|
Webmaster note: The tables and figures in Dr. Petrov's article proved to be difficult to reproduce on-line. Please refer to the print version of his article to view the tables and figures.
Abstract
Recent investigations are reviewed dealing with the creativity in the arts, which are analyzed in the light of information theory. The main contraposition of "analytic" and "synthetic" mechanisms of the information processing occurs connected with the prevalence of left- or right-hemispheric activity. Appropriate stylistic features reveal themselves in the results of creative processes of composers, architects, painters, dramatists, poets, etc., i.e. the stylistic features of their works. According to the theoretical model, each sphere of creativity should show periodical "switches" between the above two types of stylistic prevalence. This mutability is correlated with changes of generations, hence it should possess the full period of oscillations of about 40 - 50 years. Besides, monotonous trends were predicted, that is, the long-range "synthetic growth" (in each kind of creativity) needed to compensate the "analytic growth" in the entire socio-psychological sphere. The evolutionary dependencies obtained characterize both separate features of creativity in each kind of art and its "index of asymmetry". The resulting evidence favors both the 50-year stylistic cycles and the "due" monotonous trends in each of the spheres considered. The synchronism of the changes observed in different spheres confirms the theoretical prediction concerning the importance of the "general style" of the epoch.
Among numerous evolutionary models of the creativity in art (see, e.g., Martindale, 1990), those that are the most prospective are those based on the "information approach". This approach, being of rather general character (see, e.g., Golitsyn & Petrov, 1995), embraces behavior of different systems, that is, with systems dealing with various kinds of primary materials. Hence, certain "invariants" may occur possible, relating to those features of creativity, which are inherent in different branches of the socio-cultural life, and exactly such results are described below. In the present paper the investigations are summarized which were fulfilled during last 30 years, mainly in the framework of St. Petersburg and Moscow Groups of researchers of creative processes.
The model
Although there were certain differences in starting positions, theorists belonging to the above both directions came to almost identical models describing the information processing which is inherent in any developing complex system. Researchers of St. Petersburg school deduced the theoretical model of a "tower for calculations" with different paradigms at each level (Maslov, 1983). In the framework of Moscow school on the basis of the "principle of the information maximum" (Golitsyn & Petrov, 1995), the model of a "pyramid-like hierarchical structure of information processing" was built (Petrov & Kamensky, 1972). Conclusions of both directions of investigations can be described by the following model.
Any complex system of information processing, irrespective of its "material nature" (its components being neurons, or persons, or technical elements, etc.) is nothing else than a multilevel structure. Each M-th level of this structure processes information and transmits a part of it upwards to the next, M+1-th level, and this higher level formulates the criteria to choose information for its own needs (and gives these criteria to the lower, M-th level) and also transmits part of the information processed to the next higher (M+2-th) level, and so on. So two types of processes regarding information are inherent in such a system: First, information processing at a given level, organized according to definite rules (paradigm); small portions of the information received are processed in consecutive order, with very high precision; this type of activity is characterized by analytic and rational features, an important role for logic, and so forth. Second, information transmittal from one level to a higher one; this activity is characterized by a change of paradigm (rules of information processing); rather large amounts of information are processed parallel to each other, but with relatively low precision; this activity may be described as synthetic, emotional, intuitive, and so forth.
A significant difference thus exists between these types, the first being primarily analytic; the second, mainly synthetic. In the case of a human being, these two types may be attributed respectively to the left- and right-hemispheric activity of the brain. Naturally, such an identification is relative and concerns mainly functional features of processes rather than their localization in the brain. Nevertheless, many clinical experiments corroborate the connection between the predominance of the left or right hemisphere and the prevalence of left-hemispheric processes or right- hemispheric ones. The hypothetical distinction between these two types of information processing permit one to deduce their main characteristics, some of which are summarized in Table 1.
On the basis of these characteristics Maslov (1983) deduced two "sets of gnosiological preferences" (Table 2) which should be observed when strong prevalence exists for either left-hemispheric processes or right-hemispheric activity (referred to hereafter as L- and R-type processing). The distinction between the two types of information processing permits us to draw three hypothetical conclusions: First, at every given moment, a society needs to have a definite degree of domination of one (L or R) of the two types in the entire socio-psychological sphere, as well as within each of its branches (creativity in music, painting, poetry, etc.). It is always desirable to use a certain common "style of thinking" instead of a Tower of Babel of different styles. Of course, this degree of L (or R) domination need not refer to all intellectual phenomena, but it should encompass the majority of them; in this case any communication becomes easier, otherwise, communications become more complicated, and the efficiency of information processing decreases. Second, the dominating type (or "style of thinking") has to change from time to time, because each type possesses rather limited possibilities with respect to further progress of the information processing. If any given paradigm of information processing dominates too long, its possibilities become exhausted, and no new innovations (inventions) can arise within this paradigm. Therefore, the cycles must "switch" (more or less periodically) between L and R dominating styles of thinking. Third, these switches, in turn, possess a limitation, as far as their frequency is concerned, because the real carrier of any style of thinking is a human being who belongs to a definite generation and therefore has a definite degree of L (or R) domination. So any style of thinking should last at least one generation, giving up its dominance only after the decay of the generation which used it. As soon as each given generation dominates the socio-psychological sphere for approximately 20 to 25 years (because of the duration of the reproductive cycle), the full period of the predicted oscillations has to be 40 to 50 years or even more. Now let us turn to concrete indicators of L- and R- domination which can be observed empirically.
Evolution of creativity: Procedures of studies and the results obtained
To describe left- or right-hemispheric prevalence in different spheres, some parameters were chosen, relating to the creativity of eminent artists (painters, composers, poets, etc.). For each kind of art a special iterative procedure was derived, using experts as specific "instruments" (Golitsyn & Petrov, 1997; Petrov, 1992, 1994, 1996; Petrov & Boyadzhiyeva, 1996; Pokorskaya, 1992; Shepeleva, Batov, & Petrov, 1997). The results of such procedures were sets of parameters, substantial for the phenomenon in question in each kind of art. For instance, for painting, the following 10 parameters were deduced, with each parameter having a form of a binary opposition, left and right pole of which responded to left- and right-hemispheric prevalence, respectively:
a) Inclination to normative work - Inclination to originality, peculiarity;
b) Rationality - Intuition;
c) Strict form - Free form;
d) Laconism, asceticism of expressive means - Variety, diversity of expressive means;
e) Graphic features - Picturesque, coloristic features;
f) Restrained, static features - Expressive, dynamic features;
g) Discrete elements - Continuous transitions between elements;
h) Inclination to cool part of spectrum - Inclination to warm part of spectrum;
i) No color gradations - Significant color gradations;
j) Smooth painting - Textured painting.
On the basis of such parameters, procedures were derived to measure the creativity of each artist. Though these procedures involved expert estimations, the results obtained were objective. At last, the "index of asymmetry" was introduced for each artist, evidencing the degree of left- or right-hemispheric prevalence in his/her creativity.
The method derived was used to study the evolution of different fields of creativity. For each field, an appropriate sample was compiled, depending on the specific character of the sphere considered. In most investigations the compilation concerned the so-called "elite" of the field considered, i.e. the most eminent composers, painters, dramatists, etc. Though in several investigations the compilation followed other strategies. For instance, when studying the evolution of poetry, the sample involved several thousands poets, that is, "mass creativity" - see, for example Ivanchenko & Kharuto (2000). To compile such "elite sample," usually fundamental encyclopaedia for each field were examined, and a preliminary list of eminent creative persons was prepared. Then out of this list those persons were chosen (by experts) who strongly influenced the evolution of the field were considered. For instance, in the field of music (Petrov & Boyadzhiyeva, 1996) on the basis of the 20-volume encyclopaedia by S. Sadie (1980) a final list of 102 such "influential" persons was compiled, consisting both of West-European and Russian composers of the 17th to 20th centuries. For the field of painting, such a final list consisted of 240 "influential" painters of the 14th to 20th centuries, both West-European and Russian.
Each of such creative persons was evaluated by several experts; for instance, in the field of music, usually the creativity of each "influential" composer was "assessed" by 6 to 17 experts. Each expert was firstly "calibrated" using those "contrastive" persons which had been previously singled out in the experiments described. On the basis of such calibration, not only "zero points" were determined for each expert (on each scale used), but also his/her "level of competence", meaning the feeling for art style. Some experts who revealed low reliabilities were not allowed to participate in further investigations. Then the data of all the experts involved, concerning the creativity of each person studied, were aggregated. A fragment of the results of such aggregating in the field of painting, is presented in Table 3. Appropriate data for composers and poets were presented by Danilova and Petrov (1990) and Koshkin and Frizman (1991).
These results give two kinds of characteristics of each creative person: First, is the mean value of each parameter (a - j) of his/her creativity. As soon as all experts' scores were binary, the result of aggregating is nothing else but a share of L-scores out of the totality of scores put by all experts involved. Hence, this share can change from 0 to 1, or the percentage from 0 to 100%. For instance, the creativity of Canaletto (see Table 3, first line) was estimated by 9 experts. On the parameter (a) "Inclination to normative work - Inclination to originality, peculiarity" the scores of 7 experts were interpreted as L-evidences. So, the value of this parameter for Canaletto's creativity is 7/9 = 0.78 (see Table 3). Second, is the mean value of the index of asymmetry (K) for the creativity of the given person. As soon as this value was obtained by averaging over indices calculated on the basis of the data of different experts, each of these indices could range from -1 to +1. This mean value also could range in the same diapason, or if we turn to percentages, from -100% to +100%. For example, on the basis of the first expert's scores, the asymmetry index for Canaletto equals 0.60 (because this expert put L-evidences on 6 parameters and R-evidences on 4 ones); on the basis of the second expert's scores this index equals -0.33 (because he put L- evidences on 3 parameters, and R-evidences on 6 ones), and so forth. After averaging these values, the resulting mean index occurs to be 0.44, which corresponds to the percentage 44% presented in Table 3 (last right column).
Both kinds of characteristics were used to build evolutionary dependencies. The values for such curves were received by averaging appropriate values for those persons who were actively working in each given time interval; in most cases 3-year intervals were used. As for the temporal diapason of each person's "active work", due formal criteria were introduced, concerning personal "creative age". For example, the time interval of 1847- 1849 was presented in the "elite" sample for musical creativity, by 12 composers: Bruckner, Chopin, Franck, etc., so the parameters of these 12 composers were averaged to obtain their mean values for this time interval. The evolutionary dependencies were built separately for Western Europe and Russia. However it occurred that in most cases both regions revealed almost similar behavior (especially beginning from the middle of the 18th century), which is in good agreement with the data of Maslov (1983). That is why the further dependencies and conclusions relate to both regions are considered together, if not specified especially.
In conformity with the above two kinds of characteristics, two kinds of evolutionary dependencies were built (see, e.g., Petrov, 1992, 1994, 1995, 1998; Petrov & Yevin, 1991; Petrov & Boyadzhiyeva, 1996): first, evolutionary behavior of different "partial" parameters of the creative asymmetry in each field studied. For example, Figure 1a presents the evolution of such a parameter of painting, as "Rationality - Intution". Each point of this curve designates the share of L-scores placed by all the experts involved to this creative feature of all the painters studied actively working at this time. Figure 1b relates to quite an analogous dependence of another feature of the painting, namely "Restrained, static features - Expressive, dynamic features" (parameter f). Such curves were built for all the parameters of all fields studied. Second, evolutionary behavior of the "index of asymmetry" characterizing each field as a whole are presented in Figure 2. This curve describes the entire creativity in this field, meaning the evolution of its stylistic inclination either to the left-hemispheric pole or to the right-hemispheric one. Evolutionary dependencies of such indices were also built for each of the field studied. So the results obtained characterize both the evolution of some concrete features of creativity in different fields and general evolutionary behavior of each of them. Namely the last kind of dependencies is of most interest for our further consideration.
Regularities of the evolution: Cyclic behavior against monotonous trend
All the fields studied reveal a similar evolution of their indices of asymmetry, which can be illustrated with Fig. 2 for the case of painting. Such behavior contains two components. First, the stylistic "coloring" of each field possesses a pronounced monotonous trend: the growing role of the synthetic (i.e. right-hemispheric) domination. Such a trend can be easily seen in Fig. 2 where we see a decreasing curve for the index. Second, against this background, one can see featured periodic changes (or maybe semi-periodic ones, because their periods possess appreciable fluctuations). Besides, both components can be seen also when analyzing "partial" curves (i.e., evolution of separate parameters), but naturally, the curve for the index reveals these components much better - compare, for instance Figures 1 and 2.
The best way to divide ("segregate") these components is to use the middles between neighboring extreme points (i.e. the closest maximum and minimum of the curve). The result of such "segregation" is shown by Figure 3a, which presents a fragment of the periodic component of the index of asymmetry for painting. As can be easily seen, the changes observed are really close to periodic ones, possessing the full duration of cycles about 50 years, which is in good agreement with theoretical predictions. Fig. 3b shows the periodic component of the index for music, obtained by the same way.
Table 4 presents the data concerning the "waves" of L- and R-prevalence (i.e. the diapasons of positive and negative values of the periodical component of index K) for several fields. According to this Table, the mean durations of all these waves are very close: from 22 to 25 years for L-waves, and from 26 to 27 years for R-ones; the full cycle lasts from 48 to 52 years. The agreement between such waves in different fields is not absolute, but rather good. For instance, the comparison of the evolutionary dependencies for the indices of asymmetry for painting and music (curves presented by Fig.3, a and b), is evidence in favor of the coincidence of appropriate waves THIS IS CONFUSING, which is not occasional statistically (at 5%-level of significance). As a rule, the changes in painting slightly predate appropriate changes in music (approximately on 6 years). Nevertheless, some discrepancies take place, so the similar evolution of different fields can be called "soft synchronism".
This synchronism is further evidence for the unity of the entire socio-cultural sphere (this unity was fixed earlier by many historians, though only on the level of simple, qualitative observations). One can see that the socio-political "climate" reveals the same waves, as various kinds of art. Moreover, it was shown that these waves are coordinated with the so-called "Kondratiev waves" (Kondratiev, 1989), which are inherent in the economic sphere. Moreover, these waves can be observed in different fields and can provide a basis for the definition of the stylistic "epoch" in the evolution of style. It seems natural to define the epoch as the temporal range between the two neighboring changes of sign of the evolutionary curve for the index of asymmetry. Such an approach follows the above mentioned logic of "invariant" definitions. For instance, on the basis of the formal definition, it is possible to single out such a stylistic epoch of R- domination in painting, as the range of 1858 - 1916 (see Table 4), which embraces the creativity of rather pronounced right-hemispheric impressionists and their followers.
Though quite opposite, left-hemispheric directions in painting also appeared at the end of this range, including cubism; nevertheless, the general prevalence belonged to painters of R-orientation.
Such wave-like behavior was recently supported by two additional, purely statistical investigations that dealt not with qualitative parameters of creativity, but with quantitative ones. In one of these investigations (Petrov, in press), the geometrical sizes of paintings (i.e. their width and length) were examined, and it occurred that these parameters also showed 50-year cyclic behavior tightly correlated with the above stylistic oscillations, in agreement with the model of the "arousal potential" (Berlyne, 1971; Martindale, 1990). In another investigation, the "intensity of literary life" was measured as the number of lines (in literary encyclopaedia) devoted to poets and prose-writers of different epochs. Here also 50-year cyclic behavior was detected, tightly correlated with the above stylistic waves (Kharuto, Majoul, & Petrov, 2000).
Turning to the long-range monotonous component of the evolution, we should mention, first of all, the general direction of this trend which is identical in all kinds of art: constant growth of R-domination (i.e., decreasing L-domination - see, e.g., the negative slope of the evolutionary curve for the index of asymmetry in painting, Fig. 2). Such direction is not surprising, if we take into account the so-called "compensatory" function of art. Really, in order to arrange the entire mental life (both individual and social), it is desirable to maintain a certain "balance" between its analytic and synthetic constituents. Meanwhile, usually each system of mental life is constantly moving towards the growing role of analytic processes (especially in such fields as technology, communications, politics, etc.). One of the main ways to compensate for this tendency is to apply means of artCONFUSING; so the last one should increase its R- features. That is why we observe the inclination toward R- prevalence in all fields of art, both in the evolution of their "partial" parameters and in their indices of asymmetry.
To support this thesis, let us compare the long- range components in painting, for two national schools: French and Russian (Gribkov & Petrov, 1997). Figure 4 shows the evolution of this component in both national schools (over the 18th to 20th centuries), presented by one of their "partial" parameters ("Rationality - Intution") and their indices of asymmetry. To "segregate" the periodical component and the long-range trend, the above mentioned procedure was used. It occurred that the decreasing slope of L-features is much more steep in French paintings than in Russian. This difference was observed when comparing both the evolution of all 10 "partial" parameters and the evolution of painting as a whole. In all cases appropriate differences were statistically significant at the level better than 5%. Evidently, it can be explained by the fact that in France the evolution of the entire socio-psychological sphere (especially in the fields of technology and politics) was much more dynamic than in Russia (where the social life and technological development were "sleepy").
Due to such "compensation" (though of course incomplete), the entire sphere of culture reveals the "divergent evolution" which has been theoretically deduced in the framework of the information approach (Petrov, 2000): each field increases its own, genuine (specific) features; as a result, we observe an "Expanding Cultural Universe". For example, according to Martindale (1990), the "primordial" component of creativity reveals long-range decline in science, whereas in the arts it is increasing. In addition, two main directions of prose (oriented either on the structure of its ouvres or concrete material, i.e. either on L-processes or R-ones) reveal a "splitting" and divergent evolution (Petrov, 1994a).
The totality of the data reviewed confirm the "divergent" concept of the evolution. Moreover, such data can be used to forecast the future of culture and art (Mazhul & Petrov, 1999).
References
Berlyne, D.E. (1971). Aesthetics and psychobiology. NY: Appleton-Century-Crofts.
Danilova, O.N., & Petrov, V.M. (1990). Brain asymmetry and creativity: A model and attempt of quantitative investigation. In L.Halasz (Ed.), 11th International Congress on Empirical Aesthetics. Proceedings. Budapest: Institute for Psychology of the Hungarian Academy of Sciences. P. 45-48.
Golitsyn, G.A., & Petrov, V.M. (1995). Information and creation: Integrating the "two cultures". Basel, Switzerland: Birkhauser Verlag.
Golitsyn, G.A., & Petrov, V.M. (1997). Styles of creativity: Measurement of changes, their cultural determination, and the problem of "free will'. In Yu.S.Zubov & V.M.Petrov (Eds.), Aesthetics: Information approach (pp. 123-136). Moscow: Smysl.
Gribkov, V.S., & Petrov, V.M. (1997). Hemispherical asymmetry in creativity: Long-range trend in painting. In V.P.Ryzhov (Ed.), Empirical aesthetics: Information approach (pp. 125-133). Taganrog: Taganrog State University of Radio Engineering.
Ivanchenko, G.V., & Kharuto, A.V. (2000). Evolyutsionnaya dinamika sistemy poeticheskogo tvorchestva (Russkaya poeziya 1800 - 1980). In L.Dorfman, C.Martindale, V.Petrov, P.Machotka, D.Leontiev, & G.Cupchik (Eds.), Tvorchestvo v Iskusstve - Iskusstvo Tvorchestva (pp. 485-501). Moscow: Nauka; Smysl. (in Russian). [Evolutionary dynamics in the system of poetical creativity: Russia, 1800 - 1980].
Kharuto, A.V., Majoul, L.A., & Petrov, V.M. (2000). Cyclic literary life in Russia: Prose of the 18th to 20th centuries. In V.M. Petrov & V.P. Ryzhov (Eds.), Information Paradigm in the Human Science (Proceedings of the International Symposium) (pp. 163-165). Taganrog: Taganrog State University of Radio Engineering.
Kondratiev, N.D. (1989). Problemy Ekonomicheskoy Dinamiki. Moscow: Ekonomika (in Russian). [Problems of Economical Dynamics]
Koshkin, V.M., & Frizman, L.G. (1991). Byt' poetom (opyt statisticheskoy literaturometrii). Chelovek, N 3. (pp. 79-82) (in Russian). [To be a poet (a trial of a statistical literature measurement)]
Martindale, C. (1990). The clockwork muse: The Predictability of artistic change. NY: Basic Books.
Maslov, S.Yu. (1983). Asimmetriya poznavatelnykh mekhanizmov i yeyo sledstviya. Semiotika i Informatika, N 20. (pp. 3-34) (in Russian). [Asymmetry of cognitive mechanisms and its consequences]
Mazhul, L.A., & Petrov, V.M. (1999). 21st century art: Components of the midterm prognosis. In 20th Century Art: Achievements, Traditions, and Innovations (pp. 345-347). St. Petersburg: State Hermitage.
Petrov, V.M. (1992). Evolution of art and brain asymmetry: A review of empirical investigations. In G.C. Cupchik & J. Laszlo (Eds.), Emerging visions of the aesthetic process (psychology, semiology, and philosophy) (pp. 255-268). Cambridge, England: Cambridge University Press.
Petrov, V.M. (1994). Problemi di stile: espressioni caratteristiche dell' emisfero sinistro e di quello destro nell' arte. Rivista di Psicologia dell' Arte, Nuova Serie, anno XV, nn 3/4/5. P. 23-34.
Petrov, V.M. (1994a). Perception of prose: A struggle with the matter of the texts (A trial of an empirical investigation). SPIEL (Siegener Periodicum zur Internationalen Empirischen Literaturwissenschaft), Jg.13, Heft 1. P. 1-21.
Petrov, V.M. (1995). Evoluzione dell' arte e asimmetria cerebrale. Rivista di Psicologia dell' Arte, Nuova Serie, 16, 4-9.
Petrov, V.M. (1996). Quantitative estimates of left- and right-hemispherical dominance in art. Leonardo, 29, 201-205.
Petrov, V.M. (1998). The evolution of art: An investigation of cycles of left- and right-hemispherical creativity in art. Leonardo, 31, 219-224.
Petrov, V.M. (2000). Divergent evolution of cultural sphere: Cross-media and cross-genre interactions. In P.Locher & L.F.Smith (Eds.), 16th Congress of the International Association of Empirical Aesthetics. Proceedings. (pp.159-161). Upper Montclair, NJ: Montclair State University Press.
Petrov, V.M. (in press). Sizes of paintings: Evolution in the light of the arousal potential. Rivista di Psicologia dell'Arte, Nuova Serie, 32,
Petrov, V.M., & Boyadzhiyeva, L.G. (1996). Perspektivy Razvitiya Iskusstva: Metody Prognozirovaniya. Moscow: Russky Mir (in Russian). [Perspectives of Art Development: Methods of Forecasting]
Petrov, V.M., & Kamensky, V.S. (1972). An informational model of emotions accompanying perception. In 20th International Congress of Psychology. Abstract Guide. (p. 443). Tokyo: Science Council of Japan.
Petrov, V.M., & Yevin, I.A. (1991). Emotional unity of the informational sphere: Socio-psychological "climate", art, and economics. In L.Ya. Dorfman, D.A. Leontiev, V.M.Petrov, V.A.Sozinov (Eds.), Art and Emotions (pp. 160-169). Perm: Perm State Institute of Culture.
Pokorskaya, Ye.Ya. (1992). The search for parameters in dramatic art. In L.Ya. Dorfman, D.A. Leontiev, V.M. Petrov, & V.A. Sozinov (Eds.), Emotions & Art (pp. 315-327). Perm: Perm State Institute for Arts and Culture.
Sadie, S. (Ed.) (1980). The new Grove dictionary of music and musicians, vol. 1 - 20. London: MacMillan.
Shepeleva, S.N., Batov, V.I., & Petrov, V.M. (1997). Left- and right-hemispherical features in poetry: A method of measurement. In L.Dorfman, C. Martindale, D. Leontiev, G.Cupchik, V.Petrov, & P. Machotka (Eds.), Emotion, Creativity, and Art, Vol. 1, (pp. 287-302). Perm: Perm State Institute of Arts and Culture.
Figure captions
Figure 1. Evolutionary curves for "partial" features of painting - share of L-evidences (S) for the parameters:
a) Rationality - Intuition;
b) Restrained, static features - Expressive, dynamic features.
Figure 2. Evolutionary behavior of the index of asymmetry (K) for painting.
Figure 3. Fragments of the periodical components of indices of asymmetry (K) for painting (a) and music (b).
Figure 4. Comparison of the long-range components in French and Russian painting (designated by triangles and rhombes, respectively):
a) Parameter "Rationality - Intuition" (S);
b) Index of asymmetry (K).
