Summer Research Dispatch: Ritwik Banerji on Improvisation
Each year, the Berkeley Center for New Media is thrilled to offer summer research awards to support our graduates in their cutting edge work. Below, Ritwik Banerji describes how he used the funds to work on his musical improvisation project.
Ultimately, this project deals with the question of just how desirable novel musical material is in the context of improvisation. On the one hand, improvisatory musical practices privilege the production of novel material in the course of performance. On the other, novelty is not at all what improvisers want so much as the correct and consistent of reproduction of stock materials which satisfy the constraints of a given musical practice.
In order to examine the tension between novelty and conventionality, this project focuses in particular on the development of a broad array of semi-improvisatory simulations of the typical jazz “rhythm section” (i.e., piano, bass, and drum set). In typical practice, rhythm section players improvise along with the soloist as the entire ensemble explores the harmonic and rhythmic possibilities of a given composition while also preserving the sonic legibility of the composition’s basic structures (particularly the progression of chords over time.) While rhythm player’s can neither predict the soloist’s actions nor their responses to the soloist, the basic hypothesis of this project is that there is more consistency and conventionality in how rhythm sections engage with the soloist than is typically suggested by the continual framing of these activities as “improvisatory.”
This project poses this hypothesis by creating a set of fixed walking bass lines (i.e., melodic sequences which rather clearly outline the composition’s given harmonic framework). While each line is fixed, the system randomly jumps from one line to another as the sequence plays out. At any given point in a 32-bar form, for example, the system may randomly jump to another line in the collection of precomposed bass lines. This was done in order to investigate just how desirable it really may be for bass player’s to create novel patterns rather than to confine their “improvisations” to a series of collocations.
Funding from BCNM was instrumental in allowing for the completion of a wide variety of such simulated accompaniments and enabling this project to be presented as part of the UC system-wide generativism and music research group’s activities at UC San Diego. While more testing of this system with actively performing jazz musicians is needed, collaboration with scholars at UCSD and Berkeley further established the viability of the main hypothesis that much of improvisatory jazz accompaniment relies on collocations rather than spontaneous composition, as is often suggested.