as Gestalt
entities, but as properties of objects, as in an object-representation approach
to drawing, then symmetry can be considered as a “cue to depth” (Kont-
sevich, 1996). Wagemans (1993), for example, has demonstrated that skewed
symmetries act as “cues to depth”. But if symmetry is considered in a dynam-
ic-structure approach, symmetry of a structure is a cue to flatness, or more
generally a cue to topology, an indicator of a surface. Perhaps the symmetries
used in many cultures (Washburn & Crowe, 1988) to decorate various sur-
faces such as textiles and pottery provide cues to the surface layout of the
decorated objects (such as pottery), whether rigid or otherwise (such as tex-
tiles). Thus, while the symmetry of an object can act as a cue to the depth of
an object, symmetry of a pattern on a flat surface acts as a cue to the flatness
of the surface.
Different types of drawings should differ in how much and what kind of
use they make of object-representation principles, such as projections and
station points. Hagen (1988) has described various kinds of projection sys-
tems used in different types and styles of artwork (see Fig. 1). In the present
study it is predicted that life drawings should use more projective projections,
while diagrams should use other kinds of projections.
Drawing types should also differ in how they use dynamic-structure prin-
ciples. Washburn and Crowe (1988) have described a number of classifica-
tions of symmetry patterns used in various cultures (see Fig. 2). It is pre-
dicted that diagrams should have the most use of these types of symmetry
while life drawings should have the least.
The present study looks at four kinds of drawings; from life, memory,
fantasy, and diagrams in terms of their use of symmetries, projections and
station points. Art students participated in the study as did students untrained
in art. Art students should have more fluency with the various types of draw-
ings used here, while the untrained participants should be more rigid in their
use of drawing techniques, showing less variation across drawing types. Pre-
vious research (see Humphrey, 2002, for a review) has shown that females
more often prefer certain types of one-dimensional symmetries than do
males. We might thus expect a female proclivity for use of symmetry in
drawings. Males, on the other hand, might show greater use of projective pro-
Pictures in the Mind / 275
jections, as expected from their known superiority in three-dimensional spa-
tial tasks (Kimura, 1999).

Figure 15-1. Illustrations of three types of projections discussed by Hagen (1988):

a) metric (orthogonal), b) affine (parallel), c) projective (perspective).


Nine female (mean age = 22.2 years, SD = 4.8) and nine male (mean age
= 21.3 years, SD = 1.3) art students at Maryland Institute, College of Art and
25 female (mean age = 19.2 years, SD = .82) and 16 male (mean age = 20.4
years, SD = 3.2) non—art students at King’s University College at the Uni-
276 / Aesthetics and Innovation
versity of Western Ontario served as participants. King’s students received
bonus credits in their Introductory Psychology course for an assignment re-
lated to their participation.

Figure 15-2. Illustrations of seven one-dimensional band patterns as described by

Washburn and Crowe (1988): 1. p111 translation, 2. pm11 vertical reflection, 3. p1m1
horizontal refection, 4. pmm2 vertical reflection, horizontal reflection, and bifold rota-
tion, 5. p112 bifold rotation, 6. pma2 vertical reflection and bifold rotation, 7. p1a1
glide reflection.


Each participant was provided with a black marker (PMOP Series 860
Nylon Fiber Point Porous Pen- 863-11) and four sheets of blank white paper
8 ½ by 11 inches.
Pictures in the Mind / 277

Either individually, or at individual desks, participants were first asked to

draw something from life, then something from memory, then something
from fantasy, then a diagram, using a subject of their own choosing for each
drawing. Participants were allowed to interpret the four categories of drawing
in any way they chose, but were prompted with alternative wording for each
category, if requested, as follows. Life , “something you can see right now”;
memory “something you remember seeing”; fantasy, “ something you can
imagine”; diagram, “ a plan for something”. Participants were given approx-
imately fifteen minutes to complete each drawing, then they were asked to go
on to the next.


Drawings were analyzed for the use of symmetries, station points, and
projections, as described below.
Symmetries. Drawings were analyzed for the presence or absence of the
seven one-dimensional band patterns from Washburn and Crowe (1988) as il-
lustrated in Figure 2.
Projections. Drawings were analyzed for their use and the frequency in
each picture of a variety of kinds of projections, as defined by Hagen (1988),
some of which are illustrated in Figure 1. The five projection categories used
here are metric (orthogonal), affine (parallel), projective (perspective), diver-
gent (reverse perspective) and any other projections used in the drawing.
Station points. Drawings were analyzed for the number of station points,
as defined by Gibson (1966). A station point refers to the place where rays of
light from an optic array converge. In any particular environment there are in-
finite numbers of possible station points. Drawings, however, are constructed
as if from a limited number of station points, and sometimes from only one
station point.

Art students used many more marker strokes and in a more complex fash-
ion than did untrained students. Their drawings were also more diverse in
style and more realistic than those by untrained students.
278 / Aesthetics and Innovation
Symmetry Types in All Drawings

Statistical significance is considered to be p < .05. In a Manova with pro-

gram (art vs non—art) and sex as between-subject variables and types of
drawings as a within subject variable there was a significant difference
between art and non—art students in the total numbers of symmetries used
(see Fig. 3), with non—art students using more kinds of symmetries than did
art students, T = .43, F(4,52) = 5.6, p = .001, particularly on life, F(1,55) =
14.9, p < .001, and fantasy, F(1,55) = 33.8, p <. 05, drawings.

Symmetry Types in Life Drawings

In a Manova with the number of drawings showing each of seven sym-

metry types as the dependent variable and sex and program (art vs non—art)
as group variables there was a significant effect of program, T = .45, F(7,49)
= 3.1, p < .05. Non—art students made significantly more drawings (see Fig.
1) with translation, vertical reflection, horizontal reflection, vertical and hori-
zontal reflection, and bifold rotation, F(1.55) = 9.6; 16.8; 11.5; 10.2; 8.6 re-
spectively, p < .01.

Symmetry Types in Memory Drawings

There were no effects of sex or program when memory drawings were

analyzed for number of drawings showing each of the seven symmetry types.

Symmetry Types in Fantasy Drawings

There were no effects of sex or program when fantasy drawings were ana-
lyzed for number of drawings showing each of the symmetry types.

Symmetry Types in Diagrams

There were no effects of sex or program when diagrams were analyzed

for number of drawings showing each of the symmetry types.

Symmetry Trends in Drawings

Art Students. In a Manova of art students’ data only, with type of drawing
and type of symmetry as within subjects variables, there were significant
overall effects of type of drawing, T = 1.2, F(3,15) = 5.9, p < .01, and of type
of symmetry, T = 2.2, F(6,12) = 4.4, p < .05. There were significant linear
Pictures in the Mind / 279
and cubic trends across types of drawings, F(1,19) = 13.1 and 7.1, p < .01,
.05, respectively. There were significant linear, quadratic, and fifth order

Figure 15-3. Symmetries used in drawings by art students and non-art


trends across types of symmetry, F(1,19) = 21.2, 5.8, 17.8, p < .001, .05,
.001, respectively.
Non—art Students. In a Manova of non—art students’ data only, with type
of drawing and type of symmetry as within subject variables, there were sig-
nificant effects of type of symmetry, T = 3.9, F(6,35) = 22.5, p < .001, and of
the interaction between type of symmetry and type of drawing, T = 2.4,
F(18,23) = 3.1, p < .01. There were significant linear, quadratic, cubic, fifth
280 / Aesthetics and Innovation
order and sixth order trends across types of symmetry, F(1,40) = 87.3, 23.7,
34.5, 113.2, and 14.9 respectively, p < .001.

Projections and Station Points

Station Points. In a Manova with number of Station Points in drawings of

life, memory, fantasy and diagrams with sex and program as grouping vari-
ables, there was a significant effect of program, T = .31, F(4,52) = 4.13, p <
.01. Art students used significantly more station points in memory drawings
and diagrams, F(1,55) = 6.9, 7.7, p < .05, .01 respectively, than did non—art
Projections in Life Drawings. In a Manova with numbers of life drawings
using metric, affine, projective, divergent and other projections as dependent
measures and sex and program as grouping variables, there was a significant
effect of sex, T = .31, F(5,51) = 3.25, p < .05, and a significant interaction
between program and sex, T = .24, F(5,51) = 2.48, p < .05 (see Fig. 4). Sig-
nificantly more divergent projections were produced by females than by
males in life drawings, F(1,55) = 5.12, p < .05. On the other hand, signific-
antly more “other” projections were produced by males than by females in
life drawings, F(1,55) = 12.21, p < .001. In a significant interaction between
program and sex, more female art students produced metric projections than
did male art students, while slightly more male non—art students produced
metric projections than did female non-art students, F(1,55) = 4.81, p < .05.
For projective projections, however the interaction was in the opposite direc-
tion. While more art males produced projective projections, it was the non-art
females who used projective projections, F(1,55) = 5.15, p < .05.
Projections in Memory Drawings. In a Manova with numbers of memory
drawings using metric, affine, projective, divergent and other projections as
dependent measures and sex and program as grouping variables, there was a
significant effect of program, T = .72, F(5, 51) = 7.37, p < .001. Significantly
more art students than non-art students used projective projections in memory
drawings, F(1,55) = 24.3, p < .001. In fact, non-art students did not use any
projective projections in memory drawings.
Projections in Fantasy Drawings. In a Manova with numbers of fantasy
drawings using metric, affine, projective, divergent and other projections as
dependent measures and sex and program as grouping variables, there was a
significant effect of program, T = .27, F(5, 51) = 2.7, p < .05. Significantly
more art students than non-art students used projective projections in fantasy
drawings, F(1,55) = 7.7, p < .01. In fact, non—art students did not use any
projective projections in fantasy drawings.
Pictures in the Mind / 281

Figure 15-4. Projections used in drawings by art students and non-art


Projections in Diagrams. In a Manova with numbers of diagrams using

metric, affine, projective, divergent and other projections as dependent meas-
ures and sex and program as grouping variables, there was a significant inter-
action between program and sex, T= .26, F(5,51) = 2.67, p < .05. While more
art females produced diagrams with affine projections, it was the non-art
males who produced more affine projections, F(1,55) = 7.14, p < 01. For pro-
jective projections in diagrams, however, the art males produced more than
did the art females, but non-art students produced none, F(1,55) = 4.5, p <
282 / Aesthetics and Innovation
Projection Trends in Drawings

Art Students. In a Manova of art students’ data only, with type of drawing
and type of projection as within subjects variables, there were significant
overall effects of type of drawing, T = .9, F(3,15) = 4.5, p < .05, and of type
of projection, T = 4.4, F(4,14) = 15.4, p < .001, and of the interaction
between type of drawing and type of projection, T = 17.9, F(12,6) = 9.00, p <
.01. There was a significant linear trend across types of drawings, F(1,17) =
7.7, p < .05. There were significant linear, quadratic, cubic and quadratic
trends across types of projections, F(1,17) = 16.6, 28.5, 17, 35.8 respectively,
p < .001. There were also significant interaction trends, specifically, linear
linear, linear quadratic and linear cubic, F(1,17)= 25.6, 17.3, 15.5, respect-
ively, p < .001.
Non-art Students. In a Manova of non-art students’ data only, with type of
drawing and type of projection as within subject variables, there were signi-
ficant effects of type of drawing, T = .36, F(3,38) = 4.6, p < .01, of type of
projection, T = 3.2, F(4,37) = 29.3, p < .001, and of the interaction between
type of drawing and type of symmetry, T = 3.5, F(12,29) = 8.5, p < .001.
There was a significant linear trend across types of drawings, F(1,40) = 13.9,
p < .001. There were significant linear, quadratic, cubic, and fourth order
trends across types of projection, F(1,40) = 94.2, 110.6, 51.6, and 5.8, p <
.001, .001, .001, .05 respectively. There were also significant interaction
trends, specifically, linear linear, linear quadratic and linear cubic, F(1,40) =
33.1, 46.3, 9.7, p < .001, .001, .01 respectively.

Comparisons Between Drawing Types

Metric Projections. In a Manova on metric projections with type of draw-

ing as a within subject variable and, program (art vs nonart) and sex as
between subject variables, there was a significant effect of drawing type, T =
1.8 F(3,53) = 31.6, p < .001, and a significant linear trend across drawing
types, F(1,55) = 7.6, p < .001.

Correlations between Symmetries and Projections

As can be seen in Table 1, the use of divergent projections was somewhat

positively correlated with the use of most symmetries, except translation,
when all types of drawings are considered together. The use of “other” pro-
jections was negatively correlated with the use of all types of symmetries in
Pictures in the Mind / 283
Table 15-1. Correlations between symmetries and projections in drawings

Symmetry Metric Affine Projective Divegent Other

p111 translation .08 .10 .08 .07 -.53

pm11 v reflection .08 .05 -.05 .27 -.50
p1m1 h reflection .05 -.01 .04 .21 -.50
vh ref
pmm2 bifold .11 -.09 .06 .18 -.44
p112 bifold rot .12 -.09 .05 .23 -.42
pma2 v ref bifold .17 .18 -.06 .20 -.25
p1a1 glide ref .11 -.02 -.08 .34 -.21

The drawings made by art students here are qualitatively different from
those made by untrained students. They appear to be more elaborated and
complex, filling more of the paper and with greater numbers of marker
strokes. They also appear to be more realistic.
There were differences between art and non-art drawings on measures of
one-dimensional symmetry in most types of drawings. Non-art students used
many more symmetries in their drawings, particularly in life drawings. Art
students’ use of symmetry in drawings is particularly obvious only in dia-
grams. Thus, art students used symmetrical “cues to flatness” mostly in dia-
grams, as one might expect. Untrained students, however, used them in all of
their drawings, which makes them look flatter and less realistic.
Art students used more station points particularly in memory drawings
and diagrams. We might expect more station points in diagrams, as these use
more metric projections, requiring a different station point for each object or
area in the diagram. Indeed we can see trends across drawing types of life,
memory, fantasy and diagrams that show greater use of metric projections
and multiple station points with increasing abstractness of the drawing type.
But these trends are particularly clear for the drawings by art students. Un-
trained students, on the other hand, show the greatest use of metric projec-
tions and multiple station points in fantasy drawings, followed by diagrams.
284 / Aesthetics and Innovation
Perhaps untrained students do not understand the nature and purpose of dia-
In general, all participants used more metric projections than any other
kind. These were most prevalent in diagrams and fantasy drawings and least
prevalent in life and memory drawings, as we might have expected.
The use of projections differed with both sex and training, particularly in
life drawings. While females used more divergent projections in life draw-
ings, males used more unconventional projections in life drawings that did
not fit into any of the categories of projections we used.
The greatest use of projective projections in life drawings was made by
males trained in art. While females trained in art used more metric projec-
tions in life drawings, it was untrained males who used more metric projec-
tions in life drawings.
Projective projections were used by untrained students only in life draw-
ings. They were never used by these students in any other type of drawing.
Art students actually used projective projections in all types of drawings. Per-
haps the use of these technical projections from memory requires specific
training. Non-art students seldom used projections other than metric in most
of their drawings. The few projections they used may have been attempts at
projective geometry that failed. Their impoverished use of projections cer-
tainly contributed to the less realistic aspect of their drawings.
Diagrams frequently did not look like diagrams when made by untrained
students. But this does not mean that art students always produced the kind of
diagrams we might have expected. In fact, other than metric, projective pro-
jections were the most frequently used in diagrams by art students, particu-
larly male art students. Art females on the other hand, showed a slightly
greater proclivity than their male colleagues for affine projections in dia-
grams, while this sex difference was the opposite for untrained students. The
meaning and function of diagrams appears to be different for different stu-
dents. Many participants seemed to be just drawing another picture. While
there was some evidence for the use of metric projections, symmetries and
multiple station points in diagrams, these were mostly, but not exclusively,
used by those trained in art. Diagrams may be a type of drawing requiring
specific training.
In summary, there is an indication of increasing abstractness and decreas-
ing realism across the types of drawings used in this study. Students un-
trained in art seemed to be less fluent in the differing types of drawings. Non-
etheless, there were some surprising similarities between the art and non-art
students’ drawings in terms of the use of projections in particular. Symmet-
Pictures in the Mind / 285
ries, however, were much less often used by art students in all types of draw-
ings except diagrams.
Art students made their drawings more realistic by the use of projective
projections as a cue to depth and by limited or very little use of symmetry as
a cue to flatness. They did not even use skewed symmetry as a cue to depth.
There are certainly other techniques than the few studied here that are avail-
able to artists to use in drawings for various purposes. This study is the begin-
ning of explorations of how artists and others make pictures.
Finally, it is interesting that sex differences were obtained in the use of
projections, but not in the use of symmetries. Previously, Humphrey (1997)
found a female proclivity for the use of symmetries in fantasy drawings made
with colored markers by untrained adults and children.
The use of projections, however, showed a proclivity on the part of art
males for the use of projective projections, both in life drawings and in dia-
grams. We would expect males to prefer projective drawings on the basis of
known sex differences in three-dimensional spatial tasks such as targeting,
navigating, and mental rotation (Kimura, 1999). Why this is particularly ob-
vious in males trained in art and not in untrained males is unclear. Perhaps
there is a floor effect here resulting from the fact that untrained students are
simply not fluent enough in drawing skills to show these three-dimensional
spatial sex differences.

We would like to thank John Kindree for drawing collection and analyses
and King’s University College at the University of Western Ontario for fund-
ing of this research.

Arnheim, R. (1988). The power of the center: The new version. Berkeley:
University of California Press.
Arnheim, R. (1974). Art and visual perception: The new version. Berkeley:
University of California Press.
Gibson, J. J. (1966). The senses considered as perceptual systems. Boston:
Houghton Mifflin.
Hagen, M. (1988). Varieties of realism. Cambridge: Cambridge University
286 / Aesthetics and Innovation
Humphrey, D. (2004). Symmetries in development: The eye is quicker than
the hand (pp. 7–18). In D. Washburn (Ed.) Embedded symmetries: Natur-
al and cultural. Amerind New World Series. Albuqurque: University of
New Mexico Press.
Humphrey, D. (1997). Preferences in symmetries and symmetries in draw-
ings: Asymmetries between ages and sexes. Empirical Studies of the Arts,
15, 41–60.
Kennedy, J. M. (1993). Drawing and the blind: Pictures to touch. New
Haven: Yale University Press.
Kennedy, J. M. (1974). A psychology of picture perception. San Francisco:
Kimura, D. (1999). Sex and cognition. Cambridge, MA: MIT Press.
Kontsevich, L. (1996). Symmetry as a depth cue. In C. Tyler (Ed.), Human
symmetry perception (pp. 331–347), VSP.
Kubovy, M., & Strother, L. (2004). The perception of friezes: Going beyond
geometry (pp. 19–26). In D. Washburn (Ed.), Embedded symmetries: Nat-
ural and cultural. Amerind New World Series. Albuqurque: University of
New Mexico Press.
Wagemans, J. (1993). Skewed symmetry: A non-accidental property used to
perceive visual forms. Journal of Experimental Psychology: Human Per-
ception and Performance, 19, 364–380.
Washburn, D., & Crowe, D. (1988). Symmetries of culture. Seattle: Uni-
versity of Washington Press.
Willats, J. (1997). Art and representation. Princeton NJ: Princeton University

Automatic Affective Evaluation of Pictures and


Stefano Mastandrea

The present work deals with the automatic processing of pictures and words
associated with a particular type of affective meaning. In Experiment 1, pic-
tures were associated with the words according to an affective quality of the
objects represented. Participants had to perform an affective categorization
task (positive/negative) of thirty pairs of stimuli composed of pictures and
words linked in three different associations: congruent, incongruent and neut-
ral, using the picture/word variant of the Stroop paradigm. Results show
evidence of automatic affective processing: latency times were influenced by
incongruent distractors, and pictures were faster than words. In Experiment 2
we wanted to see if automatic evaluation would occur even without an expli-
cit evaluative goal. Ten pictures presented as a prime were associated to three
words: the name of the picture, an affective meaning of the picture, and a
neutral word. Participants had to respond to a lexical decision task. Results
show that latency times of names were faster than those of emotions and
neutral words; moreover, positive emotions were processed faster than negat-
ive emotions and neutral words. In both experiments there is clear evidence
of an asymmetry concerning the valence of the stimuli with positive stimuli
being processed faster than negative ones.

We are constantly surrounded by objects of any type to which we con-

sciously attribute different meaning (semantic, affective, aesthetic). One im-
portant aspect of our relationship and comprehension of pictures and objects
is to see if even our iconic world, which often needs to be interpreted sym-
bolically and metaphorically, can be processed automatically.
The role attributed to automatic processes is becoming increasingly im-
portant in different areas of contemporary research going from perception to

288 / Aesthetics and Innovation
emotion, from affect to attitudes. Bargh (1989) states that the distinction
between automatic and controlled processes is not always clear; automatic
processes can be characterized by the fact of being non-intentional, uncon-
trollable, efficient, and unaware, even though not all automatic processes
present these features.
Automatic affective processing can be defined as the capacity for an or-
ganism to automatically distinguish external stimuli into two large dichotom-
ic categories, good or bad, pleasant or unpleasant, positive or negative (Bar-
gh, Litt, Pratto & Spielman, 1989; Klauer, 1998; Klauer & Musch 2003; Öh-
man, 1987; Zajonc, 1980). If we see a spider or a flower, do we have to think
intentionally about what we think of it or can the associated affective evalu-
ation (good/bad; pleasant/unpleasant; positive/negative) be aroused uncon-
sciously and effortlessly? Experimental evidence sustains that evaluative as-
sociations stored in long term memory are automatically prompted any time
we encounter an attitude object. Research into automatic affective processing
has received a boost from the seminal work by Fazio, Sanbonmatsu, Powel
and Kardes (1986) on affective priming. The work by Fazio et al. (1986) was
rooted in studies of semantic priming. Meyer and Schvaneveldt (1971) ob-
served that the presentation of a prime word (robin) speeded up the recogni-
tion of a semantically related target (swallow). Affective priming consists of
attributing a positive or negative evaluation to a target word preceded by an
affective prime: if prime and target are congruent (e.g. sun—holiday), affect-
ive categorization will be facilitated.
Research on automatic evaluation has focused, with few exception, on
words and verbal language. In the late 1940s, the North American ‘New
Look’ movement focused on the importance of motivational, affective and
emotional components referred to individual differences in automatic pro-
cesses in visual perception (Bruner & Postman, 1947; McGinnies, 1949).
After that very few researchers dealt with pictures and a major part of studies
on picture processing used pictures of human, in particular facial expressions
(Hansen & Hansen, 1988; Murphy & Zajonc, 1993; Niedenthal, 1992; Öh-
man, 2002; Stenberg, Wilkin & Dahl, 1998; White, 1995).
De Houwer and Hermans (1994) explored automatic processing of pic-
tures with affective meanings using the Stroop paradigm in the picture—
word version. They used pictures of six animals (taken from Snodgrass &
Vanderwart, 1980), three positive (duck, rabbit, and bird) and three negative
ones (spider, snake, and beetle), with the corresponding name overimposed.
In the picture—word interference task the affective categorization of the
words was delayed in case of valence incongruity between word and picture,
while the words did not cause any delay in the affective categorization of the
Automatic Affective Evaluation / 289
picture. In another work on affective priming, Hermans, De Houwer & Eelen
(1994) used mainly human coloured pictures with a positive and negative
valence (from the international affective picture system by Lang, Öhman &
Vaitl, 1988) belonging to different categories. Results demonstrated that tar-
get pictures preceded by extreme affectively congruent picture primes were
evaluated faster than in incongruent or neutral conditions.
Giner-Sorolla, Garcia & Bargh (1999) extends the concept of automatic
evaluation to less extreme valenced pictures presented as prime stimuli in or-
der to influence latency time of the picture targets; moreover excluding expli-
cit evaluation purposes (Bargh, 1989), Giner-Sorolla et al. (1999) found that
also in a pronunciation task, word valenced targets were influenced by more
and less extreme picture primes.
The present work has the aim to extend the study of automatic affective
evaluation also to pictures of everyday objects in which the pictures are not
associated with the words normally used to name them, but with words asso-
ciated due to an attribute, characteristic or affective quality of the objects. For
example, the picture of a bidimensional red heart was associated not with the
word “heart”, but with the word “love”», as a possible affective meaning rep-
resented by the picture; or the picture of a cup (sport prize) associated with
the word “victory.”

Experiment 1



Seventy Italian participants, aged between 20 and 30, participated to the

experiment as a partial fulfillment of course requirements.


Simple colored pictures of common non-human objects, to assess whether

pictures and words with an affective relationship, as specified above, can be
automatically processed, were used. Ten affective qualities (emotions and
emotional antecedents) and their representation through pictures were con-
sidered. The ten affective qualities were subdivided into five pairs with an
opposing positive and negative valence: happiness/sadness, success/illness,
love/fear, victory/danger, interest/disgust. A picture of balloons was chosen to
depict happiness; a cemetery for sadness; a bottle of champagne for success;
290 / Aesthetics and Innovation
red heart for love, a red cross for illness; a bomb for fear; a sports cup for
victory; a triangular sign for danger; a boxed gift for interest and an insect
for disgust (see Fig. 1). Pictures and words of a neutral valence belonging to
furniture and vegetable categories were added as control, neutral stimuli.
Picture pairs were balanced by referring to structural characteristics; in
particular, complexity (simply defined as the number of single elements
present in the picture) and chromatic features (number of colors for every
The expressive characteristics of each picture was checked by asking 18
pilot participants to rate the ten affective characteristics (happiness, sadness,
etc.) represented by each picture on a 5-point scale; each expressive quality
associated to the corresponding picture (happiness/BALLOONS,
sadness/CEMETERY, etc.) reached a high mean evaluation score (from 4.1 to
Each item consisted of a pair of stimuli composed of a picture and a
word; in each pair of stimuli, the relationship between picture and word re-
ferred to three possible relationships: congruent, when picture and word had
the same valence, either positive or negative; incongruent, when picture and
word had the opposite valence; neutral, when the control stimulus had a neut-
ral value. Pictures were associated with thirty words to give a total of thirty
pairs of stimuli.
The pairs of stimuli were subdivided into two groups: Picture condition,
when the task required was the affective categorization of the picture; in this
case the picture was the target and the word was the distractor. Each of the
ten pictures, positive and negative, was associated with three words, in order
to obtain a relationship of congruence, incongruence and neutrality, for a total
of thirty pairs of stimuli. Word condition, when the task required was the af-
fective categorization of the word (the word being the target and the picture
the distractor), in a congruent, incongruent and neutral relationship. For ex-
ample, in the picture condition, the picture of balloons was associated with
the corresponding emotion happiness (congruent), with the opposite emotion
sadness (incongruent) and the neutral word chair (neutral). In the word con-
dition, the word happiness was associated with the picture of the balloons
(congruent), with the picture of the cemetery (incongruent) and with the pic-
ture of the chair (neutral).
The pictures were 5-9 cm tall and 4-8 cm wide. The words were printed in
Helvetica block capitals, 24 points, and were composed of from 5 to 9 letters,
6 mm tall and from 3.0 to 4.8 mm wide.
Automatic Affective Evaluation / 291

Figure 16-1. Positive (left column) and negative (right column) black and white pic-
tures (of the original colour pictures used in the experiment) and positive and negative
associated words
292 / Aesthetics and Innovation

The experimental design was mixed, 2x3x2: the first factor was condition,
with two levels, picture and word, and between subjects. The second factor
was congruency and referred to the kind of relationship between the stimuli
of the pair, with three levels, congruent, incongruent and neutral, within sub-
jects. The third factor was the target valence, with two levels, positive or neg-
ative, within subjects. Participants were divided into the two conditions and
every participant responded only to one experimental condition, picture or


Participants were seated in front of a colour monitor of a Power Macin-

tosh computer. The first task that participants had to go through was to evalu-
ate the valence of the pictures and the words printed on a card of the same
size as those later appearing on the screen. This task was performed to make
sure that stimuli would be evaluated with the same valence, positive/negative,
that had been attributed by us. Participants first evaluated stimuli correspond-
ing to their task condition, picture or words, and then stimuli of the other con-
dition. Participants that did not answer according to the established criteria
were discarded.
The second task of this preliminary phase was to associate each picture (if
the task was the categorization of the picture) with the corresponding word
for meaning, in order to match 10
pairs of stimuli, picture/word, for a meaningful association. A correspond-
ing pairing, to match each word with a picture, took place when participants
responded in the word condition.
Afterwards, participants read the instructions on the task to be performed.
They were told that pairs of stimuli, always composed of a picture and word,
would be shown in the central position of the computer monitor. Picture and
word would appear randomly at the top or bottom. Before the stimuli ap-
peared, a fixation point would indicate the centre of the screen; immediately
afterwards, the real stimuli would appear. It was stressed that attention had to
be focused only on the picture, ignoring the word, in the picture task, and the
opposite in the word task. On every trial the complete presentation sequence
was composed of: a fixation point (500 msec), blank screen (500 msec), pair
of stimuli (up to 2000), blank screen (500 msec).
The affective categorization task consisted of an evaluation, as fast as
possible, of whether the picture or word expressed, represented or communic-
ated a positive or a negative sensation. If the answer was positive, the parti-
Automatic Affective Evaluation / 293
cipant had to press a green button on the keyboard; if the answer was negat-
ive, then a red button had to be pressed. The positive and negative buttons
were inverted for half of the participants. A warming up phase with two
blocks of ten stimuli each (different from those of the real experiment) fol-
lowed to get familiar with the task and the equipment. Once everything was
clear, participants went through the real experiment. SuperLab 1.6.4 software
package was used to record reaction times.


The hypothesis was that even with stimuli of this kind, in which the rela-
tionship between picture and word was given by an affective association, the
incongruence between target and distractor would produce a delay in the af-
fective categorization task, with an asymmetry of the answer regarding pic-
tures and words, with picture evaluation being faster than word evaluation. In
fact, according to the Glaser and Glaser’s model (1989) the irrelevant stimu-
lus (distractor) interferes only if it has a privileged access to the system relev-
ant for the response; in this case, the word distractor has no privileged access
to the semantic system and it should not cause any delay in the incongruence
pictures situation.
Regarding valence, a well-known asymmetry in the literature refers to the
different evaluation of positive and negative stimuli. Robust and stable data
(Kanouse & Hanson, 1971; Peeters & Czapinski, 1990; Pratto & John, 1991;
Taylor, 1991) refer to a “negativity effect” in which negative stimuli, having
a greater impact than positive ones, are evaluated slower than positive ones.
We expected to obtain results in line with these findings.

The general criteria for accepting answers took into account the prelimin-
ary appraisal of the valence stimuli, latency of the reaction time, and errors.
In the picture condition there were 36 participants in the experiment: 3
participants were discarded because the picture valence evaluation was differ-
ent from the established criteria and so they did not take part in the experi-
mental phase. Thus, 33 participants were taken into consideration for this
condition. There were 34 participants in the word condition: 2 were discarded
due to a different evaluation of the word valence, and so 32 participants were
With regards the second task of the preliminary phase, picture/word pair-
ing (see Table 1), it was clear from the diagonal of the matrix that most of the
294 / Aesthetics and Innovation
associations between stimuli mirrored the affective relationship we expected
(heart/love, bomb/fear, etc.).

Table 16-1. Numbers of pairing between pictures and words: Words are in rows and
pictures are in columns. The diagonal shows that the major numbers of association
(out of 64 participants) are made respecting the criterion hypothesized.


happiness 53
sadness 61
success 58
illness 63
love 63
fear 52
victory 58
danger 56
interest 59
disgust 63



Data outside the diagonal are limited to a very few associations of other
kinds and have a very low value. Therefore, information on the composition
of the pairing is evident and does not need further statistical analysis;
moreover, the asymmetrical structure matrix of the data ties in with the use of
an asymmetrical scaling technique like Multidimensional Unfolding, often
giving rise to an anomalous solution with many missing values as in this
A general criterion adopted for answer acceptance was to exclude from
the analysis all reaction times under 250 msec (a time interval considered too
short for a valid answer, attributable then to anticipated answer error) and
above 1500 msec, to reduce the influence of outliers. There were no answers
Automatic Affective Evaluation / 295
under 250 msec and answers above 1500 were calculated as errors. Errors
(real errors and errors coming from too long reaction times) for picture condi-
tion were 3,9% and for word condition were 3,2 % of trials and were ex-
cluded from analysis. Only in the word condition, one participant exceeded
the level of 10 % (level of errors accepted for each participant) and was ex-
cluded from analysis. Thus, the analysis was carried out on 33 participants in
picture condition and 31 in word condition, for a total number of 64 parti-
The means of the reaction times (in milliseconds) in the picture and word
condition, as a function of the different levels of congruence and valence, are
reported in Figure 2.

Figure 16-2. Means of reaction times (in ms) as a function of condition (picture and
word) and congruence (congruent, incongruent, neutral)

An analysis of variance 2 (condition: picture and word) x 3 (congruence:

congruent, incongruent and neutral) x 2 (valence: positive and negative) was
performed; the condition factor was between subjects, while the other two,
congruence and valence, were within subjects.
The condition factor had a significant effect, F(1,62) = 25.90, p < .001; in
fact, the total time required for the affective categorization of pictures (675
msec) was much faster than the one needed for words (817 msec), with a
total difference of 142 msec. The main effect for congruence was significant,
296 / Aesthetics and Innovation
F(2,124) = 19,01, p < .001; congruent stimuli (721msec) were evaluated
faster than both neutral (739) and incongruent ones (780). As regards
valence, a significant main effect was found, F(1,62) = 11,49, p < 001; posit-
ive stimuli were evaluated faster (734 msec) than negative ones (762 msec).
The interaction between condition and congruence was significant, F(2,
124) = 11,11, p < .001; the mean of the reaction times for the congruence
factor referred to pictures was 675 msec, and 817 msec for words, with dif-
ferent values for the three levels of congruence. The interaction between con-
dition and valence was not significant, while the interaction between congru-
ence and valence was significant, F(2, 124) = 7,45, p < .001, with positive
stimuli (734) faster than negative ones (762). The three-way interaction
between condition, congruence and valence was not significant.
More specifically, a 3 (congruence) x 2 (valence) repeated measure AN-
OVA was performed separately for the two conditions, picture and words. As
regards the picture condition, the congruence main effect was not significant;
in effect, only a slight difference of 15 msec in favor of congruent pictures
compared to incongruent was observed. The main effect for valence, F(1,32)
= 9,70, p < .004, was significant; positive pictures (661 msec) were evaluated
faster than negative ones (692 msec). The interaction between congruence
and valence was not significant.
As regards the word condition, a 3 (congruence) x 2 (valence) ANOVA
showed a main effect for the congruence factor, F(2,60) = 20,94, p < .001,
with a difference of 103 msec between congruent (772 msec) and incongruent
words (875 msec), with neutral words at 809 msec. For the valence factor,
F(1,30) = 3,41, p < .07, there was a tendency towards significance. The inter-
action between the two factors, congruence and valence, was significant,
F(2,60) = 6,37, p < .003.
Words were chosen referring to the affective association with pictures, so
for the frequency in written Italian was not balanced between positive and
negative words (Institute of Computational Linguistic of Pisa, 1989). To
check possible influence of word frequencies on reaction times a correlation
between latency times and word frequencies was performed (r = -.21, ns);
results thus avoid a possible effect of word frequency on valence effect.

Results clearly indicate that even with stimuli of a certain complexity in
which the picture/word association is of an affective type, as those used in
this study, there is evidence of automatic affective evaluation. Target affective
categorization latency times are facilitated by a congruent distractor (same
Automatic Affective Evaluation / 297
valence), in comparison to an incongruence situation. In the picture condi-
tion, the difference in reaction times between congruent and incongruent is
small (15 msec); on the other hand, in the word condition, the difference
between congruent and incongruent is very high (103 msec) and statistically
significant. The difference in the evaluation of picture and word is very
marked. Pictures are categorized much faster than words: pictures have an
overall mean of 675 msec, while words of 817 msec, with a difference of 143
msec. This result is further confirmation of the very stable data in several
works on cognitive categorization of pictures and words (Glaser &
D?ngeloff, 1984; Lupker, 1979; Smith & Magee, 1980).
The categorization task needs a conceptual processing of the stimulus,
picture or word. Direct access of a picture to this kind of information in the
semantic system brings a substantial lowering of reaction times as compared
with a word. These results are also confirmed in the affective categorization
task and further evidence has been obtained in this experiment. When, as in
this case, the picture has to be affectively categorized, referring to the valence
represented, a certain number of nodes corresponding to different steps of
processing must be activated: structural description, semantic or conceptual
representation, inclusion in the affective category, and execution of the re-
sponse. According to some affect theories (Bower 1981), affective informa-
tion is represented in a network similar to the semantic system; the greater
speed of the picture affective evaluation compared to the word would confirm
the fact that pictures have a privileged access to the semantic system.
De Houwer and Hermans (1994) affirmed that if the affective information
is represented within the semantic system, pictures should have more rapid
access to this type of information
than words, as the Glaser and Glaser (1989) model postulates. These res-
ults would seem to confirm the idea that affective information is represented
and stored in the semantic system because affective pictures have quicker ac-
cess to this type of information with respect to words.
Regarding the valence of stimuli, responses to positive stimuli, regardless
of whether they refer to pictures or to words, are faster (734 msec) than those
to negative ones (762 msec) with a significant effect of valence in the AN-
OVA. Implication of valence asymmetry will be discussed later.

Experiment 2
In experiment 1 participants had to judge the target as positive or negat-
ive, so for they had to perform an explicit evaluative task. Results showed
evidence of affective automatic evaluation of pictures and words. But, as Bar-
298 / Aesthetics and Innovation
gh (1989) states, automatic evaluation should occur also when an evaluation
aim is not present. Giner-Sorolla et al. (1999, experiment 2), answering to the
question if automatic evaluation is a goal dependent phenomenon, found that
in a pronunciation task of the target word (so for with a non-evaluative task)
picture primes interfere with target words. In this second experiment we want
to observe if in the absence of an explicit evaluative task, by having parti-
cipants merely answer a lexical decision task, it is possible to automatically
process target stimuli combined with different associations to prime pictures.



Twenty Italian participants, aged between 19 and 29, volounteered parti-

cipated in the experiment.


From the ten pictures used in the previous experiment, two pictures were
changed: the red cross, because of a valence meaning ambiguity (in fact, in
the evaluation phase of Experiment 1, 9 participants gave it a positive evalu-
ation “is good in case of need”) was replaced by a line drawing of a syringe ;
the picture of the bomb was replaced by a gun to balance chromatic features
of positive stimuli. Every picture was associated to three words: the name of
the picture (for example the picture of the syringe and the word syringe); an
affective meaning of the pictures (the picture of the syringe and the word ill-
ness); a neutral word. Overall stimuli were composed of 10 prime pictures
and 30 word targets. Another 10 pictures linked to 30 non-word targets as
fillers were added.


Participants were told that a series of stimuli would be shown in the cent-
ral position of a colour monitor of a Power Macintosh computer. On every
trial the complete presentation sequence was composed of: a fixation point
(500 msec), blank screen (500 msec), prime picture (200 ms), blank screen
(100 msec) and the target word (up to 2000 msec), with a SOA of 300 ms.
The task consisted of a lexical decision task of the target words, by press-
ing one of two buttons on the keyboard, as fast as possible. The word/non-
word button were inverted for half of the participants. A warming up phase
with two blocks of ten stimuli each (different from those of the real experi-
Automatic Affective Evaluation / 299
ment) followed to get familiar with the task and the equipment. SuperLab
1.6.4 software package was used to record reaction times.
Differently from the Stroop task, where interference is obtained if there is
an asymmetry in speed or in automaticity of processing between target and
distractor (Klauer 1998), in the priming paradigm, according to spreading ac-
tivation hypothesis, interference should take place if prime and target are se-
mantically or affectively related. Priming would better fit the study of auto-
matic evaluation in this second experiment in the absence of an explicit eval-
uative task.


The hypothesis is that the names of the pictures should be processed

faster than emotions and neutral words because names should be the
strongest association to pictures for spreading activation and concreteness.
Emotions should be faster than neutral words because pictures would spread
the activation to words affectively linked as in the affective priming.

As a whole, evaluation of names is faster than emotions and neutral
words. An ANOVA for the factor condition at 3 levels (names, emotions, and
neutral) shows a significant effect,
F(2,38) = 5.32, p < .009. Post hoc t-test, Names vs. Emotions is signific-
ant, t(19) = -4.33, p < .001; as well as Names vs. Neutral, t(19) = -4,33, p<
.001; the difference between Emotions (as a whole without subdividing them
for valence) and Neutral is not significant.
More in detail it can be observed that there is a strong difference between
the evaluation of positive and negative emotions; evaluation of positive emo-
tions is significantly faster than negative emotions, t(19) = -4.41; p < .0001,
and than neutral words, t(19) = -2.48, p < .02 (see Fig. 3).
To check for a possible influence of word frequency on reaction times a
correlation between latencies times and word frequency was performed
(r = -.22, ns); the results avoid a possible effect of word frequency on valence
300 / Aesthetics and Innovation

Figure 16-3. Means of reaction times (in ms) as a function of names, emotions and
neutral words


Results of this experiment show that latency times of names are faster
than emotions and neutral words; the difference between emotions as a whole
and neutral is not significant; so the hypothesis seems only partially con-
firmed. But more in detail it can be observed that there is a clear difference
between latency times of emotion stimuli attributable to stimuli valence: pos-
itive emotions are processed faster (539ms) than negative ones (611ms);
there is in fact a significant difference between positive emotions versus neut-
ral but not between negative emotions versus neutral. Experiment 2 replicates
the findings about the valence asymmetry found in Experiment 1 with posit-
ive stimuli, pictures, and words always faster than negative ones.

General Discussion
Results from the two experiments demonstrate that automatic processing
can take place when pictures are associated to a word with an affective mean-
ing and not only when pictures are associated to the name of the object. The
two studies presented show that a different class of stimuli from those usually
investigated, coming from our everyday experience, are able to be automatic-
ally processed. In the first experiment evidence of automatic affective pro-
Automatic Affective Evaluation / 301
cessing for pictures and words in an explicitly evaluative Stroop task has
been found. Differences in picture and word processing have emerged as well
as an asymmetry of valence categorization. In the second experiment, using
the priming paradigm, in the absence of an evaluative task, evidence of auto-
matic processing has been found only for a specific class of stimuli: the
names of the objects and the positive emotion words linked to the pictures
compared to neutral words.
One constant in the two experiments is the positive-negative asymmetry
in favor of a major speed of positive stimuli either pictures or words.
Kanouse and Hanson (1971) referred to a negativity effect due to the greater
impact of negative stimuli compared to positive stimuli of the same intensity.
According to Peeters and Czapinski (1990), the stronger impact of negative
stimuli is due to a behavioural adaptive function. The greater weight of negat-
ive stimuli will not only activate an avoidance reaction, but will also produce
a defense and control attempt of the negative object. Pratto and John (1991)
found that participants took longer to name the ink colour of negative words
than positive ones that represented traits of personality. In studies on impres-
sion formation and person perception, the stronger weighing of negative than
positive information has been explained in informational terms by the higher
informativeness (or diagnosticity) of negative information. Negative informa-
tion tends to be perceived as more diagnostic than positive information be-
cause people’s expectations of events and outcomes in the world are gener-
ally positive. For example, people expect others to behave in socially desir-
able or at least socially appropriate ways (Kanouse & Hanson 1972). Matlin
and Stang (1978) suggests that desirable events tend to be viewed as com-
mon, frequently occurring and typical whereas bad events tend to be uncom-
mon, infrequent, and atypical. According to Taylor (1991), negative events
appear to elicit more physiological, affective, cognitive, and behavioural
activity and prompt more cognitive analysis than neutral or positive events.

Despite the complexity of our iconic world, which most of the time, needs
to be interpreted consciously (symbolically and metaphorically), we have
tried to demonstrate that we possess a particular skill by analyzing it pre-at-
tentively and automatically.
Given these results, it would be interesting to address further research on
more complex objects with an aesthetic value, to see whether even these
could be automatically evaluated as a first basic approach to aesthetic percep-
302 / Aesthetics and Innovation
Bargh, J. A., Litt, J. E., Pratto F., & Spielman, L. A. (1989). On the pre-
conscious evaluation of social stimuli. In K. McConkey & A. Bennet
(Eds.), Proceedings of the XXIV International Congress of Psychology
(Vol. 3, pp. 357-370). Amsterdam: Elsevier/North Holland.
Bargh, J. A. (1989). Conditional automaticity: Varieties of automatic influ-
ence in social perception and cognition. In J. S. Uleman & J. A. Bargh
(Eds.), Unintended thought (pp. 3-51). New York: Guilford.
Bower, G. H. (1981). Mood and memory. American Psychologist, 36, 129-
Bruner, J. S., & Postman, L. (1947). Emotional selectivity in perception and
reaction. Journal of Personality, 16, 69-77.
De Houwer, J., & Hermans, D. (1994). Differences of affective processing of
words and pictures. Cognition and Emotion, 8, 1-20.
Fazio, R. H., Sanbonmatsu, D. M., Powel, M. C., & Kardes, F. R. (1986). On
the automatic activation of attitudes. Journal of Personality and Social
Psychology, 50, 229-238.
Giner-Sorolla R., Garcia, M. T., & Bargh J. A. (1999). The automatic evalu-
ation of pictures. Social Cognition, 17, 76-96.
Glaser, W. R., & Düngeloff, F. J. (1984). The time course of picture-word in-
terference. Journal of Experimental Psychology: Human Perception and
Performance, 8, 875-894.
Glaser, W. R., & Glaser, M. O. (1989). Context effects in Stroop-like word
and picture processing. Journal of Experimental Psychology: General,
118, 13-42.
Hansen C.H., & Hansen, R. D. (1988). Finding the face in the crowd: An an-
ger superiority effect. Journal of Personality and Social Psychology, 54,
Hermans, D., De Houwer, J., & Eelen, P. (1994). The affective priming af-
fect: Automatic activation of evaluative information in memory. Cogni-
tion and Emotion, 8, 515-533.
Institute of Computational Linguistic of the CNR (1989). Corpus di Barcel-
lona. Pisa: ILC.
Kanouse, D. & Hanson, L. (1971). Negativity in evaluations. In E. E. Jones,
D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.),
Attribution: Perceiving the causes of behaviour (pp. 47-62). Morristown,
NJ: General Learning Press.
Klauer, K. C. (1998). Affective priming. European Review of Social Psycho-
logy, 8, 67-103.
Automatic Affective Evaluation / 303
Klauer, K. C., & Musch, J. (2003). The psychology of evaluation. In J Musch
& K. C. Klauer (Eds), The psychology of evaluation (pp. 7-49). Mahwah
N J: Lawrence Erlbaum Associates.
Lang, P. J., Öhman, A., & Vaitl, D. (1988). The international affective picture
system. Gainesville FL: The Center for Research in Psychophisiology,
University of Florida.
Lupker, S. J. (1979). The semantic nature of response competition in the pic-
ture-word interference task. Memory and Cognition, 7, 485-495.
Matlin, M. W., & Strang, D. J. (1978). The Polyanna principle. Selectivity in
language, memory and thought. Cambridge, MA: Schenkman.
McGinnies, E. M. (1949). Emotionality and perceptual defence. Pychologic-
al Review, 56, 471-482.
Meyer, D. E., & Schvaneveldt, R. W. (1971). Facilitation in recognizing pairs
of words: Evidence of a dependence between retrieval operations. Journ-
al of Experimental Psychology, 90, 227-234.
Murphy, S. T., & Zajonc, R. B. (1993). Affect, cognition and awareness:
Priming with optimal and suboptimal stimulus exposures. Journal of Per-
sonality and Social Psychology, 64,723-739.
Niedenthal, P. M. (1992). Implicit perception of affective information. Journ-
al of Experimental Social Psychology, 26, 505-527.
Öhman, A. (1987). The psychophysiology of emotion: An evolutionary-cog-
nitive perspective. Advances in psychophysiology, 2, 79-127.
Öhman, A. (2002). Automaticity and the amygdala: nonconscious response to
emotional faces. Current Directions in Psychological Science, 2, 62-66.
Peeters, G., & Czapinski, J. (1990). Positive-Negative asymmetry in evalu-
ations: the distinction between affective and informational negativity ef-
fect. European Review of Social Psychology, 1, 33-60.
Pratto, F., & John, O. P. (1991). Automatic vigilance: The attention-grabbing
power of negative social information. Journal of Social and Personality
Psychology, 61, 380-391.
Smith, M. C., & Magee, L. E. (1980). Tracing the time course of picture-
word processing. Journal of Experimental Psychology: General, 109,
Snodgrass, J. G., & Vanderwart, M. (1980). A standardized set of 260 pic-
tures: Norms for name agreement, image agreement, familiarity and visu-
al complexity. Journal of Experimental Psychology: Human Learning
and Memory, 6, 174-215.
Stenberg, G., Wiking, S., & Dahl, M. (1998). Judging words at face value: in-
terference in a word processing task reveals automatic processing of af-
fective facial expressions. Cognition and Emotion, 12, 755-782.
304 / Aesthetics and Innovation
SuperLab 1.6.4 (1994). Silver Spring, MD: Cedrus Corporation.
Taylor, S. E. (1991). Asymmetrical effect of positive and negative events:
The mobilization-minimization hypothesis., Psychological Bullettin, 110,
White M. (1995). Preattentive analysis of facial expression of emotion. Cog-
nition and Emotion, 9, 439-460.
Zajonc, R. B. (1980). Feeling and thinking. Preferences need no inferences.
American Psychologist, 35, 151-175.

Computer Sound Analysis in Musicology:

Its Goals, Methods, and Results

Alexander V. Kharuto

The acoustical approach to musical sound studies has a long history. Now,
computers have become one of the most powerful and universal instruments
for the examination of musical sound. In their professional work, musicolo-
gists have to analyze musical performances. For different types of music,
they would analyze the interpretation, or deviations from a “standard” per-
formance, or make notation of a folklore music piece, etc. In order to satisfy
musicologists’ needs, the author has since 1995 developed a special computer
program, which estimates sound pitch with an error not more than 4-5 cents
(i.e., 1/20 of a halftone) and obtains parameters of tones, glissandos, and vi-
brato elements. With the help of the program, numerous investigations have
been done. Some examples described below are analysis of folklore singing,
sound pitch rows and changes in traditional performance style, statistical ana-
lysis of vibrato in academic singing, and analysis of Tuva throat singing.

The acoustical approach to musical sound studies has a long history.

“Live” sound was first analyzed only by ear, and later, by means of Helm-
holtz’s acoustical resonators (Helmholtz, 1877). Than, in the electrical age,
sound could be transformed into the form of electrical oscillations and ana-
lyzed with the help of electrical resonators and filters. Finally, the invention
of sound recording devices allowed analyzing sound recordings instead of
“live” sound. Now, computers have become one of the most powerful and
universal instruments for musical sound examination.
New technical devices introduced at each stage of this history have
provided increasing information from the sound. The possibility of listening
to the same sound recording many times enabled musicologists to distinguish

306 / Aesthetics and Innovation
more details of performance. Using technical devices such as electrical filters
or special spectrum analyzers, and now the modern computer equipment for
oscillation analysis provides objective and more and more precise measure-
ment of a musical sound’s physical properties.

Properties of Musical Sound

Professional sound perception by musicologists or music teachers deals
with specific analyses of artistic performance. In the case of academic music,
the score of which is well-known, a musicologist would be interested in per-
formance peculiarities, in details of interpretation of a musical piece by a cer-
tain performer. In the case of teaching, similar properties would be analyzed
as deviations from a “standard” performance of the musical piece. By analyz-
ing of vocal sound some other parameters would be added, such as voice son-
ority, vocal vowels, power uniformity, saturation of spectrum formants with
overtones, etc. Folklore performance analysis has the goal of creating the
score, which did not exist before because of the oral nature of this kind of art.
This notation means determination of the pitch, intensity, and duration of
each sound and also fixation of performance style properties.
The computer-aided analysis must provide measurements of the sound
parameters listed above. In order to satisfy the musicologists’ needs, the au-
thor developed a special computer program named SPAX (reg. #2005612875
of Federal Institute of Industrial Property of Russia).

Computer Methods of Musical Sound Analysis

As is well known, the sound spectrum contains information about timbre
as well as about all other sound parameters. Therefore, the sound parameters
for each point of time can be derived from the “current” spectrum. First, the
sound would be recorded in digital form onto a computer's hard disk. Then,
the program calculates a sequence of spectrums with a time stepping of about
5-10 milliseconds. The sound spectrum has a convenient graphic representa-
tion as a sonogram like the one shown in Figure 1. The darker the sonogram
point, the more power the spectrum component has at this frequency and this
point in time. Points compose horizontally oriented lines, which conform to
the current main tone and overtones.
In Figure 1, the vertical line approximately at the time point 1.1 s is the
time marker used in the program for measurements. In the right field, the mo-
mentary spectrum for the time point 1.1 s is shown (vertical axis—frequency,
Computer sound analysis in musicology / 307
horizontal from left to right—power of components). The overtones are rep-
resented on the momentary spectrum as “peaks”. Not every sound has an
evident pitch. We define as “musical” those sounds which have “sound
pitch”. This type of oscillations obeys the law of temporal periodicity. The
period may have a complex form, and therefore the spectrum will contain
harmonic components in the frequencies f1, 2xf1, 3xf1, etc. (only), where f1
is the main tone of the musical sound (named also “fundamental”). This tone
will be perceived as the sound “pitch”. From the viewpoint of music, the har-
monics are “overtones” of this tone. Hereby, the periodical character of oscil-
lations in the time domain and that of spectrum structure are the persistent
properties of “musical sound”.

Figure 17-1. Results of computer analysis of a Russian folklore song: The spectrum in
form of a sonogram

Let’s note that the “main” role of the main tone f1 has only an informa-
tional sense, but not an energetic one. Indeed, in the presence of 2xf1, 3xf1,
etc. spectrum components, f1 component falling out does not change the peri-
od of sound oscillations, and, therefore, the perceived sound pitch. (Some
sound distortions like these are typical for telephone communication: the low
main tone frequency of a man’s voice will not be transmitted, but we discern
a man’s and women’s voices.)
The computer algorithm used in author’s program detects the sound pitch
as the mean distance between overtones and provides the estimation of sound
308 / Aesthetics and Innovation
pitch with an error no more than 4-5 cents (i.e., 1/20 of a halftone, a well-
known practical limit of human perception ability). Comparison of many
computation results with expert’s evaluation (based on hearing) showed a
close fit of it. During musical performance the passing pitch varies continu-
ously. The computer program plots the graph of this time function (so-called
«melogram») shown in Figure 2.

Figure 17-2. The melogram: Results of computer analysis of the spectrum in fig. 1

There are some instruments which form non-equidistant overtone sys-

tems, for example, bells and gongs. In these cases one can recognize rather
sound register then a certain pitch.

Types of Sound Variation and their Measurement

Numerous melogram analyses allow finding out the following persistent
structures: horizontal segments (sounds with approximately constant pitch),
sloping segments (glissandos), oscillating segments (vibration of sound
pitch), and various combinations of them. These “elementary” sounds form
an “alphabet” of variation types. From the acoustical point of view, these ele-
ments have also numerical characteristics, such as duration, mean pitch,
range of variations (for glissando and vibrato), and frequency (for vibrato).
The manner of using of these elements and the variety of parameters charac-
Computer sound analysis in musicology / 309
terize the peculiarities of musical performance (Smirnov & Kharuto, 2000;
Kharuto & Smirnov, 2002). The author’s program SPAX provides visual re-
cognition of the type of sound pitch variation and makes it possible to meas-
ure numerical parameters for elements of every type.
The horizontal elements of a melogram will be depicted by usual notes on
the score. Possible variations of real sound pitch may be negligible for music-
al instruments with fixed tuning. But, because of accidental or intentional
pitch variations, sound pitch constancy would be denoted only approximately
for other instruments and singing voice. In this case, the program SPAX
measures mean pitch value and both maximum and mean square deviation of
pitch from the mean line. Figure 3 shows an example of such measurement.
As a first approximation, glissando elements may be described, by the co-
ordinates of the two ends of a glissando line. However, in practice glissando
never forms strongly straight lines. The program SPAX calculates parameters
of mean straight lines of glissando and also maximum and mean square devi-
ations of real glissando from the calculated mean line. An example of these
measurements is shown in Figure 4.
The next “complicative” element is musical vibrato, a periodical change
of the sound pitch during a tone or glissando performance. The computer pro-
gram SPAX provides a measurement of vibrato parameters. As a result, the
parameters of vibrato mean line (tone-like or glissando-like) will be fixed and
the maximum and mean square deviation from this line and the frequency
and maximum magnitude of vibrato itself are also fixed. In order to capture
the complex form of vibrato, the magnitude of the first 7 harmonics of the vi-
brato form are also calculated and fixed. Examples of vibrato parameters
measurement are shown in Figure 5, “a” and “b”. In first case, vibrato is per-
formed “in the tone”, and in second case, glissando forms a “mean line” for

Computer Analysis of Sound Pitch: Statistical Approach

Precise computer analysis shows that the pitch confining is not ideal even
in “tone-like” elements: the melogram contains some variations. These fea-
tures may be caused by psycho-physiological, individual, and regional (tradi-
tional) reasons. Taking into account accidental variation of elementary
310 / Aesthetics and Innovation

Figure 17-3. The measurement of sound mean tone

(Russian folklore song)

Figure 17-4. Measurement of a melogram fragment with glissando

(Russian folklore song)

Computer sound analysis in musicology / 311

Figure 17-5. Measurements of vibrato: (a) with a horizontal mean line and
(b) glissando mean line

sounds and numerical characteristics, it is suitable to use statistical ap-

312 / Aesthetics and Innovation
The analysis program measures numerical and statistical characteristics of
every element of a melogram. These are mean pitches of “tones” and mean
line of glissando and deviations of sound pitch curves from the mean line as
well as range of variations for glissandos and vibratos.
Statistical distributions of the sound pitch and of its changing velocity are
also accessible with the same program, SPAX. These statistical parameters, in
a aggregate, may characterize the style of performance of the musical piece.
It is shown below that the statistical distribution of the sound pitch helps
to discover the pitch row used by a folklore performer. Pitch changing velo-
city statistics may also be useful for distinguishing folklore singing styles.
Statistical analysis of the vibrato in academic singing shows that the vibrato
parameters are not stabile in groups of performers or in one given singer.

Computer Sound Analysis: Folklore Singing

In many folklore music melograms the sound pitch of tone-like elements
forms a lattice structure, named “the pitch row”. As a rule (Kharuto &
Smirnov, 2000), the pitch row of folk music performers does not match the
12-tone equal tempered row of European academic music. The sound pitch
rows analysis of many performers from Russia (including Siberia and the
North Russian regions) shows that they are principally equal-tempered.
However, the interval between pitch levels is less then 1/12 and varies down
to 1/30 of an octave. An example of a 17-microtone pitch row used by a Rus-
sian folk singer is shown in Figure 6.
In order to estimate the performer’s pitch row, the computer program uses
the one-dimensional statistical distribution of the sound pitchs. An example
of such a distribution, superimposed on the melogram fragment is shown in
Figure 7. In this graph horizontal lines designate sound pitches used by the
performer. Their length (measured from right picture edge) is proportional to
the probability of each given sound pitch.
In case of fixed-tuned instruments (like a piano), the sound pitch distribu-
tionwould have a form of s simple lattice with components corresponding to
standard notes of 12-tone equal tempered pitch rows. Distributions of vocal
sound pitchs are more complex because of intermediate pitch levels. Those
pitch levels are produced by glissandos, vibratos, as well as pitch instability
specific to the human voice.
Computer sound analysis in musicology / 313

Figure 17-6. Example of 17-microtone pitch row used by folk singer (Russia)

Figure 17-7. Statistical distribution of sound pitch (Russian folklore song)

In some folklore types of music no persistent structure of pitch stages can

be discovered at all. Such examples represent the performance of North Rus-
314 / Aesthetics and Innovation
sian lamentations (Kharuto & Smirnov, 2002). The North Russian folk music
tradition is well known because of keeping its integrity for a long time. The
location remote from the center of Russia and lack of communication protec-
ted this tradition from the influence of European civilization until the middle
of the 19th century. Then, the dissemination of literacy caused changes in
folk singers’ repertoires. At the end of 19th century, the diffusion of accordi-
ons with its fixed tuning began to affect the singing manner. Now this tradi-
tion is being destroyed under the influence of different causes. This influence
extends in the form of broadcasting of audio records (radio, phonograph) as
well as through the school education renewed after the end of World War II.
In our work (Kharuto & Smirnov, 2002), three groups of records represent
the examined tradition in expedition materials: 1) sound recordings of the
elder generation of performers (1910-1920s years of birth), 2) sound record-
ings of the “middle” singers generation (most active as performers now), 3)
sound recordings of children groups, which adopted the tradition lately.
An example of a melogram, typical for the eldest performer generation is
shown in Figure 8.
While analyzing precise melograms of the eldest performers generation
one can see a certain persistent form with continuous type of pitch varying.
This form of curve repeats for each couplet and has a number of “hills” the
tops of which coincide with the moments of accents of the vocal text. The as-
cending glissandos are steep, and the descending changes are slower.
The same analysis has been made for the sound recordings of the
“middle”and “youngest” generations of singers. As a whole, the results are
like those in the previous case, but contain some simplifications. An example
is shown in Figure 9. The sound pitch field reveals pronounced pitch row
stages, and former steep “hills” are replaced by a sequence of short glissan-
dos and tone-like fragments.
The ethnomusicologist’s conclusions given above can be formulated in
terms of mathematical statistics. New possibilities of analysis have been ad-
ded into the program SPAX especially for this task,—the analysis of statistic-
al distribution melody curve velocity properties. The substitution of steep
“hills” with great raise and fall of sound pitch (by elder performers) through a
sequence of smaller “cupola” and following sound with constant tone (by
younger performers) can be described through changing of the pitch moving
Computer sound analysis in musicology / 315

Figure 17-8. Melogram and statistical distribution of sound pitch for a performer of
“eldest” generation (North Russian lamentation)

Figure 17-9. Melogram and statistical distribution of sound pitch for a performer of
“middle” generation (North Russian lamentation)
316 / Aesthetics and Innovation
steepness. Two examples of statistical distribution of this parameter for the
“eldest” and “middle” generation of performers are represented in Figure 10,
(a) and (b).
As a rule, distributions for elder generation of performer are shifted from
zero to “plus”. This means that steep ascending is more typical than the slow
one, but descending is slow as usual. In addition, these distributions are more
dispersed. This property can be explained through the variety of steepness
during every glissando fragment (i.e., non-linearity of glissando curves). The
examples of the younger performers’ generation tend to more symmetrical as-
cending and descending movements and lower spreading of steepness. In oth-
er words, the influence of European academic music caused some loss of
sound pitch control complexity, and this loss may be fixed and measured ob-

Computer Analysis of Parameters of Academic Singing

Musical vibrato is one of the usual strokes in academic singing. Profes-
sional singers decorate every sound with it. This skill is gained during music-
al education. The goal of the computer analysis was the experimental study of
musical vibrato performed by professional singers in the academic classic
With the help of the computer program SPAX, 11 sound recordings of ro-
mances and arias performed by female singer G. B. (soprano, graduated from
the Gnessins musical institute in Moscow, Russia) have been analyzed. Also
fragments from Monteverdi’s opera “Orpheo” (conductor J. E. Gardiner), the
vocals of Orpheo (R. Johnson), the 1st shepherd (M. Tueker), and Prosepine
(D. Montaque) have been analyzed. For comparison purposes, some frag-
ments of singer M. Callas opera sound recordings have been used. Only frag-
ments with “a capella” singing have been selected for the study. The frag-
ments where the voice is much louder than other instruments, can also be
used for the sound pitch analysis. Because a concert sound recording (piano
and vocal) contains many fragments of this kind, 191 fragments of a concert
of the singer G.B. have been selected. The opera sound recording contains
less such fragments. Only 37 pieces from Monteverdi’s opera “Orpheo” and
21 fragments from the M. Callas’ sound recording have been used.
Vibrato frequency and amplitude have been measured for every fragment,
which contain vibrato. These experimental data have been analyzed with the
help of the statistical program SPSS. The vibrato frequency and amplitude
statistical parameters for each performer are shown in Table 1. These data
Computer sound analysis in musicology / 317

Figure 17-10. Statistical distributions of the steepness of sound pitch changing for
“eldest” (a) and “middle” (b) generations of performers (North Russian lamentation)

a show significant variety of vibrato parameters, its frequency as well as its

amplitude, for different performers. The comparison of vibrato frequencies
318 / Aesthetics and Innovation
and amplitudes statistical “standard deviation” for different performers shows
a great variety of these parameters, too.

Table 17-1. Statistical parameters of vibrato

Source recording, Frequency Amplitude

M(f), Hz SD(f), Hz M(A), SD(A),
cent cent

Concert recording of 6.0 0.7 92.9 33.2

“Orpheo”—Opheo 5.9 0.5 68.6 19.9
“Orpheo”—the 1st 5.7 0.6 74.0 27.6
“Orpheo”—Prosepine 7.3 1.2 31.6 10.4
Phonogram of 6.2 0.6 86.3 21.0
M. Callas

The study of statistical distribution of vibrato frequencies and amplitudes

for the singer G.B. (191 fragments) is shown in Figure 11, (a) and (b). The
distributions have hill-like forms, but the Kolmogorov-Smirnov statistical
test does not confirm the Gauss-type of the distribution curve.
The distribution of vibrato parameters of “Orpheo”-performers are of the
same type. The study of M. Callas’s vibrato discovers another distribution
form of frequency and amplitude, as shown in Figure 12, (a) and (b).
The results of this investigation show that the musical vibrato in vocal
academic performance does not belong to “pre-learned” elements. The vi-
brato parameters vary in wide bands and may be influenced by the individual
characteristics of performers and also by emotional tension corresponding
with the analyzed fragment of musical pieces.
Computer sound analysis in musicology / 319

Figure 17-11. Statistical distributions of vibrato frequencies (a) and amplitudes (b)
(concert recordings of G.B.—191 fragments)

Figure 17-12. Statistical distributions of (a) vibrato frequencies and (b) amplitudes
measured in the sound recording of M. Callas (21 fragments)
320 / Aesthetics and Innovation
Analysis of Timbre Variations: Khoomei and the “Two
The cultural traditions of Asia include a specific technique of singing,
named “khoomei”. It is the well-known Tuva traditional throat (overtone)
singing. One can meet the same singing technique also in Mongolia. The tra-
ditional khoomei usually contains two alternately changing kinds of sound-
ing: a “recitative” (“half-sung” text) and a vocalization (without text). This
second kind of singing often sounds like a flute and makes the listener to hear
something like two or more voices together.
Some researchers supposed that it is the effect of independent oscillations
of the “true” vocal chords and the “false” vocal chords in human’s throat,
which produce two different sounds (Axenov, 1964).
Computer analysis shows that there is only one low sound (bourdon),
which forms a great deal of equidistant overtones (see Fig. 13). In case of two
sound sources there will be two independent overtone systems with different
frequency steps.
During the recitation parts the sound pitch varies as in usual singing (Fig.
14a). The analysis of the vocalization shows that the sound has a relatively
stable pitch (Fig 14b).
The “flute-variation” effect, which is perceived by the listener, is pro-
duced by adroit selection (made by the human voice apparatus) of one or
more overtones, while other ones are muted. The computer program allows
controlling the threshold of overtone power. The result of computer sound
pitch analysis made with only the most powerful overtones, is shown in Fig-
ure 14c. The main tone at “gis3” corresponds with the most powerful over-
tone before the time point t = 22.5 s (see Fig. 13b). Than, the “main tone”
rises to “cis4” (after t = 22.5 s).
Let’s note that in this case, the only accessible “sound material” is the nat-
ural pitch row based on constant distances between pitch stages (in Hertz),
instead of the more usual equally tempered pitch row (where the ratio of fre-
quencies is constant).
Computer sound analysis in musicology / 321

Figure 17-13. Sonogram of Tuva throat singing (khoomei): (a) recitative part, (b) frag-
ment from a “flute-like” vocalize part.
322 / Aesthetics and Innovation

Computer sound analysis in musicology / 323

Figure 17-14. Computer analysis of khoomei sound pitch for the sonograms represen-
ted on Figure 13: a) the beginning of the recitative part, b) the low (bourdon) sound of
vocalize (the analysis based on all overtones), c) “flute-like” variations in vocalize
(the analysis based on most powerful overtones).

Computer sound analysis methods provide new opportunities in musico-
logical research and raise some new problems in “traditional” European mu-
sicology. High precise analysis of sound pitch allows fixing many details of
musical performance. In the case of musical folklore investigations, sound
pitch rows have been studied which contain always more then 12 microtones
in an octave. Changes in traditional performance style, caused by the influ-
ence of European musical culture have been captured and measured. Statist-
ical approaches in academic vocal vibrato analysis showed the instability of
vibrato characteristics for a set of professional singers. The spectrum vari-
ation analysis of khoomei sound disproved the “two-voice” hypothesis and
clarified the nature of “many voices” sounding and perception.
324 / Aesthetics and Innovation
Axenov, A. N. (1964). Tuvinskaya narodnaya musyka (Tuva folklore music).
Moscow: Musyka. (in Russian).
Helmholtz, H. L. F. von (1877/1954). On the sensation of tone. New York:
Kharuto, A. V. (2003). Folk music sound: Methods and results of computer
analysis. In L. Dorfman, V. M. Petrov, E. Grigorenko (Guest Eds.), Cur-
rent trends in Russian approaches to art and culture: Bulletin of Psycho-
logy and the Arts, 3, 35–37.
Kharuto, A. V. (2004). About elementary musical sounds used in vocal per-
formance (Theoretical and empirical study). In J. P. Frois, P. A. Marques,
& F. Marques (Eds.), Proceedings of the XVIII Congress of the Associ-
ation of the Empirical Aesthetics. (pp. 152–156). Lisbon: IAEA.
Kharuto, A. V., & Smirnov, D. V. (2002). Information approach in examining
of evolution of North Russian folk musical tradition. In T. Kato (Ed.),
Proceedings of IAEA-2002 (17th Congress of the International Associ-
ation of Empirical Aesthetics (pp. 313–318). Takarazuka: Takarazuka
Smirnov, D. V., & Kharuto, A. V. (2000). Nelineiny zvukoryad v muzikalnom
folklore: obtshaya zakonomernost i individualnost (Non-linear sound
pitch row in music folklore: Common consistent pattern and individual-
ity). In Z. E. Zhuravleva, V. A. Koptsik, & G. Yu. Reznichenko (Eds.),
Languages of science—languages of art (pp. 347–352). Moscow: Mo-
scow State University. (in Russian).



Sociocultural Oscillations and their Analogies

with Physical Waves

Colin Martindale

A number of social systems or the artifacts that they produce show long last-
ing oscillations with a period of around 50 to 60 years. I begin by describing
my theory of and empirical studies of the evolution of art and literature. Lit-
erature and the arts show long lasting oscillations in the degree to which their
content is conceptual vs. primordial or concrete. Theoretically these oscilla-
tions are brought about by an unremitting pressure for innovation. I then re-
view a number of areas such as science and the economy in which oscilla-
tions theoretically caused by innovation may be found. Finally, the similarity
of these sociocultural oscillations to physical oscillations such as light and
sound is examined.

Without innovation a social system quite obviously would not change. If

the system were under pressure to innovate, common sense would lead us to
expect that, if possible, we should merely observe a series of innovations.
However, pressure to innovate seems very often to lead to cyclic oscillations
in the products of the system. I begin by describing my theory of aesthetic
evolution. Artists are under a constant pressure to innovate. They generally
manage to accomplish this, but because of the means of production of innov-
ations, oscillations in the content of artworks are invariably observed. This
seems not to be an isolated case. I review research on the history of science,
economics, and other domains where it can be seen that innovation induces
not only unilineal trends but also long lasting oscillations. I end the chapter
by examining whether these sociocultural oscillations are similar to the phys-
ical oscillations found for example in light and sound. Some analogies are

Evolutionary oscillations / 327
pointed out and possibilities for discovering even more analogies are enumer-

Aesthetic Evolution
I have elsewhere described a theory of aesthetic evolution (Martindale,
1975, 1990) patterned after the theories of Darwin (1859, 1871). In order for
evolution of any type to occur, three factors must be present: variation, long-
lasting selection criteria, and mechanisms to preserve selected variants. The
theory applies to all of the arts, but is easiest to explain if we deal with the
case of poetry. Clearly, there is variation in poetry. All poets do not write the
same thing. Poetry is also preserved. It is not discarded once read but kept for
long periods of time.

Selection Criterion of Novelty

All sorts of forces act upon poets, but across the long term they seem to
randomize out with one exception: the selection criterion of novelty. Once
something has been said, later poets cannot say it again. Thus, whether it be
an obsession or a nuisance, it ends up determining the course of poetic his-
tory. I am hardly the first person to emphasize the selection criterion of nov-
Obviously, poets cannot continue repeating the same metaphors or writing
the same thing. They must say something different than what prior poets said.
Mere difference is not sufficient. They must say something more novel or in-
teresting rather than something less novel or striking. This is the case in all of
the arts. For example, Gőller (1888) ascribed change in architecture to what
he called Formerműdung or form fatigue: people tire of seeing the same thing
endlessly repeated and want to see or hear something new.
The Russian Formalists and Czech Structuralists saw that the very defini-
tion of literature and art builds into these activities an incessant pressure for
change. Art and literature will die unless they change continually. However,
these theorists admitted that they could not predict the direction in which they
would change. This would not be a problem if literature and the arts con-
stantly changed but did not change in any specific direction. However, as de-
scribed below, they do change in specific directions. Furthermore, all of the
literary and artistic traditions that I have studied change in the same ways
(Martindale, 1990). Mukařovskỷ (1940) and Tynjanov and Jakobson (1929)
argued that the direction of change is due to extra-literary forces. In so doing,
328 / Aesthetics and Innovation
they assumed that in some sense art reflects or is influenced by society. As
we shall see, this is not the case for the high arts.

Trends in Unpredictability in the Arts

I and other theorists have argued that poetry and the other arts by defini-
tion must become more deformed, deautomatised, or novel across the course
of time. Is this in fact the case? If not, then we had a nice idea that deserved
to be true but is not, and we have nothing to explain. If so, then we need an
explanation for why and how this has happened.
Cohen (1966) measured several devices for increasing unpredictability or
novelty in samples of French poetry from the 18th century, early 19th century,
and late 19th century and found very clear trends. For example, using an an-
imate adjective to modify an inanimate noun rose from 3.6% though 23.6%
to 46.3%. Enjambment refers to lack of punctuation at the end of a line of
verse. When French poetry is read aloud, it is conventional to pause briefly at
the end of each line. Thus, enjambment brings about a conflict between
sound and sense in that there will be pauses in the middle of grammatical
phrases. For the three periods, Cohen found enjambment in 11%, 19%, and
39% of lines. He found a similar trend in noncategorical rhymes (e.g., rhym-
ing a noun with a verb). In this case, the percentages were 18.6%, 28.6%, and
Martindale (1975, 1990) studied samples of poetry by 21 French poets
born between 1770 and 1909. As in the studies of American and British po-
etry described below, the poets were selected on the basis of eminence. The
general procedure has been to divide the epoch being studied into successive
20-year periods and to study the N poets born during each period to whom
the most pages are given in the relevant Oxford anthology of verse. Once po-
ets are selected, random samples are taken from their collected works. The
poets so selected very clearly belonged to the generally accepted canon of
great French poets. I measured the frequency with which words with opposite
connotations (e.g., good vs. bad, strong vs. weak, active vs. passive) occurred
in the same sentence. Such incongruous juxtapositions showed a very strong
monotonic uptrend across time that was quite unlikely to have occurred
merely due to chance. This measure could indicate incongruity where there
was, in fact, none, as in a sentence such as “the strong father saved his weak
daughter.” Reading the poetry shows that the measure was not in general get-
ting at sentences like this but was measuring what I wanted it to measure.
Martindale (1990) used a more subtle but less error prone measure of un-
predictability in a study of poetry by 51 American poets born between 1750
Evolutionary oscillations / 329
and 1949. I created a composite measure composed of indices such as the
hapax legomena percentage (percent of words occurring only once in a text
(hypothetically a measure of unpredictability), variability of word length (an-
other measure of unpredictability), variability of phrase length (yet another
measure of unpredictability), average number of word associates (a measure
of ambiguity or at least of words with multiple meanings), polarity (a meas-
ure of extremity of a word along the dimensions of good vs. bad, active vs.
passive, and strong vs. weak), etc. This measure increased strongly across
I applied the same measure to samples from the poetry of 170 British po-
ets born between 1290 and 1949. The measure increased at an accelerating
rate across the entire time span with no evidence of declines in any of the
successive 20-year periods into which I divided the series. Interestingly, there
were no changes in rate of acceleration. Thus, rate of change in the unpredict-
ability of British poetry has been accelerating, but it has been doing so at the
same rate since before Chaucer. It is also notable that the measure increased
across the periods when the British neoclassical style held sway. The neoclas-
sical poets stressed order and brought it to some aspects of poetry but more
than made up for this by increasing unpredictability in other aspects of their
poetry. It is always well to keep in mind Plato’s warning that poets are prone
to telling lies.
The novelty rule applies to popular as well as to high art. Martindale
(1990) found increases in novelty in the top 10 American popular music lyr-
ics written from 1950 through 1972. The increase is unlikely to have arisen
by chance but is less strong than those found in serious poetry. No one makes
much money writing poetry, so poets tend to write for one another. One can,
however, make a lot of money writing popular music. Those who write it are
probably influenced less by the pressure for novelty and more by the desire to
write whatever the external audience wants to hear.
Martindale (1990) reported quantitative studies of traditions ranging from
European music and painting to Gothic architecture, ancient Egyptian paint-
ing (often falsely thought not to have evolved), Athenian vase paintings, and
Japanese prints. In these studies, rating scales rather than content analysis
were used. The results were always the same. Measures of novelty or surpriz-
ingness increased monotonically across time.

Means of Production of Novelty

We have seen that artists do in fact increase the novelty of their works
across time. Now we shall consider how they have managed to do this. Let us
330 / Aesthetics and Innovation
consider the situation of the poet. Poetry is written within the confines of a
contemporary style. As a simplification, I shall define a style as consisting of
a roughly defined lexicon of words that can be used in poetry and a loosely
defined set of rules as to how these words may be combined. Each poet must
attempt to write poetry more novel than poetry written previously. This is be-
cause of habituation: the audience (mainly other poets) tires of reading the
same or very similar things repeatedly. Novelty may be important for the poet
or a mere nuisance. Its importance lies in the fact that it is a constant pres-
sure. No matter what else a poet wants to say, he must say it in a way that has
not been said before. This is crucial. No matter whether it has been a strong
or a weak force, it has been present ever since poetry was first written. At dif-
ferent times, various forces have acted upon poetry, but they have been trans-
itory. Thus, they had at most a temporary impact on poetry. The consequence
is that the pressure for novelty, because it was constant and always pushing in
the same direction ended up determining the main direction of poetic evolu-
tion. Whether evolution be biological or sociocultural, only constant forces
determine its course. In explaining the direction of literary history, two other
assumptions are helpful. Habituation is a gradual process: we tire of things
slowly rather than immediately. Thus, a poet does not have to write extremely
novel things. Only a moderate amount of novelty is required. Second, like all
other organisms, poets follow Zipf’s (1949) principle of least effort. We ex-
pend the least possible effort in achieving a goal. We are not lazy, but we
avoid needless expenditures of effort in reaching a goal.

Trends within a Style

Consider a poet who wishes to combine words A and B. Once he has

thought of A, he will use as B the first word that comes to mind provided that
the resultant combination has not been used before and does not violate styl-
istic rules. We know from word-association tests that the mental representa-
tions of words are connected to varying degrees and that these connections
are about the same for all people in the same culture (Martindale, 1980,
1991). For example, presented with the word, ‘sun’, most people will think of
‘day’, a few less will think of ‘moon’, less of ‘bright’, and so on (Palermo &
Jenkins, 1964). Very early poets had an easy time of it, as they could say
things so obvious that they are now considered trite—e.g., ‘the bright sun il-
luminated the day’. As time went on, writing about the sun obviously became
more difficult, as poets had to think of things more and more remotely associ-
ated to ‘sun’. The 17th century poet Edward Benlowes compared the sun to a
burning coal-pit in the sky. He was desperate, as all the more obvious things
about the sun had already been thought of. How were they thought of?
Evolutionary oscillations / 331
Psychology of Creative Thinking

If poets must think of more and more novel things to say, we should ask
how novel utterances are thought of in the first place. Martindale (1980,
1995, 1999) showed that all plausible theories of creativity or the production
of novelty are really identical but merely expressed in quite different vocabu-
laries. It is generally agreed that consciousness or states of mind vary along a
continuum ranging from ordinary, wakeful, reality-oriented, rational, prob-
lem-solving, conceptual cognition, through several types of fantasy and then
reverie to dreaming. The further one gets from waking problem solving cog-
nition, the more thinking becomes irrational, free associative, unconcerned
with purpose or problem solving, and dominated by concrete images rather
than abstract concepts. Theorists as diverse as Berlyne (1965), Freud (1900),
Nietzsche (1872), Werner (1948), and Wundt (1896) have given this con-
tinuum various names. For want of a better term, I refer to it as the conceptu-
al—primordial cognition continuum.
In order to have a creative idea, one must regress from conceptual toward
primordial cognition. To have an even more creative idea, one must regress
even further toward primordial cognition. This works only up to a point,
however. Extremely primordial cognition is too diffuse and holophrastic to
yield creative ideas. Thus, creative ideas usually arise from states of reverie
and virtually never from dreams (Martindale, 1995).
In extremely conceptual cognition, ideas or words may be compared to
atoms in a crystal. The atoms (ideas or words) are precisely fixed in place in
extremely conceptual cognition, so that a new combination is completely im-
possible. If we heat the crystal, it will turn into a liquid, and the bonds
amongst atoms will loosen. Now the probability that two remote atoms will
combine with each other is increased. This is analogous to a movement to-
ward more primordial cognition. This movement increases the probability
that remotely associated words will be combined. If we apply even more
heat, our liquid will eventually become a gas. Now virtually any atom (word
or idea) may combine with any other. This corresponds to a state of extreme
primordial cognition. On the mental level, it leads not to novelty but to non-
sense, as the ideas combined will be randomly related. This metaphor
between crystals, liquids, gases, and mental states can be turned into an expli-
cit neural-network theory of creativity (Martindale, 1995).

Intra-Stylistic Trends

From what I have said, we may derive the hypothesis that in order for po-
ets working within the same style to satisfy the novelty rule, successive poets
332 / Aesthetics and Innovation
must tend to regress to more and more primordial states of cognition in order
to write acceptable poetry.
The first poets using a style will, because of the principle of least effort,
write poems using the most obvious word combinations. Later poets will
have to engage in more primordial thinking in order to think of more novel
combinations of words. Yet later poets will have to engage in even more
primordial cognition in order to find yet more novel word combinations.

Stylistic Change

There is a limit to this method of increasing novelty by thinking at a more

and more primordial level. Too much primordial thought leads to less rather
than more novel utterances.
Changing the poetic lexicon. There is a clear way around this problem of
being caught in an evolutionary trap; novelty may also be increased by chan-
ging poetic style. There are two ways of doing this. First, poets may change
the topic (Miles, 1964). In other words, they may add and drop words from
the poetic lexicon. Once poets have said all they can think of about kings and
heroes, they may simply stop writing about them and write about common
people. Wordsworth’s poem, ‘The Leech Gatherer’ is a nice example of this
tactic. The leech gatherer did not need to say anything interesting and cer-
tainly did not do so. To attain novelty and elicit attention, Wordsworth merely
had to introduce such a character into the poetic lexicon. There is not much to
say about leech gatherers, so they made only a cameo appearance in the his-
tory of poetry.
Changing stylistic rules. Style change can also involve changing the rules
for combining words rather than changing the poetic lexicon. This usually in-
volves dropping old rules rather than adding new ones. A very clear example
of this occurred in French poetry around 1900. Until then, if two words were
to be compared in a simile they had in fact to be alike in some at least arcane
manner. French poets very explicitly dropped this rule around 1900 (Mar-
tindale, 1975). The new more permissive rule was that any words at all could
be combined in similes regardless of whether they were alike in any way.
This made writing poetry very easy in that anything could be compared with
anything else. French poets followed the principle of least effort in following
the new rule. Consider Eluard’s “the earth is blue like an orange.” He was us-
ing words or ideas that are closely associated. The earth viewed from outer
space appears blue, and ‘blue’ and ‘orange’ are very close word associates
(Palermo & Jenkins, 1964). No great effort or regression toward primordial
cognition was needed to think of Eluard’s comment. Other surrealist meta-
phors and similes from this era make a certain sense in that they are invari-
Evolutionary oscillations / 333
ably derived from close word associates that may be opposite in meaning but
are closely related in everyone’s’ minds. An example would be Breton’s “seas
red like the egg when it is green.” In this case, ‘red’ and ‘green’ are close as-
sociates, as are ‘sea’ and ‘green’ (Palermo & Jenkins, 1964).

Summary of Predictions

Theoretically the novelty, unpredictability, or entropy of poetic utterances

or art works in general should increase across time. Empirical evidence that
this has in fact occurred was presented above. I have described the develop-
ment of poetic styles in reverse order. When an old style shows signs of ex-
haustion, we expect the gradual introduction of a new style. This involves
adding and dropping words from the poetic lexicon and changing the rules of
the game. During this phase of the new style, we expect primordial content to
decline. Enough novelty or surprise can be gained merely by using the new
words and following the new rules. Once the new style is firmly established,
we expect a decrease in the addition and deletion of words from the poetic
lexicon. Now poets must find new things to say by engaging in more and
more primordial cognition. It will be noted that styles are biphasic. In their
early stages, primordial content should decline. In their late stages, primordi-
al content should increase. Analogous trends should be found in all the arts.
Just as biological evolution is species-specific, aesthetic evolution is genre-
specific. Though it may be influenced by other genres, we expect evolution to
occur mainly in a specific genre in a specific nation.

Tests of the Theory

It is reasonable to assume that a portion of the words one uses when he

writes or speaks will reflect the state of consciousness he was in when he
wrote or spoke. If I am in a conceptual state of cognition, one would expect
that I should use a lot of words having to do with concepts and abstractions.
On the other hand, more primordial cognition involves not concepts but con-
crete images and objects. One would expect that I should write about these. It
could not be otherwise. If I am not thinking about concepts, I could not very
well write about them.
Based upon the works of the theorists mentioned above who have dealt
with states of consciousness, I created a computerized content analysis dic-
tionary to measure how primordial the content of language is (Martindale,
1975). Conceptual cognition is measured by how often words referring to
time, instrumental or purposive behavior, moral imperatives, social behavior,
abstract concepts, restraint, and order are found in a text. Primordial cogni-
334 / Aesthetics and Innovation
tion is measured by the frequency of occurrence of words having to do with
raw sensations, disorder in the external world, body parts, concrete objects,
timelessness, basic drives such as sex, etc. The general measure of primordial
cognition is computed simply by subtracting the percentage of words measur-
ing conceptual cognition from the percentage of words directly measuring
primordial cognition.
I showed that this dictionary is valid by doing studies that found that
more primordial content was found where various theorists said it should be
found. As an example, psychedelic drugs hypothetically put one into a more
primordial state of cognition. Several studies showed that people given such
drugs used more words indicative of primordial cognition. Martindale (1990)
summarizes the large number of studies validating the computer dictionary.
I applied the primordial cognition dictionary to the samples of 18th
through 20th century French and American poetry and to the much larger
sample of 14th though 20th century British poetry mentioned above. The res-
ults were very clear in all three cases. Primordial content oscillated across
time in a manner that could not be attributed to chance. It rose while styles
agreed upon by literary critics were in effect and fell as a new style was intro-
duced. For example, in the case of British poetry, it rose while the metaphys-
ical style was dominant and began declining as this style was gradually re-
placed by the neo-classical style. Once the neo-classical style was firmly es-
tablished, it began to rise again. Results for the British series of poets are
shown in Figure 1. In the figure, I also show results for the series of British
painters and musical composers. In the latter cases, primordial content was
measured with rating scales rather than with content analysis. Similar oscilla-
tions in primordial content were found in the other samples of artistic tradi-
tions mentioned in connection with increases in novelty.
In the statistical study of British poetry, there were enough epochs to
measure the number of words added, dropped, and retained from the prior
epoch. This was a difficult task, but the results were very clear. There were
oscillations in number of words added and dropped (presumably measuring
stylistic change). These were out of phase with the oscillations in primordial
content. This suggests that when a given style is in effect, rather few words
are added or dropped. As a new style is introduced, the number of words ad-
ded and dropped increases. A given style stays in effect for several decades
with impact value mainly being determined by increased novelty due to prim-
ordial cognition. Once the old style is exhausted, words are added and
dropped at an increasing rate, and the index of primordial content declines,
because the new words allow novelty to increase with less rather than more
Evolutionary oscillations / 335
primordial cognition. If the words themselves are new, it is not necessary to
combine them in new ways.

Figure 18-1. Primordial content in successive 20-year birth periods beginning in 1290
for British poetry (top), music (middle), and painting (bottom). Positively accelerated
increases in primordial content have been removed from all three series so the oscilla-
tions may more clearly be seen. Amount of primordial content is not directly compar-
able across the three series.

Range of the Theory

One may object that everything I have said is true, but that I have merely
described several trivial trends that account for very little of poetic or artistic
history. At least in the case of poetry, I was able to compute statistics that
show that almost half of the history of the poetic traditions described above
can be explained by the theory about novelty and primordial content.
As for the other half of the variation in an artistic tradition, presumably a
good bit of it may come from personality differences, differences in sub-
genres, and such forces. Some of it comes from the influence on the one sort
of art upon another within the same nation. For example, British music is in-
fluenced by British poetry and painting though, using 20-year aggregation
336 / Aesthetics and Innovation
periods, poetry and painting did not influence each other and were not influ-
enced by music. However, in another study of British poetry, painting, music,
and architecture using 40-year aggregation periods cross-media synchrony
was observed.
I assume that the cross-media synchrony arises mainly because the vari-
ous arts are oscillating along the primordial cognition axis so will show a
loose synchrony if the periods of their oscillations are similar. Petrov (1998,
2001) has studied oscillations in analytic vs. synthetic content, which is very
similar to conceptual vs. primordial content. He argues for greater synchrony
than I do, because he assumes that the arts will reflect the mentality suffusing
an entire culture. Probably because of slight differences in method that we
have not noticed, Petrov has tended to find more synchrony than I have.
An art form in one nation may be influenced by the same art form in other
nations. For example, Martindale (1990) reports a study of compositions by
252 composers who worked in Great Britain, Italy, Germany (including Aus-
tria), and France (including Belgium) who were born in 21 consecutive peri-
ods from 1490—1909. In order to examine cross-national influences, it was
first necessary to remove secular (linear and quadratic) and autoregressive
trends from primordial content. If this is not done, spurious correlations
would be found (Haugh, 1976). The population of Bulgaria and the gross do-
mestic product of Ireland have increased at an accelerating rate across time
but are not related to one another. Unless the secular increases are removed
from both time series, they would show a high but spurious correlation.
Autoregressive trends describe the cycles found in a time series. Unless they
are removed, spurious correlations would arise if two series happened by
chance to be oscillating at the same rate. After detrending, the average prim-
ordial content score for one country was correlated with mean primordial
content in the other three countries for the same and the three prior periods.
Italy influenced other nations but no significant influence on Italian com-
posers was found. For the other nations, extranational influences accounted
for 61% of the variance in detrended British music, 65% in German music,
and 85% in French music.
At least in the case of British poetry, painting, and music, I was able to
show that trends in primordial content (and for the case of poetry) trends in a
variety of types of content having nothing to do with primordial content were
unrelated to extra-artistic factors. There is no reason to expect the high arts to
reflect society, and in fact they do not.
Virtually no serious theorist concerned with art or literature has argued
that art is in fact very closely related to external society. It is commonly
thought that Marxist theorists argue for such a relationship. In fact, they ar-
Evolutionary oscillations / 337
gue against it. From Marx and Engels (e.g., 1947) through Trotsky (1925),
mainstream Marxist theorists have explicitly said that the superstructure (art
and literature) is quite autonomous from the base structure (economy), so it
follows its own rules of development. As Eagleton (1976, p. 51) put it, art re-
flects society in “the way in which a car reproduces the materials of which it
was built.”
British poetry from around 1320 to the present was correlated with vari-
ous external time series (e.g., prices, wages, wars, average temperature,
philosophical emphasis). The correlation between poetic content and these
external time series was extremely weak. If I took the proper caution that my
findings were not due to pure chance, the conclusion would be that nothing I
could measure in poetry was related to anything that others had measured in
the extra-literary world. The same was true for painting and music. In con-
trast, as would be expected, the lyrics of American popular music are correl-
ated with a number of social and economic indicators. As Golitsyn (2000),
Petrov (2002), and Petrov and Mazhul (2002) have shown, popular art and
the high arts are quite separate genres that follow rather different laws.
The high arts generally seem to operate virtually in a social vacuum under
normal circumstances. The vast majority of people take no interest in them,
and what a poet says in a poem that virtually no one is going to read poses no
threat to the stability of society. However, totalitarian regimes with a desire to
control almost everything can certainly destroy an art form or halt its natural
evolution. Martindale, Kwiatkowski, and Vartanian (1999) studied socialist
realist painting in the Soviet Union. It showed no increases in novelty and no
evidence of evolution. We might say that it began with a painting of a peasant
woman driving a tractor and ended with a painting of her daughter driving a
slightly newer tractor. Since the state controlled both style and content, evol-
ution was not possible.
Architecture, especially large buildings, cannot be created in a social va-
cuum, as the huge sums involved in construction must be provided by large
business firms or governments. Maslov (1983) has found correlations
between political climate and architectural style. Totalitarian governments are
especially prone to build monuments to themselves. Had Germany won
World War II, the plan was to turn much of the center of Berlin into a com-
plex of monumental buildings that was purposely designed to be more impos-
ing than ancient Rome (Spotts, 2003). The plans, which survive, do not sug-
gest much of anything that reflects National Socialist ideology aside from
their grandiosity.
338 / Aesthetics and Innovation
Period of Oscillations

Martindale (1990) summarized a la