Live Algorithms for Music

Live algorithms are an ideal concept: computational systems able to collaborate proactively with humans in the creation of group-based improvised music. The challenge is to achieve equivalence between human and computer collaborators, both in formal terms and in practice (evident to both performers and audience alike). The fundamental question is the capacity for computational processes to exhibit “creativity.” The problems inherent in computer music performance are considered, in which computers are quasi-instruments or act in proxy for another musician. Theories from social psychology and pragmatics are explored to help understand live music-making as a special case of social organization; namely, Kelley’s covariation model of Attribution Theory and Grice’s Maxims of Cooperation. This chapter outlines a description of how human beings and computers might engage on an equivalent basis and proposes how social psychology theories, rendered in formal language, can point to new horizons in human-computer performance practice.