WALTER SIEGFRIED
DANCE, THE FUGITIVE FORM OF ART

Aesthetics as Behavior

( in: Beauty and the Brain Birkhäuser, Basel 1988)

To analyze dance as an early form of aesthetics means to accept the idea that the aesthetic is not a quality of tangible objects alone. By stressing the fluent and fugitive side of art, dance emphasizes action and not a resulting object. The notion that artistic behavior does not always result in an object is often neglected, but here it is our main thesis. We shall focus on the dynamic character of art; aesthetics will be considered as behavior. Dance is most suited to visualize this dynamic side of aesthetics, which also can be objectified to demonstrate that something persisting has been created. The regularity of the generated rhythmic patterns or the stabillty of certain body orientations among the dancers provide prime examples. Dance, the fugitive art form, shows by such qualities as repetition, regularity, and constancy a tendency towards externalization. Since the beginning of this century, the value of isolated works of art has been questioned and a reverse process can be observed. Interest is shifting from the graspable work of art towards the action of producing it. This first appeared in the expressionist school of art, and the result of the expression, the painting, remained an object of art. Since then, action has been increasingly stressed (action painting"). but the result is still important. Finally, the action itself has become the aim (situationists 1, happenings, performances).

In emphasizing, the behavioral aspect of the aesthetic, we shall focus on the specific qualities of dance movements and examine what separates them from ordinary motor activities. Such an analysis of dance movements is futile, however, without a concept for classifying the movements. Different classification systems have been proposed. most of the earlier ones in the context of choreocraphy.

To get a clearer picture of the relationship between dance movement and ordinary movement, it is not enough to describe the isolated dance movements, which are often formally identical to everyday activities. Rather, we must look at the dancing group as a whole and ask what transforms those similar looking movements into something other than everyday activity. Two approaches, both holistic in their roots, have been chosen to determine the typical quality of dance movements: phenomenology and ethology.

Phenomenology, with its rich tradition in observing the human body in motion2-4 presented a holistic approach. Only after a phenomenological description of dance movements had been undertaken5.6, did it become clear in this investigation that it was not the movement itself but the context in which it occurred that necessitated further observation. In this investigation, context" refers not to functions or cultural influences but to the movement context, i.e., what the dance movement is related to. What is the system within which the dance movements take place? The hypothesis of this study, is that the system to which the dancers and their movements relate must be created by the dancers themselves and is the result of a group process. The reference points of the dance movements are elaborated by the dancers or, as Maxine Sheets5 puts it: A dance cuts into everyday time and space ... Consequently, one can speak of the specific temporal flow of a specific movement and the time of the dance as a whole as created temporalitv and created time ... Time and space are not containers in which dance occurs: they are intrinsic dimensions of the illusion. virtual force, and as such, they are immediately created with the creation of that force.....". The measurable description of this spatio-temporal creation, as far as methodologically possible, is the essence of this study.

Regarding the ethological approach, it was precisely the analysis of movement pattems, especially of ritualized movements, with which ethology, a new branch of biology,, contributed a new dimension to the theory of evolution.7 As Lorenz showed in his article On Dance-like Movements in Animals,"8 ritualized animal movements resemble human dance. Because this study concentrates on human dance, the etholocical perspective here focuses on the description of movement pattern, on the analysis of their coordination process, and on possible laws that govern this process rather than examining the whole richness of dance events. This approach, based on the most simple (and probably, also the most original) forms of human aesthetic behavior, links dancing to the highly developed communicative behavior of animals and provides the best basis for the question of how behavior and aesthetics may work together in human dancing. In the domain of art, dance is the most suitable medium to demonstrate the cooperation between behavior and ideas, a cooperation that is also present - but less visible - in normal human movement. Aesthetic ideas, as conveyed through books, painting, sculpture, etc., are presented by the vehicle the human body in dance without the presence of an extraneous object. Therefore, the traditional analysis of aesthetic objects must be replaced in this study by the analysis of aesthetic behavior.

Investigating an Ephemeral Phenomenon

The working hypothesis of this study defines group dancing, as a behavior in which dancers create, maintain. and vary a common space-time structure (S-TS), which, once created, is binding on the participants. This model of a specific S-TS links behavior and aesthetics by providing a tertium comparationis (the point in which two seemingly different phenomena are similar) between highly developed, ritualized animal movements and the early artistic expression of man in dance. Spatial as well as temporal structures established by the participants are crucial for both phenomena and provide a basis for comparison. In both cases, they provide a framework within which to behave.

What is meant by this spatio-temporal framework" within which human dancing takes place? One of the most basic requirements of group dance is to be in the right place at the right moment, and this is essentiell what is meant by the S-TS. But who determines the right place and the right moment? For ballroom dancing" this is obvious: the right moment is indicated by the rhythm of the music and the right place results from learned dance movements, fixed by choreography. But if, for exarnple, in children's play dances, there is no space-defining choreography and no time-keeping music, what then? We will show how the S-TS is established successively by the individuals dancing in the group. But first we must explain more clearly the meaning of the term space-time structure We will initially analyze the space structure and then the time structure, since both processes are strongly interrelated.

The space structure is defined as. the continuina spatial relationships amoung the dancers. Two methods will be used to show its development. First, drawings that were directly copied from film frames provide insight into the successive building up of the dance space. Second, a transcription method introduced by Kendon9 and Deutsch10 represents each body as a transactional segment," thus define, the shape of an individual's contribution to a joint interactional space" (Fig. 1).

Deutsch described the construction of the transactional segment as follows: We can arrive at just what the shape of this space is by the following: Imagine a person standing, direct with his hands at his sides. First we anchor this person in space and time by drawing a connecting line between the two points on the ground created by dropping, a perpendicular line from the lateral extension of each shoulder. The length of this line (line a in Fig. 1) is approximately two feet for an average adult. We call this line the base of the transactional segment. From each of the two endpoints of this base we extend lines drawn at 45-degree angles to the straight ahead (line b in Fig. 1). These lines were drawn at 45-degree angles because, as we saw, this angle is the maximum allowable discrepancy in the head-body orientation of an interactant.

We call these lines the legs of the transactional segment. Finally a line is drawn parallel to and two feet in front of the base (line c in Fig. 1 ). This line is drawn thus because two feet is half the maximum distance (four feet) between two people interacting at Hall's personal distance": this line thus equally divides the space between the two interactants. We call this line the crown of the transactional segment. If we extend this crown in both directions until it interesects each leg, we create a bounded space which the leg being two feet ten inches long and the crown, six feet long. We hypothesize this transactional segment as an individual`s contribution to the joint interactional space of a face to face interaction.

Just as face-to-face interactions may be viewed within this conceptualization of interactional space, so may dance groups. lf one transcribes typical space configurations of simple dance forms by the interactional segment, the space structure and its characteristics are easily recognizable (F.2).

Such simple space configurations (floor parterns) as e.g.: a) the circle; b) and c) the line: d) and e) face-to-face differ fundamental, from one another vis-a-vis the relative positions of the bodies to each other. For instance, all participants are equal in the circle, in the line, there are a first. a last, and some middle positions: and in the face-to-face configuration, the pairs are emphasized.

In patterns A and B of Fig.3, the dancers always stand on the same spot in objective space and only change their orientations to each other slightly. But this slight spatial alteration provides a radical chance in the meaning of the dance space: the members are embedded suddenly in a new spatial context. The changed context creates a new situation. For example, B1 is a pronounced face-to-face situations, which could be characteristic of a courtship dance. B2 has professional character and B3 with its clear double front can be found in war dances. Here are three versions in which each dancer remains on his spot in objective space but participates in very different dance spaces

Given the concept of our research, it was clear that neither the analysis of stage dancing nor of ballroom dancing, can be used to demonstrate the creation of space or time structures because, in those dances, the choreography or traditional positions dictate the space structure, and the given music dictates the time structure. We, therefore, looked for dance events that occurred without music. In such situations, the dancers are forced to create their common place themselves, and, during such opening, phases. the relative positions among the dancers are not yet totally fixed so that the creation of the space structure can be observed. Children's dances and play dances meet these requirements.

Among, the many documents in the Human Ethology Film Archives of the Max-Planck-Society, the film sequences of southem African dances provided the richest examples for our research. In this region, we find living dance traditions, and dancing, children can often be observed.15 However, only a few sequences of these films could be used for complete analysis.16 Other research material consists of documents of experimental dance groups, which consisted of medical students. The size of the groups varied from four to fifteen. In these dance groups, we played with spatial and temporal elements by a grouping in different ways and producing different kinds of rhythms. Protocols, photographs, and tape recordings, partly transcribed by oscillograms, were the documentary tools used. As mentioned, this paper tries to emphasize dance not only as an aesthetic object but also as an act in which to participate. Therefore, our own experiences in dance groups were important.

The materials used in this study are obviously heterogeneous and are only examples to help illucidate the model proposed in the hypothesis. We do not intend to explain the meaning of the dances. To understand dancing fully, one must live within the dancing culture not only as an observer, but as a participant. Without such living participation, analysis intended to crasp the whole sense of dancing, is open to all kinds of misunderstanding us. Difficulties begin with the word dance," whose corresponding terms in different cultures refer to different phenomena.17 Because our study looks for very simple things, such as spatial orientation within a group or the temporal development of the pulse of a dance, the meaning of the analyzed dances can, at first, be neglected in our approach.

The Hypothesis

The core of this study's hypothesis is summarized in the term common space-time structure." This structure is established at the very beginning of the group dance and is then binding on the participants by defining the spatial and temporal reference points for their mo-vements. The development of the S-TS is the context of our endeavor.

Because dance is a fluent rather than a static, palpable art form, such as sculpture, the S -TS cannot be regarded as an object; it must be viewed as an ongoing process. Its aesthetic qualities must, therefore, be rooted in its dynamic creation and development. Focusing on the dynamic aspect of the dance. the S-TS must not only be created but also maintained and varied. Without active maintenance, the once established S-TS may, break down very quickly; either the spatial coherence will become loose or the synchronized movements disintegrate into individual rhythms, causing a chaotic destabilization of the common tempo.

Maintaining the S-TS, i.e., continuing the same rhythm in the same spatial positions, soon becomes boring there must be variation in the spatial configuration as well as in the temporal flow. lt is especially these elements of playful variation that give dance the qualities that belong so fundamentally to aesthetic behavior and provide intrinsic reinforcement - thrill, pleasure, excitement, innovation, display of individuality, and so on. They need, of course, a background from which to evolve as variations. This background is not just space and time in general but the specific S-TS created and maintained by the participants.

Dance is not only the creation of a S-TS. As a dynamic process, it must constantly be reculated by the participants while it is being developed. This regulation may, at one time, consist of variation to avoid boredom. At another time, it may stress the basic S-TS to ensure continuity of the dance in a risky situation. Our description may be summarised in the following hypothesis: dancers create, maintain, and vary a common S-TS, which, once created, is binding on the participants. In the following sections, we shall present three aspects of the findings corresponding to the hypothesis:

Creation of the S-TS

Maintenance of the S-TS

Variations within the S-TS

In the last section, we shall discuss the consequences of this definition of dance for the concept of aesthetics. Special attention will be paid to the fact that dancers are the creators of the work of art, which they themselves represent and from which they personally benefit. This approach stresses perception as an active process. In dance, receptivity and activity do not exclude each other, but provide mutual reinforcement. This cooperation of perception and motion coincides perfectly with our concept of aesthetics, which stresses the unity of perception and behavior within a social context.

The Creation of the Space-Time Structure

Dance croups are defined by their specific space structures. If one decides to dance, one goes into the position required, but if dancing is happening without big decisions and defined beginnings, then finding the right" position in space is a process. For example, the oryx-antelope dance or gemsbuck dance of the Iko-Bushmen (Kalahari Desert, Botswana) (Fig. 4) may be initiated by one man (A) stamping clearly the basic rhythm in the middle of a relaxed group (frame 0234).

A second man (C) stands up and begins to dance facing, A, while B and D are still standing around in relaxed positions. A young member of the sitting group stands up (frame 0346) and enters the dance to challenge dancer A for a short period of the dance, thus an axis is created with A on one side and the boy and C on the other side. Now D moves closer to C, orientating his body axis towards A (frame 0652). All heads seem to be asking, for participation of the sitting group. In frame 0924, a member of that group steps next to C, while B changes his orientation so that now, for the first time, the basic space structure is realized - the antelope" in opposition to the group (frames 0924/1214).

Similarly, the girls, already moving in the typical pattern of a dance, do not yet have strong mutual spatial relationships (Fig.5). Only later do they create, step-by-step, a circle (Fig.6). At the beginning, neither body axes nor faces are correlated among the dancers, whereas later their mutual orientation creates the framework within which to behave. The time required to establish such a space structure varies from case to case and depends on many variables, such as mood, knowledge of the intended space structure, will of the participants to dance the same specific form, and so on.

Returning now to the group of girls that established a circle as a space structure, we ask how they created their time structure. In other words, how do they synchronize their individually generated movements? The girls, dancing in this circle, jump up and down, first in the middle position (m). later forward (1f) and backward (b).

Figure 7 provides a schematic idea of this spatial development, which introduces the time factor by the jumping movements of the dancers. The rhythm, or pulse, of each individual can be defined as the succession of the jumps. Each jump is considered as one beat of the pulse. The jump is defined by the deepest point of flexion in the knees (Fic,.8). It coincides perfectly with the feeling, of the pulse beat in this dance that works with fallen movements, with the force of gravity. Furthermore, the point is easily detectable by single-frame analysis.

If one uses this method to define the pulse beats for each individual in a dancing group and puts them together on a graph (Fg.9), the creation of the common time structure is visible. At the beginning (frames 100ff), each individual jumps in her own rhythm, later (frames 600ff), the individual pulse sequences are synchronized.

These two processes - synchronization of pulse beats (creation of time structure) and orientation into the dance space (creation of space structure) are not separate. They are treated separately here only for purposes of analysis. To demonstrate their unity, one must show where the pulse beats occur in the dance space (Fig.7). If one combines this spatial signature (m,f,b) with the time signature (the succession of the pulse beats), the creation of the S-TS becomes clearly visible (Fig. 10 and 11). These two charts are the visualization of the first and most important part of the hypothesis: the successive creation of the common S-TS. At the beginning, (frames 100 ff), the jumps are hap hazard in space as well as in time, but the S-TS is clearly established around frame 560 - that is, after about 22 seconds.

The schematic transformation used here is a simple and rough version of what we were looking for in the methodological section. It shows how the framework, or the system of reference points, is created by the dancers themselves. The objective time represented in the succession of film frames is negligible, as it is not important whether something happens in frame 600 or 610. The importance for the dance is whether that something happens simultaneously among all four dancers or not. Similarly, it is not the coordinates in objective space that are of interest but the specific space structure of this dance form. This is the relative spatial relationship of the dancers to each other. Only when one understands the integrative form toward which the movements tend to be directed does the specific dance space become perceptible, allowing us to analyze, retrospectivelv, the creation process of the space structure by providing the reference points. These spatial and temporal reference points for the notation system can appear only step-by-step during the process of creating the S-TS. No extemal force integrates the dancers. The coherence of the group is not given, it must be realized while dancing, as the S-TS, which is then binding on the participants. The emphasis lies on the realization," as it is clear that its structure cannot always be an absolutely new creation.More often it will be provided by tradition. but, unlike painting, it is not just there"; it must be realized new each time.

Maintenance of the Space-Time Structure

As we saw, the common S-TS must be created by the dancers. Once created, it must also be maintained. A created pulse can easily be destroyed or a broken if people lose interest.

To maintain the space structure means to control and continue the established spatial relationship of the dancers to each other. For instance, dancers in a line formation will control the distance between each other by taking shorter or longer steps. By such controlled distance, they guarantee continuity of the line formation. In some dances of the Trobriand lslanders (New Guinea), in which more then twenty participants dance behind one another, the control maneuver of one dancer influences all succeeding dancers in such a way that a correction can be seen as a wavelike movement going through the whole line of dancers.

An extreme example of the maintenance of a space structure has been observed during a courtship dance of the Himba people (South Africa). The women stand in a semicircle, which is opposed to the semicircle of the men. Two children behind the dancing women attract their attention. The women turn toward them and stand, therefore, in a very uncomfortable position (Fig.12). That this position is not a customary one is shown by Kendon, who reported to Deutsch ... that all interactants will not usually hold, for any sustained period of time, a discrepancy in their head-body orientation of more than 45 degrees. Rather, after momentarily adopting such a discrepant headbody orientation, a person will reorient himself so that his relative head-body orientation is less than 45 degrees. This position can and will be held for long periods of time in relative comfort......"10 In our case, the unusual position is maintained over a period of about 16 seconds. I suggest that the women maintain it in order to make the dance space (D) continue to exist. With their body orientation ( ),they maintain the dance space (D), while with their head orientation ( ), they control the children space (Ch).

The maintenance of the time structure (the stabilization of a once-established tempo) can only be investigated if there is something like stability in the tempo behavior of dance groups. Some speak of such underlying stable str-uctures as meter."17 For our purpose, tempo" is more adequate, as it indicates the possibility of increasing. decreasing, or maintaining the speed. The term meter" suggests that the stability is an extrinsic quality. This may be an effect of looking at music not as an experience but as a written text, with bars suggesting that musical time must occur within this outer framework. Music, however, does not have to fit within a metrical context: it creates its own context by establishing what we refer to as a maintained time structure. In dance, the maintenance of such created time is crucial to plan the movements of the body parts.

If one measures the pulse beats of the grasshopper dance (Iko) in the same manner as proposed in Figs. 8 and 9 with the method of deepest flexion of the knee, one sees that the created pulse is very stable (Fig.13). Two quarters correspond to x = 1150 milliseconds (ms) with a standard deviation of s = 60 ms. This is astonishing, if one sees the enormously complex movement patterns that are created in this stable pulse (Figs. 14-16).

A similar stability of time structure was found in our experimental dance groups. This was not expected, since the members were not trained in dancing and did not have strong dance traditions. For instance, in a group of four dancing students, the oscillogram analysis showed a high stability of time structure and a standard deviation from the pulse similar to the one of the grasshopper dance (Fig. 17).

lt seems that stabilization of the dance tempo does not create large problems, particularly if it is compared with the difficulties in maintaining a certain tempo while singing, in a chorus, for instance. The rhythmic movements that swing the body masses and stabile the dance tempo are absent in singing. One could go further and suggest that the inertia of the moving bodies not only enhances the stability of a tempo but actually hinders tempo changes. This argument can be sustained by the following observation: in our dance, if the students were totally free in their tempo choices, an established tempo would normally be maintained. If the students changed their tempo, it was not through a commonly controlled increase or decrease of speed, but rather through a short period of chaotic destabilization, from which a new stable tempo emerged. It is as if each individual accelerates or decelerates the first tempo in his own manner, thereby causing the general acoustic mass (Fig 18a) out of which a synchronization at a new new tempo level occurs (Fig 18b).

One could speak of a trend toward stabilization of tempo levels. There are traditional dances in which sudden changes in tempo levels have become fixed rules, as in the proviously discussed Himbra courtship dance. After a period of allegro (about 480 ms), the speed is almost doubled as soon as a soloist leaves the common circle, which initiates a prestissimo of the dance (about 270ms).

After variably long pause, the group begins again with the allegro tempo,which is always between 460 and 500ms

This rather long description of the phenomenon of tempo levels does not suggest that this process is the only possible way to change tempi. Further experiments with the students showed that we are able to produce constant accelerandi or ritardandi while dancing, and many elaborate dance forms (e.g. a wallaby dance of the Australian Aborigines from the Wik tribes) work in fascinating ways with just this possibillty of constant accelerando or ritardando.

The point here is that in simple dance events there is a tendency to maintain tempo levels. lf tempo changes, it often occurs in such a way, that an established tempo (A) breaks down in a chaotic phase out of which a new stable tempo (B) emerges.

Does the relationship between two or more tempi such as this follow certain rules? Do these tempi correspond to the time proportions of low integer ratios, e.g., 1:2, l:3, 3:4, that have been found in examples of Western classic music14,18 and later in examples of non-Western music?19 We found such time proportions of low integer ratios in some experiments with our student dance groups (Fig.21). The students accelerated in such a way that, after a short phase of destabilization, a second tempo was established in the proportion 3:2 compared to the first tempo. We need, of course, many more examples of such tempo shifts to be able to decide whether the observed proportions indicate parallele to Epstein's findings (see chapter 4) on the temporal relationships in music.

The maintenance of the S-TS guarantees the continuity of the basic dance structure by controlling the spatial relations of the dancing bodies to each other and the temporal framework within which the movements take place. The S-TS must be maintained because of the fluent character of the dance. It is the only stable element in the dynamic process of dancing. The maintenance of the S-TS establishes invariance during all transformations. lt continues the created Gestalt within the dynamic process, thus providing the contextual frame that dictates when the dancers or parts of their bodies must be where.

Variations within the Space-Time Structure

As we demonstrated, maintenance of the S-TS guarantees the continuity of an invariant pattern that organizes the movements of the dancers. To attain this common pattern is fascinating; to maintain it by repeating the movements strengthens the feeling of community and can be exciting, too, especially in trance dances, for example. But one of the most thrilling experiences while dancing is the variation of the S-TS or even allowing it to desintegrate, to create a new one.

Variations of movements represent such alterations within the S-TS. Instead of a space used to go directly from point A to point B in the simplest dance movement pattern, the way from A to B becomes a zone of more elaborate and richer movement possibilities. The framework of invariant points in space and time remain the same, while the empty zones between the fixed points are open to all kinds of inventions.

To illustrate a variation within a given space structure, we shall describe a simple dance of G/wi children (South Africa): One girl crosses the children's playground with typical dance movements and is followed by other girls and boys. They create a line formation and synchronize their movements, thus establishing and maintaining a S-TS. They then begin to move more freely, one boy does not seem happy with his sixth position in the line formation. breaks out, and runs alongside the line (Fig.20). He then jumps in front of the first dancer. The former leading dancer then becomes the second, and the boy who moved in front becomes the new leader (Fig.21). The line formation is maintained, but, with the changed position of this one dancer, all participants are in a new situation. The framework here is the line formation, which represents an absolute schema that can be varied: it provides organized positions without indicating who has to take which position. The variation within the space structure takes place during the time between leaving the old position and reentering the line formation in the new position.

Our example for the variation within the time structure is more complex. It is based on observations of the already discussed grasshopper dance of the !ko-Bushmen. As we have seen, its pulse is very stable (Fig.13). But within this pulse, a highly developed division of time takes place (Fig.22). While the group maintains the rather simple pulse, the individual dancers in the center of the group perform complex variations. They break up the time between two pulse beats in different manners (variations 1-3). Sometimes this changing division of time can go so far that a pulse beat may be left out (variations 4-6). The different structuring of time between two or three pulse beats may be experienced as a subjective elongation or contraction of the same objective span of time. For the active dancer, such possibilities to play with time create the feeling of moving back and forth between the common pulse, breaking away from it and then falling back again into the common time structure.

The grasshopper dance example also demonstrates a possibility (frequently used in Himba and ! ko-Bushmen dancing) to divide the functions of maintainance and variation among the participants. A group of dancers maintains the S-TS and invites one or two participants, either by fixed rules or by stimulation, to perform rich individual variations. After a certain time, the soloists return to the core of the group maintaining the S-TS to allow others to invent variations. The solo phase stresses the possibilities for competition and individual display; the group phase puts emphasis on the community, the binding character of dance. This ambiguity and tension between competition and community is an important topic in Andrew Strathern's conception of aesthetics discussed in chapter l2 of this volume.

The variations need the unchanging background provided by the maintained S-TS. If there is no such continuity, the variations would not be possible. In our example of G/wi children, when a child left the line formation, the dance would end if all participants did the same at the same time, and no one maintained the line. We have observed that the variations get richer as soon as the common S-TS reaches a certain stability. If the variations begin before, this stability has been reached, they risk disintegration of the S-TS, as the points to relate to become vague. In such phases of impending disequilibration, we have observed regression to the most simple movement patterns, which are then performed with formal exaggerations similar to those during the opening phases of dances.

In other words, the elements of maintenance and variation are not independent. The labile equilibrium opens manifold possibilities of relationships between two extreme poles. lf maintenance is too strong, dancing becomes rigid and boring (except, of course, in those situations in which maintenance has a specific sense, as in trance dancing). lf the variations are overemphasized, the movements become totally individualized. Although the latter may be very expressive, the sense of community, which is essential for group dancing, is lost. Other possible relationships between stabilization and variation within a group could be analyzed.20

Research on this relationship between maintenance and variation within the body of just one individual dancer could reveal interesting phenomena. lt may be possible that some body parts maintain simple movement patterns, allowing others to perforrn rich variations. The temporal development could be similar to the one in group dancing. Variations are only possible on the foundation of a strong invariance. The typical movement patterns of Black African dances are summarized by Dauer21as: relaxation, isolation, polycentricity, and multiplication. Based on our analysis, one should now ask how these processes develop temporally within the body of one dancer. We suggest that isolations of movements are only possible after the creation of a strong basic movement to which the isolations are related.

Variation normally takes place within a given S-TS. Sometimes, however, a new structure can be heard or seen within the variations of the old structure. lf these emerging new rhythms or space formations are emphasized by some of the participants, a struggle between the two structures begins. In this labile phase, the old S-TS may break down totally, so that the new structure is heard and seen clearly and may then be maintained by all the participants, who provide the basis for new variations.

Dance as Aesthetic Behavior

We have analized simple dance forms from the perspective of our approach - aesthetics as behavior. Dance has been viewed as a fluent art form that is regulated by the dancers while it is being developed. An important question remains: does this perspective provide some insight for understanding aesthetics?

This view of aesthetics stresses other factors than those elaborated through the analysis of literature, painting, sulpture, etc. We will discuss here the two most important ones: the dynamics of aesthetics and the social aspect of aesthetics.

Theodor Adorno22 said that there is no question that art does not begin with works". he then went on to speak of aesthetic behavior." This term is ideal to characterize what happens in dance: perception (aisthesis) and behavior - often thought of as being, the opposite - are strongly linked in dance. The S-TS must be created by the dancers, and this creation can only take place if already created elements (spatial configurations as well as temporal structures) are perceived. This unity of perception and movement was described by Viktor von Weizsäcker as early as 194023, and Rudolf zur Lippe24 expanded on this matter in his anthropological aesthetic."

Aesthetic behavior in dance means perception of the ongoing movements and modification of one's own movements while perceiving those of others. lt seems that we share this behavior with animals, as some birds modify their movements while dancing" in relation to the mate's movements. One of the most beautiful examples is the mating dance among the male birds of paradise, in which they coordinate their movements on the pairing tree. But this behavior does not comprise the creation of a S-TS because there is no continuity that indicates intentionality. A S-TS can only be established if the projection of temporal and spatial structures is possible: the dancer must know in advance exactly when he wants to be where. It is not enough to react to a partner; this will always be too late.

Intentionality is, therefore, crucial. It enables the dancers to plan their movements to be in the right place at the right moment. It is astonishing, how often children, energetically perceiving and moving, try to get into such a STS (Figs.20 and 21), and it is wonderful to see their happiness when they are in phase with the group. After a certain time, the maintained S-TS may become boring, so that varlations within the structure may begin. Sometimes these variations lead to the end of the dance; sornetimes they are the beginning of a transformation phase from which a new S-TS evolves.

The creation of common norms, their maintainance, their variation, and their desintegration are not only stages of dance events, they also correspond to important phases in problem solving25 and creative processes.26 The stages can be detected in both the individual process of creation, as in the development of Picasso's Guernica"27 and in phenomena of art history, such as the construction, variation, and destruction of linear perspective.28,29As dance leaves no traces or constructions, there is nothing left to be destroyed. In dance, something, is or is not. To dance means to transform.30 Destruction is linked to construction - to material objectification that can be destroyed. The density of such objectification on our planet demands objectless" art, through which we can relearn transformation instead of destruction.31

The first essential change caused by the dance model" was to abandon the common supposition that aesthetics predominantly deals with ideas fixed in works of art (objectifications). The triangle - creator, work of art, and receptor - is replaced by the moving subject that is at once creator, vehicle of the message, and receptor. The comprehensive term for this changed attitude is aesthetic behavior. The specificity of this behavior lies in the fact that it is not a simple action or reaction but a permanent play with the self-created rules.

The rules provide a framework within which the dancer can cause shivers by trying new exciting movement combinations. The power lies in his hands.The pure shock does not provoke the aesthetic thrill but the fear that there is no more possibility to handle it. Therefore, the playful character of aesthetic behavior is important. lt permits regulation of thrill, stimulation, and the degree of novelty, which make art experiences so wonderful.

Another common supposition concerning the relationship between aesthetic events and the social context has been amended. The belief still prevails that creation occurs in solitude. From our perspective of group dancing, creation cannot be an act of just one individual- only the consent of the participants leads to a group dance (Figs. 23 and 24). The consent is expressed in the S-TS, which can be considered a unifying norm. Tbe group-binding function" of dance, described by Eibl-Eibesfeldt23, may have one of its roots in this common S-TS. Strong and weak individuals must integrate, if they do not want to be excluded from the dance event. The norm binding the paritcipants does not mean that individual display is suppressed. lt is this very common S-TS that allows and even demands individual variation.33 This is especially true for small groups. The rhythms of very big groups dominate the individual movements without being influenced by their specific dynamics. The rules become laws and lose their playful character. The dance has changed into a military march. Instead of encouraging variation, it dictates uniformity.

Human movement - a very simple form of creation - needs the answering movement of other human beings. Only if these movements are authentic expressions of personalities and not the jerking movements of marionettes, can real communication occur. Only in small groups will individuals be able to learn to trust their own rhythms, dynamics, and movements. Sometimes, society delegates the creation of aesthetic norms and their maintenance to a few specialists. Then the aesthetic domain loses not only wide ranges of individual variations, it also risks being increasingly secluded from everyday life. The public may be confronted with works of art, whose meaning is no longer comprehensible to them. The works of art denote a foreign world.

The masses, who are condemned to passive spectatorship or are forced to perform in a preconceived manner, lose interest in experimenting with their individual possibilities. Instead of revealing their personalities to others and having fun, they become narcissistic, or worse, just boring.

Acknowledgements

This research project was made possible by the Swiss National Foundation. I am especially obliged to Prof. Irenäus Eibl-Eibesfeldt - Research Unit for Human Ethology in the Max-Planck-Society in Seewiesen, BRD - who generously allowed me to use his film documents and to Prof. David Epstein who, as a guest of the Institute of Medical Psychology at the University of Munich, encouraged me to go on with my work. He contributed as well as Joyce Nevil-Olesen and Barbara Herzberger to the editing. Furthermore, I thank Prof. Edwin Wilmsen for his help during the translation, Renate Krell for her photographic collaboration, Dr. Wulf Schiefenhövel and Dieter Heunemann for their information about the social background of the Southern African dancers, Peter Heinecke for his technical help during the experimental dance groups and for the oscillogram production, and Daniel McKee for his computer instructions. Last but not least, I thank the medical students from Munich and the city dancers for their active dancing, in our groups!

References and Notes

1. Nautilus (ed) (1976) Situationistische Internationale. MaD, Hamburg

2. Buytendjik F (1956) AllgemeineTheorie der menschlichen Haltung und Bewegung.

Springer, Berlin

3. Plessner H (1970) Philosophische Anthropologie. S.Fischer, Frankfurt

4. Sartre JP (1943) L'ettre et le néant. Gallimard, Paris

5. Sheets M (1 966) Tle phenomenology of dance. Universitv of Wisconsin Press, Madison

6. Siegfried W (1977) Mensch - Bewegung - Raum. Thesis, University of Zürich

7. Lorenz K (1975) Die Rückseite des Spiegels. Piper, München

8. Lorenz K (1952) Über tanzähnliche Bewegungsweisen bei Tieren. Studium Generale 1:1-11

9. Kendon A (1975) Organization of behavior in face-to-face interaction. Mouton. The Hague

10. Deutsch R (1977) Spatial structurings in everyday face-to-face behavior. Asmer. Orangeburg

11. There are also situations in which the surrounding space influences the dance space. In cosmic dances. the orientations and the movements of the dancers are guided by qualities of cosmic space (e.g., the sun or the moon). In processional dances. a spatial configuration created by the dancers moves through a surrounding space (i.e. the village).

12. Golani 1 (1951) Auf der Suche nach Invarianten in der Motorik. In: Immelmann K (ed) Das Bielefeld-Projekt. Paul Parey, Berlin

13. lf we speak of tempi in metrical terms. it will be expressed throughout as the duration in milliseconds between two beats of the pulse.

14. Epstein D (1 983) Das Erlebnis der Zeit in der Musik. Die Zeit. vol 6. Schriften der Carl Friedrich von Siemens Stiftung, München

15. Sbrzesny H (1976) Die Spiele der !ko Buschleute. Piper, München

16. Selection was limited by the multitude of criteria that had to be met for the planned investigation and by the fact that the films had originally been made to fulfill other purposes. To mention some problems: the films began at the moment when the dance group had already been constituted: people danced out of the frame, dancers stood in front of one another so that individual movements could not be analyzed: etc.

17. Hanna JL (1 979) Movements toward understanding humans through the anthropological study of dance. Current Anthropology 20:313-339

18. Epstein D (1979) Beyond Orpheus: Studies in musical structure. MIT Press, Cambridge

19. Epstein D (1 985) Tempo relations: A cross-cultural study. Music Theory Spectrum 7:34-71

20. It would be interesting to know whether the general principle that a strong background S-TS allows richer variation within this S-TS can also be delegated to one element of the S-TS. Does a strong space structure allow the movements more temporal variation? Does a strong time structure free them from spatial limitations?

21. Dauer AM (1969) Zum Bewegungsverhalten afrikanischer Tänzer. Research Film. 6(6):517-526

22. Adomo TW (1984) Aesthetic theory. Routledge and Kegan Paul. New York

23. Weizsäcker V von (1940) Der Gestaltkreis. Thieme. Leipzig

24. Lippe R zur (1987) Sinnenbewußtsein. Rowohlt. Hamburg

25. Rohr A.R (1975) Kreative Prozesse und Methoden der Problemlösung. Belz, Weinheim

26. Kubie LS (1965) Neurotische Deformation des schöpferischen Prozesses. Rowohlt. Hamburg

27. Arnheim R (1962) Picasso's Guernica". University of California Press, Berkely

28. Francastel P (1951) Peinture et Société. Audin. Lyon

29. Novotny F (1 938) Cézanne und das Ende der wissenschaftlichen Perspektive.

A. Schroll, Wien

30. Siegfried W (1987) Danse, Dessin, Destruction. Cahiers du Centre de Recherche Imaginaire et Création. Université de Chambéry 3:122-141

31. Siegfried W (1988) Stadttanz, Übungen zur Ganzheit. Poiesis 4:92-97

32. Eibl-Eibesfeldt I (1972) Die !ko-Buschmanngesellschaft. Piper, München

33. This relationship between the group and the individual seems to be a fertile field for research. Polly Wiessner (1984) investigates this subject in her study: Reconsidering the behavioral basis for style: A case study among the Kalahari-San. Journal of Anthropological Archeology 3:190-234