Live coding produces multiple overlapping kinds of liveness — machine liveness, embodied human liveness, and relational audience liveness — not one single kind

Liveness in live coding is not a single phenomenon. Steven Tanimoto’s degrees of liveness captures machine liveness — the immediacy of semantic feedback and real-time execution enabled by technology. But human embodied liveness (the vitality, improvisatory risk, and bodily presence of the performer) and relational liveness (the feedback loop between audience and performer) are equally present and distinct. Live coding complicates Philip Auslander’s mediatized/live distinction: it activates a feedback loop between live performer and machine, between live and mediatized. A common error: conflating live coding’s liveness with either just real-time audio (machine liveness only) or just audience presence (forgetting the code’s own temporal agency).

Examples

A TidalCycles performer modifying patterns in response to the audience’s energy, while the pattern itself generates emergent variations the performer didn’t anticipate — three kinds of liveness operating simultaneously: machine (interpreter feedback), human (bodily risk of error), and relational (audience response).

Assessment

Describe a live coding performance moment that exhibits all three kinds of liveness simultaneously. Then explain why hiding the screen would eliminate one type while leaving the others intact.