Don Herbison-Evans ,
Technical Report 422 (1991), updated 22 October 2012, 27 May 2017
Basser Department of Computer Science (now School of Information Technologies)
University of Sydney, Australia
This paper describes the various uses of computers in Dance. Some important non-graphics uses are described: in administration, lighting control, and competition scrutineering. Graphic uses described are: notation, choreography, teaching, and performance. Perhaps the most important use, however, will be in gaining an understanding of human movement, possibly using the paradigm of formal grammars.
The Arts in general, and Dance especially, are regarded by much of the computer community as frivolous and unworthy of significant investment in computing resources and effort. This, coupled with the fact that most Dancers sacrifice literacy and numeracy studies in order to devote adequate attention to their Dance training, have perhaps between them led currently to a somewhat limited involvement in computing by the Dance community. Nevertheless, the work that has been done so far indicates a considerable potential for future development. The most significant impact so far as been indirect and behind the scenes.
Among the larger Dance schools and companies, administration is now largely done on personal computers. These are used with standard office software for address lists, correspondence, budgeting, accounting and publicity (Bromelkamp, 1981). Typically the only graphics involved are drop down menus (which normally have verbal entries).
An even more widespread use of computers behind the scenes in Dance is for the control of stage lighting. A typical evening Dance performance will involve perhaps 100 or more powerful lamps lit variously over 100 or more cues. With the cues being totally non-verbal, the use of computer control of the lighting is a great benefit. Nevertheless, although the output might be described as graphic, the input to such systems still largely follows the pre-computer lighting traditions using numeric coding (Bentham, 1980).
Another successful use of computers behind the scenes has been with the scrutineering of Dance competition results. In Dance competitions typically several judges place the contestants in rank order. The evaluation of these results to obtain an overall placing for each Dance is an intricate procedure. In Australia, the Australian Dancing Society has been at the forefront of performing this scrutineering by computer, particularly for the annual Australian Ballroom Dancing Championships (Williams, 1991). This use of computers in Dance is, however, far from graphic.
A genuine use of Computer Graphics behind the scenes has been assistance for Dance notators (Hutchinson Guest, 1984). Many methods exist for writing on paper a description of a Dance, and the most general of these typically use a very complex lexicon of symbols, the detailed orthography and positioning of which on a score are significant. Several programs now exist for the two major notations (Laban and Benesh) which allow the user to create, store, modify and print scores in these notations, e.g. (Marcovici and Ryman, 1987), (Sealey, 1982). There has also been work on translating notation automatically into the movement of animated figures (Herbison-Evans, 1986) and the computer translation between notations (Politis, 1989). The more complex inverse problem of converting images of moving figures to notation has received some attention (O'Bourke and Badler, 1980). This problem is likely to be very resistant to computer assistance, given that trained human notators typically need 500 times as much time to notate a Dance as it takes to perform it (Braban, 1990). The problem is also beset by the constituent/contingent paradox: how does one distinguish what was meant to be performed from what was actually performed (Williams and Farnell, 1990).
The first significant proposal for, and demonstration of Computer Graphics in Dance was by Michael Noll in 1967 (Noll, 1967). He envisaged that choreographers might beneficially use computer generated images of figures dancing. Subsequently, suggestions have been made of the use of such figures in Dance teaching, and such figures have also been used in Dance performance.
As a choreographer's tool, the computer can be used in the same manner as a piano might be used by an orchestral composer. However, choreographers have been reluctant to make a break with tradition, preferring the older method of working with live Dancers and typically depending on them for some inspiration (Hutchinson, 1967). The only significant use so far has been by Merce Cunningham for his work `Trackers' (Dunning, 1991). This was partly choreographed using a computer system called `Life Forms', devised by Tom Calvert of Simon Fraser University, Vancouver. This allows the user to manipulate figures on a screen by using dials. Several figures can be shown simultaneously (in different colours), and their positions and movements viewed from an interactively variable perspective.
Computed dancing figures have also been proposed as an aid in teaching Dance. For example, the computer could be used to show idealised movements slowly of fast steps, that are impossible to demonstrate slowly because of problems with balance or momentum. Computers could also be used as a teaching aid for students to clarify for themselves steps with complex alternatives, such as the 16 forms of Pas de Bourree in ballet (devant, derriere, dessus, and dessous, performed starting with left or right foot, with the starting foot in fifth position in front or behind).
One of the few reported uses of computers in Dance teaching was the use of the NUDES computer animation system by a class of Dance students, each producing their own choreographic sequence on a computer (Herbison-Evans, 1982). This activity was reported as having beneficial effects for the students mainly in their discipline of having to specify to the computer exactly what movements they wanted.
One use of computer graphics in a Dance performance has been for the generation of backdrops. Substantial use of this idea was made by the San Francisco ballet in the Dance `Pixellage' (Crow and Csuri, 1985). The backdrops were designed on an Aurora 100 videographics workstation and recorded with a LogE/Dunn 35 mm film camera, together with a metronome track for synchronizing the orchestra. The results were very sophisticated, with the backdrops seeming to provide all the props, such as a ball which appears to be thrown from Dancer to Dancer.
Another use involving a computed figure was in the ballet `The Catherine Wheel', choreographed by Twyla Tharp (Gruen, 1983). This used the stylised nature of the computed figure to advantage by using it as the spiritual figure of Saint Catherine. The figure was transparent and partly triangulated, moving by cubic spline interpolation between points taken from measurements of images of a live human model dancing. Saint Catherine, in the ballet is projected at life size on the backdrop, the cubic movements giving it a peculiar unnatural quality.
An even more sophisticated use of a computed figure in performance was in the closing ceremony of the 1983 Australian Computer Conference ACS10. This included a two minute Pas de Deux between a live Dancer and a computer figure: `Digital Duet' (Herbison-Evans, 1989). The figure was programmed in the NUDES animation language, and a film made of the computer display. This was then back-projected onto a screen behind the live Dancer. Synchronisation of the movements was obtained by a skilled pianist playing to the movements of the computed figure, and then the live Dancer synchronizing her movements to the piano.
Various representations of human figures have been used in these studies (Badler et al, 1979). They are all very stylised. The most realistic proposals simulates changes in shape of parts of the body as muscles contract and extend, by using a three layer model of skeleton, muscle and skin (Anderson, 1990).
However, there is a problem in that any representation more complex than a visually ambiguous stick figure requires computing devices more expensive than those used in, say, Dance administration. As mentioned in the introduction, most Dance personnel lack a sympathetic understanding of computers, and also most Dance companies and schools operate financially close to the margin, so the required investment for any system showing realistic human figures is seldom available.
Another aspect of some Computer and Dance studies which has had a generally negative impact has been the attempts to generate choreography by computer without human intervention. These studies typically have used random number generators to produce random movements and positions (Lansdowne, 1978). The results have proved interesting challenges to perform, but have about the same aesthetic content as clouds in the sky: i.e. they are `Art Trouve' rather than `Art Compose'. Any propagation of this aspect of Dance and Computers may be viewed as an abnegation of the artistic responsibilities of the so-called choreographer who does it.
To normal moving people, i.e. those with little specialised training in such things as Dance or gymnastics, the idea that there is a need to understand movement may seem obscure. This attitude is possibly a hangover from Plato, who propagated the idea that the body (and its movement) is to be dismissed intellectually, and what is of greater value is the study of the mind. However, the consideration of the sign languages used by aurally impaired people will show that movement is at least as complex as speech. One might also consider that it takes most people twice as long to discover how to jump their own length as it takes to learn to understand a suggestion that they do so.
The study of movement is being considerably enhanced at present by using the paradigm of the formal computer language. Recent studies have taken small subsets of movements from the Foxtrot (Herbison-Evans, 1989) and from Classical Ballet (Hall, 1990) and shown how these Dances can be recognised by formal context-free grammars. These studies suggest that in the long process of human physical, intellectual and social evolution, the development of mental processes recognising the language of Dance may have preceded and been used as a model for the development of speech.
One of the most sophisticated computer graphic results to date in Dance has been this author's effort to synthesise Ballroom Dancing. The coordination of the movements of two figures dancing together with a constant five point contact between the bodies (man's left hand, right hand, right forearm, right upper arm, and diaphragm) as they perform locomotive and turning movements is a considerable challenge for a computer system. So far, the movements might be described as `Silver Medal' standard, as can be seen in the author's animation of Waltz `Reverse Turns', using the NUDES animation system. There is a long way to go before we understand movement well enough to synthesise computer graphic dancing figures with any aesthetic value.
As computers become more powerful and cheaper, more sophisticated computer systems will gradually come within the reach of even impoverished Dance institutions, and the applications described above will perhaps become commonplace. Meanwhile, research continues in many academic institutions on further aspects of Computers and Dance.
`Modelling the Human Body for Computer Animation', M.Sc. Thesis, Monash University, 1990.
Badler, N.I. and Smoliar, S.W.,
`Digital Representations of Human Movement', Computing Surveys, Vol. 11, No. 1 (March 1979) pp. 19-38.
`The Art of Stage Lighting', (2nd ed.), Pitman House (London), 1980, p. 165.
`Score Statistics', The Choreologist, No. 41, (Winter 1990), p. 11.
Bromelkamp, H. (1981),
`All in Order: Information Systems for the Arts', Foundation Centre (New York).
Crow, F. & Csuri, C. (1985),
`Music and Dance Join a Fine Artist and a Paint Machine', IEEE Computer Graphics and Applications, August 1985, pp. 11-13.
Dunning, J. (1991),
`Dance by the Light of the Tube', New York Times, 10 Feb. 1991.
Gruen, J. (1983),
`Dancevision', Dance Magazine, Vol. 57, June 1983, pp. 78-79.
Hall, N.L. and Herbison-Evans, D. (1990),
`BALLONES: A Ballet Animation Language', Proceedings of AUSGRAPH90 (Australian Computer Graphics Association, 1990 Conference).
Herbison-Evans, D., Green, R.D. and Butt, A. (1982),
`Computer Animation with NUDES in Dance and Physical Education', Australian Computer Science Communications, Vol. 4, No. 1, pp. 324-331.
Herbison-Evans, D. (1986),
`Animation of the Human Figure', Dept Computer Science (University of Waterloo), Technical Report CS-86-50.
Herbison-Evans, D. (1989),
`Digital Duet', Basser Dept Computer Science (University of Sydney), Technical Report 338.
Herbison-Evans, D. (1989),
`A Revised Grammar for the Foxtrot', Basser Dept Computer Science (University of Sydney), Technical Report 371.
Hutchinson, A. (1967),
`A Reply', Dance Magazine, Jan. 1967, pp.45-46, 81-82.
Hutchinson-Guest, A. (1984),
`Dance Notation', Dance Horizons (New York), p. 164.
`The Computer and Choreography', IEEE Computer, Vol. 11, No. 8, pp. 19-30.
Marcovici, M.M., and Ryman, R.S. (1987),
`The MacBenesh Manual' Dance Group, University of Waterloo.
Noll, A.M. (1967),
`Choregraphy and Computers', Dance Magazine, Jan. 1967, pp. 43-45.
O'Rourke, J. and Badler, N.I.(1980),
`Model-Based Image Analysis of Human Motion Using Constraint Propagation', IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), Vol. 2, No. 6, pp. 522-536.
Politis, G. (1989),
`A Translator for Human Movement Notation', Basser Dept Computer Science (University of Sydney), Technical Report 350.
Sealey, D. (1982),
`Notate: Computerized Programs for Labanotation', Journal for the Anthropological Study of Human Movement, Vol. 1, No. 2, pp. 70-74.
Williams, D. and Farnell, B.(1990),
`The Laban Script', Australian Institute of Aboriginal and Torres Strait Islander Studies (Canberra), Chapter 7.
Williams, D. (1991),
`Computer Scrutineering', Australian Dancing Society.