When a composer or songwriter finishes a new piece of music, typically the result is a fixed and defined work that can be reproduced consistently. Classically, a set of score dictates what notes to play, when to play them, for how long, how loudly, and so on. Even in jazz, a quintessentially improvisational form of performance, a fixed melody known as the “head” is played (usually at the start and end of the piece), and a lead sheet specifies the chord changes through which the music cycles. Recordings capture performances and multitrack constructions with excellent fidelity. In all these cases, there is some degree of chance, relying on the skills and inclinations of the performers or the capabilities and settings of the equipment involved, as well as the acoustic properties of the listening environment; however, some composers incorporate chance into the very structure of the music itself.
At least as far back as the late 15th Century, games existed in which short fragments of music were selected and arranged according to the results of rolling dice. A manuscript written by Wolfgang Amadeus Mozart in 1787 consists of several two-bar segments of score with various marks indicating some system was used to piece them together, and one example is even included, though the precise rules were not recorded (or have not survived) and attempts to deduce them remain inconclusive. Several other varieties of "Musikalisches Würfelspiele" (musical dice games) were credited by their publishers to Mozart, though nearly all of these are almost certainly inauthentic; one such game was produced by Nikolaus Simrock in Berlin, who was Mozart's own publisher, though its attribution to Mozart has never been confirmed.
In the early 20th Century, Marcel Duchamp used chance in the composition of two musical pieces, while Charles Ives and Henry Cowell produced pieces that required performers to arrange fragmentary music. Soon, many other composers began experimenting with chance in various ways, including Karlheinz Stockhausen, Pierre Boulez, and John Cage. “Water Walk,” a piece Cage performed on the show I've Got A Secret in January of 1960 had to be modified for television, as the composition originally required five radios turned on and off at specific instances, incorporating by chance whatever might be on the air at the time of performance, but a dispute between two labor unions on the set of the show meant the radios could not be used; as a solution, Cage instead opted to smack the top of each radio at the designated moments to signify turning them on, and to signify turning them off, he knocked them to the floor.
Developments in computing technology allow for the production of highly variable music. For example, Brian Eno's 1996 CD-ROM album Generative Music 1 is a collection of twelve compositions where each conforms to its own complex parameters but never plays out quite the same way twice. More recently, Eno also composed the elements used in the procedural music system for the videogame Spore, yielding a soundtrack that never repeats. Many programs for creating different varieties of chance music are available, both commercially and as free and open source software, with some offering the user little control while others provide extensive options.
Though it may seem easier for a composer to relinquish decision-making to chance, this is not usually the case; in fact, it's often far simpler to write a traditional fixed piece than to devise a system of parameters within which the results are bound to produce the sort of effect a composer wishes to convey. Not only does the incorporation of chance allow a musician to emphasize the broader shape of a piece rather than place too much significance in the specifics of one possible implementation of an idea, it can also offer the listener a fresher experience each time through.
I've conducted several experiments with different types of chance music, and I'd like to offer a brief example. “Inganok” is one selected result of a C++ program I wrote in 2008 to produce highly variable Csound scores, a technique I've since been revisiting and revising with Python. Of course, this simplistic mood piece reflects more on my own tastes and abilities than on the usefulness of chance music in the hands of more skilled musicians!
At least as far back as the late 15th Century, games existed in which short fragments of music were selected and arranged according to the results of rolling dice. A manuscript written by Wolfgang Amadeus Mozart in 1787 consists of several two-bar segments of score with various marks indicating some system was used to piece them together, and one example is even included, though the precise rules were not recorded (or have not survived) and attempts to deduce them remain inconclusive. Several other varieties of "Musikalisches Würfelspiele" (musical dice games) were credited by their publishers to Mozart, though nearly all of these are almost certainly inauthentic; one such game was produced by Nikolaus Simrock in Berlin, who was Mozart's own publisher, though its attribution to Mozart has never been confirmed.
In the early 20th Century, Marcel Duchamp used chance in the composition of two musical pieces, while Charles Ives and Henry Cowell produced pieces that required performers to arrange fragmentary music. Soon, many other composers began experimenting with chance in various ways, including Karlheinz Stockhausen, Pierre Boulez, and John Cage. “Water Walk,” a piece Cage performed on the show I've Got A Secret in January of 1960 had to be modified for television, as the composition originally required five radios turned on and off at specific instances, incorporating by chance whatever might be on the air at the time of performance, but a dispute between two labor unions on the set of the show meant the radios could not be used; as a solution, Cage instead opted to smack the top of each radio at the designated moments to signify turning them on, and to signify turning them off, he knocked them to the floor.
Though it may seem easier for a composer to relinquish decision-making to chance, this is not usually the case; in fact, it's often far simpler to write a traditional fixed piece than to devise a system of parameters within which the results are bound to produce the sort of effect a composer wishes to convey. Not only does the incorporation of chance allow a musician to emphasize the broader shape of a piece rather than place too much significance in the specifics of one possible implementation of an idea, it can also offer the listener a fresher experience each time through.
I've conducted several experiments with different types of chance music, and I'd like to offer a brief example. “Inganok” is one selected result of a C++ program I wrote in 2008 to produce highly variable Csound scores, a technique I've since been revisiting and revising with Python. Of course, this simplistic mood piece reflects more on my own tastes and abilities than on the usefulness of chance music in the hands of more skilled musicians!
