In September 1978 I moved from the American Midwest to Utrecht, Holland, to study for one year (or so I thought) at the Instituut voor Sonologie. Sonology had two PDP-15 computers, equipped with 12-bit digital-to-analog converters. My previous experience with digital sound synthesis consisted of a brief foray into preparing a stack of punched cards, to be sent far away for conversion, before hearing a one-second “beep” on a tape that came back in the mail. The community of composers in Utrecht, in contrast, had developed an extensive library of software for algorithmic composition and sound synthesis, primarily written in PDP-15 assembly code. Armed with these examples, I dove into the world of machine-language programming.
My most influential mentors were Gottfried Michael Koenig and Paul Berg. Both were concerned with developing programmatic models for the generation of music and sound. Paul’s 2009 article describes the period well – in it he quotes Koenig’s aphorism: “The analytical task—given the music, find the rules—is reversed: given the rules, find the music” [Koenig, 1978, p. 10].
I studied Berg’s ASP programs and PILE compiler closely and developed my own library of routines (named SPA in tribute to ASP, and because we enjoyed the drink). The conceit continued to the title of my first full composition using the library: Spattering…a Shower (1979). (Note the appearance of SPA at the beginning). The other reference in the title was to the work of Louis-Ferdinand Céline (aficionados will recognize the homage).
The work was designed to describe a class of compositions, in keeping with the model formation philosophy of Sonologists at the time. That is, each run of the software output a noticeably different version of the piece. Sound was generated during the execution of the program (in real time) and captured on tape. No subsequent editing or other processing was performed. I recall that it was procedurally palindromic in the middle section – a succession of processing subroutines was added and then removed in the reverse order.
The 12-bit converters we used and the fluctuating sampling rate at which data was fed to them (determined by the amount of time it took to produce the next sample) produced a raw, wildly digital sound. That grit was quite antithetical to the lavishly produced sound objects of the non-real-time synthesis aesthetic popular at the time (it was termed “non-standard”). We liked it. I still like it. It is with some amusement that I notice a similar sound quality now being produced on purpose, rather than of necessity. You can hear an excerpt of Spattering…a Shower in the right sidebar.
Berg, P. (2009). Composing sound structures with rules. Contemporary Music Review, 28(1), 75-87.
Koenig, G. M. (1978). Composition processes. In UNESCO Computer Music: Report on an International Project Including the International Workshop Held at Aarhus, Denmark in (pp. 105-126).