History of Electroacoustic Music

The GROOVE System

While Risset was conducting his research at Bell Labs, other music researchers there were interested in developing more responsive systems. At issue was the matter of continuous control.

The Music N software series was completely unresponsive, suffering from the "stupid computer-stupid musician" syndrome. This describes the fact that notated music serves only a rough blueprint for a trained musician, who embellishes it with gestures picked up over years of training. A huge number of small adjustments in finger or mouth position bring variation into the sound. Most musicians aren't even aware that they're doing it. It's intuitive, generally gained by years of imitating a teacher, and gradually developing a personal style.

When the computer is made into a performer, it has no such intuition or training. Computers are nothing more than fast idiots, and all they can do is what they're programmed to do. The Music V system represented music as a series of note events. As a result, computer music, while often interesting, tended to lack the expressiveness of acoustic music. Music is not made up of discrete note events, but rather by processes that unfold over time. A note might be played, but the note is subject to many variations during its duration. These variations are not well represented by a single note event description, as is found with printed music or with the score file of a Music V piece.

In 1967, F. Richard Moore, who had just graduated from Carnegie Mellon with BFAs in music composition and performance, began to work with Max Mathews at Bell. After they completed Music V, they began efforts to inject a higher degree of expressivity into computer music. They created a digital-analog hybrid system, in which a performer played an analog synthesizer, and a computer monitored the performer's manipulations of a controller interface (turning knobs, pressing keys, and so on) and stored them. It was called GROOVE (Generated Realtime Operations On Voltage-controlled Equipment). It ran on a Honeywell DDP224 computer that Bell had acquired specifically for sound research. There were 14 control lines between the computer and the analog synthesizer that were monitored every one hundredth of a second. A special control panel allowed synthesizer interconnections to be changed quickly, on the fly. It was also possible to create libraries of programming routines, so that users could create their own, customized logic patterns for automation or composition.

Composer Emmanuel Ghent (1925-2003) began work at Bell Labs in 1967. His work earlier in the 1960s had required him to build specialized machinery to represent his complex tempo changes and cue the musicians. In 1964, he created a punched paper tape reader by which rhythms could be represented by holes arranged to represent any rhythmic relationship. In 1967, he approached Max Mathews with his work, hoping to move a step further by working with the computers at Bell. At that time, GROOVE was just being put together, and Ghent worked with Moore to develop it. Ghent was composing for the the Mimi Garrard Dance Company, and he was exploring ways of synchronizing lighting and music. His Phosphones was the most famous piece to be created with the GROOVE system. One set of instructions controlled audio playback, and a complementary set controlled lighting dimmers. The timbres of the percussive sounds, and their rhythms are constantly shifting. It recalls Gann's comments about the use of the computer as an instrument to create long-term, gradual changes in sound qualities.

In 1973 Laurie Spiegel joined Bell Labs to work with the system. She and Ghent created a number of works on it.

In 1976 Mathews, in pursuit in what he called the "Intelligent Machine," used GROOVE to create a program called Conductor. Pierre Boulez, then musical director of the New York Philharmonic, wanted an improved level of expressivity over music with tape. The problem with music written for tape and performer, as heard in pieces by Babbitt and Davidovsky, is that the tape is unrelenting and the performers have to tailor their performances to it. Boulez wanted to be able to conduct a tape somehow. The Conductor program allowed articulation and expressive elements to be added to each event in a score file. In performance, these elements could be controlled by external devices, such as a joystick, a keyboard with tappable keys, or a mechanical drum.

In 1978, Bell announced that the computer used for GROOVE was to be removed from service. This ended work with the GROOVE system. But the stage was set to explore the implementation of expressive elements that were based on performance gestures into the performance of computer music. The goal from this time forward for many computer musicians was to implement an improvisatory element into computer-based works, creating logical algorithmic processes that emulated the processes of jazz improvisers.

Joel Chadabe, Electric Sound: The Past and Promise of Electronic Music, Prentice Hall, 1997.

F. Richard Moore, Elements of Computer Music, PTR Prentice Hall, 1990.