Training Camp (2016) is intended for dual screen installation and is the first of three works that use found motion capture files that supposedly depict “gay” movements. This data is plugged into an army of preset Roman soldier characters from crowd simulation plugin Golaem, and the result is a kind of call-and-response exercise routine between a single Roman soldier and a battalion. The work attempts to understand how software, it’s infrastructures and industries might facilitate, appropriate, encode or propagate gender behaviours, biases and stereotypes.
The “gay” mocap files were found within the database of the TrueBones library, and sit alongside thousands of other searchable entries (examples drawn at random include “sulkywalk,” “pullrope,” “gunfighter,” “sit_at_desk,” and “goosestep”).
Although mocap seems like a new technology, it’s as old as the computer itself. In 1964 at New York’s World Fair, Disney premiered their “Waldo Suit” which was the first motion registration system to affix tracking markers to a dancer’s body. Since the 1980s, there has been a steady stream of proprietary mocap systems claiming to offer more fidelity and flexibility with their cost effective, compact and intelligent systems. Yet mocap has still not taken off in the way early adopters imagined. Motion capture files often include a lot of noise that’s hard to clean up, take a long time to manipulate and are not always compatible with traditional keyframe animation.
In this kind of climate, many animators, game artists and VR developers use motions off the shelf, and therefore propagate an index of motions that may date back five, ten or twenty years. With no standardised archival procedure, it’s difficult or impossible to track the origins, provenance and ownership of these data sets, many of which will have had close ties to certain academic institutions, proprietary (and obsolete) technologies or long bankrupt or bought-out companies. The material history of mocap data is cloudy: we don’t know who created these motions, when they were made and for what purpose. Was the dancer simply obeying orders? Did they improvise? Who cleaned up the data? Who paid their salaries, or were they even paid at all?
This foggy archive is not a unique phenomenon: motion capture libraries are just one (acutely visual) example of the contested and controversial site of the archive in the early stages of our 21st century data revolution.