History of Electroacoustic Music
The Birth of Computer Music
The Illiac Suite
Lejaren Hiller (1924-1994) was a chemistry professor at the University of Illinois. While getting his degrees in chemistry at Princeton, he studied composition with Milton Babbitt and Roger Sessions, both of whom encouraged him to make a career in music. In 1955, when he was at Urbana as a post-doctoral student, he and Leonard Isaacson, a graduate assistant, were computing polymer configurations. By using the same techniques, they programmed the university's Illiac computer to write a musical composition. They relied on probabilities taken from information theory. The rules of counterpoint were programmed, and musical events were chosen by generating random numbers. Pitches were tested according to the rules of classical counterpoint. If a pitch didn't fit, another note was generated. This is a generate and test method of creating material with a computer. Random numbers also determined durations and instrument. Probability was also employed with a Markov chain, which is a formula by which the past determines the probabilities of the future. For example, if a selected pitch is C, then the next available pitches may be G (65%), E (15%), F (10%), or A (5%). A longer chain might also take into account not only the present note, but notes from the past as well. This type of programming is now known as algorithmic composition, referring to a composer setting up processes and letting a computer determine the specifics based on them.
The first computer piece was completed in 1957, the Illiac Suite for String Quartet. It was written by the Illiac computer, then notated manually and played by human musicians. Hiller described the piece as a laboratory notebook of experiments in algorithmic composition carried out during the years 1955-57. The piece was divided into four movements, each of which represented a particular musical experiment. The first was based on simple counterpoint. The second began with random notes, with rules gradually being applied througout the movement that imposed order, resulting in correct counterpoint by the movement's end. The third movement explored incorporating varying rhythms and string playing techniques, with pitches first being the same for each instrument while rhythms and techniques were highlighted, then were chosen at random, then from simple compositional rules, then according to twelve-tone serial rules. The fourth movement was based on probability experiments such as Markov chains.
Hiller described the piece in an article in Scientific American, causing a furor in the musical establishment. Neither the New Grove Dictionary of Music and Musicians nor Baker's Biographical Dictionary of Musicians carried entries on Hiller until after his death, even though his compositions (even those not created by computer -- quite a few) were performed worldwide.
In 1958, Hiller established the Electronic Music Studio, making Illinois second to Columbia in starting electronic music experiments in a university, and the first to conduct musical research with a computer.