Emergent Patterns in Dance Improvisation and Choreography

In a traditional choreography a choreographer determines the motions of a dancer or a group of dancers. Information theory shows that there is a limit to the complexity that can be created in any given amount of time. This is true even when building on previous work, since movements and their interactions have to be communicated to the dancers. When creating a group work, choreographers circumvent this problem by focusing either on the movements of individual dancers (giving rise to intricate movements but within a simple spatiotemporal organization) or on the overall structure (intricate patterns but simple movements) or by creating room for the dancers to fill in part of the movements. Complexity theory offers a different paradigm towards the generation of enticing patterns. Flocks of birds or schools of fish for instance are considered 'beautiful' but lack a central governing agent. Computer simulations show that a few simple rules can give rise to the emergence of the kind of patterns seen in flocks or swarms. In these models individual agents are represented by dots or equivalent shapes. To be of use to choreography and to be implemented on or rather with dancers, some additional rules will therefore have to be introduced. A number of possible rules are presented, which were extracted from 'real life' experiments with dancers. The current framework for modeling flocking behavior, based on local interactions between single agents, will be extended to include more general forms of interaction. Dancers may for instance perceive the global structure they form, e.g. a line or a cluster, and then put that knowledge to creative use according to some pre-established rules, e.g. if there is a line, form a circle or if there is a cluster spread out in all directions. Some of these rules may be applied back to other complex systems. The present paper is also an invitation to complexity theorists working in different fields to contribute additional rules and ideas.