Music
The computer has proved itself qualified in a number of fields and professions, but what of the more artistic ones? Not long ago RCA demonstrated an electronic computer as an aid to the musical composer. Based on random probability, this machine is no tongue-in-cheek gadget but has already produced its own compositions based on the style of Stephen Foster. Instead of throwing up their hands in shocked horror, modern composers like Aaron Copland welcome the music “synthesizer” with open arms. Bemoaning only the price of such a computer—about $150,000—Copland looks to the day when the composer will feed in a few rough ideas and have the machine produce a fully orchestrated piece. The orchestration, incidentally, will include sounds no present instruments can produce. “Imagine what will happen when every combination of eighty-eight keys is played,” Copland suggests. Many traditionalists profess to shudder at the thought of a machine producing music, but mathematical compositions are no novelty. Even random music was “composed” by Mozart, whose “A Musical Dice Game” is chance music with a particularly descriptive title, and Dr. John Pierce of Bell Laboratories has extended such work.
Listen:
In 1955, Lejaren A. Hiller, Jr., and L. M. Isaacson began to program the ILLIAC computer at the University of Illinois to compose music. The computer actually published its work, including “Illiac Suite for String Quartet,” Copyright 1957, New Music Editions, done in the style of Palestrina. All music lies somewhere between the complete randomness of, say, the hissing of electrons in vacuum tubes and the orderliness of a sustained tone. No less a master than Stravinsky has called composition “the great technique of selection,” and the computer can be taught to select in about any degree we desire. Hiller describes the process, in which the machine is given fourteen notes representing two octaves of the C-major scale, and restricted to “first-species counterpoint.” By means of this screening technique, the computer “composed” by a trial-and-error procedure that may be analogous to that of the human musician. Each note was examined against the criteria assigned; if it passed, it was stored in memory; if not, another was tried. If after fifty trials no right note was found, the “composition” was abandoned, much as might be done by a human composer who has written himself into a corner, and a new start was made. In an hour of such work, ILLIAC produced several hundred short melodies—a gold mine for a Tin Pan Alley tunesmith! It was then told to produce two-voice counterpoint for the basic melodies. “Illiac Suite” is compared, by its programmers at least, with the modern music of Bartok.
Purists whose sensibilities are offended by the very notion of computer music point out that music is subjective—a means of conveying emotion from the heart of the composer to that of the listener. Be that as it may, the composition itself is objective and can be rigorously analyzed mathematically, before or after the fact. From a technical standpoint there seems to be only one question about this new music—who composed it, the programmer or the computer?
An interesting sidelight to computer music is its use to test the acoustics of as yet unbuilt auditoriums. Bell Telephone Laboratories has devised such a machine in its Acoustical and Visual Research Department. The specifications of the new auditorium are fed into the computer, followed by music recorded on tape. The computer’s output is then this music as it will sound in the new hall. Critical experts listen and decide if the auditorium acoustics are all right, or if some redesign is in order.