Experiments in Musical Intelligence began in the early 1980’s as David Cope’s attempt to create algorithmically new instances of his own music. The creative model proposed with recombinancy processes in EMI suggests that fresh thoughts and invention are culturally inherited. As Cope writes in (Cope, 1996):
Earlier examples of formalized recombinancy date back to the eighteenth-century with Musikalisches Würfelspiele (musical dice games). More generally, re-ordering and re- contextualization of previously existing music material can be found in innumerable examples of musical allusions in the Western music literature. In their works composers, consciously or unconsciously, “borrow” from previously existing works, establishing a network of musical pointers within the culture that created them (Cope, 2003a). EMI “composes” by attempting to make Musikalisches Würfelspiel out of a database of music that originally was not designed as such. The merit of EMI’s success lies in SPEAC (EMI’s analysis system), the pattern matching component and the implementation of an augmented transition network (ATN) responsible for the reconstruction process. SPEAC analyses the function of note groupings in a work based on levels of tension between intervals, metrical placement, and agogic emphasis. Note groupings have their function labeled with an identifier such as statement, preparation, extension, antecedent, and consequent. The identification of a note grouping’s function allows its logical re-constructions according to an ATN, which EMI is its first music implementation.
The pattern-matching component ensures the protection of musical signatures. Musical signatures are patterns occurring in different works from the same composer constituting essential elements that define and help and to recognize a composer’s style.
Since typically recombinancy is most successful when operated on a beat-to-beat level, pattern-matching in EMI protects musical signatures (patterns larger than one beat) from disintegration during recombinancy.
EMI is a software program that, although not intelligent, has produced aesthetically convincing new music. Intelligence seeks survival by the exercise of power over a surrounding environment. In composition, intelligence equals decision making.
Every composition results from the selection of a finite set of constraints to operate on selected materials; even the most intuitive decision remains itself a decision, and consequently, a product of constraints.
Much of the fear and uneasiness surrounding EMI’s acceptance and appreciation, lies on the unwillingness to understand it’s purpose and design. EMI is a composer’s tool. It can be said to “compose”, not because it has the intelligence, the motive or the free will to do so, but because it performs the functions David Cope, the composer, programmed.
“The genius of great composers, I believe, lies not in inventing previously
unimagined music but in their ability to effectively reorder and refine what already exists.”
unimagined music but in their ability to effectively reorder and refine what already exists.”
Earlier examples of formalized recombinancy date back to the eighteenth-century with Musikalisches Würfelspiele (musical dice games). More generally, re-ordering and re- contextualization of previously existing music material can be found in innumerable examples of musical allusions in the Western music literature. In their works composers, consciously or unconsciously, “borrow” from previously existing works, establishing a network of musical pointers within the culture that created them (Cope, 2003a). EMI “composes” by attempting to make Musikalisches Würfelspiel out of a database of music that originally was not designed as such. The merit of EMI’s success lies in SPEAC (EMI’s analysis system), the pattern matching component and the implementation of an augmented transition network (ATN) responsible for the reconstruction process. SPEAC analyses the function of note groupings in a work based on levels of tension between intervals, metrical placement, and agogic emphasis. Note groupings have their function labeled with an identifier such as statement, preparation, extension, antecedent, and consequent. The identification of a note grouping’s function allows its logical re-constructions according to an ATN, which EMI is its first music implementation.
The pattern-matching component ensures the protection of musical signatures. Musical signatures are patterns occurring in different works from the same composer constituting essential elements that define and help and to recognize a composer’s style.
Since typically recombinancy is most successful when operated on a beat-to-beat level, pattern-matching in EMI protects musical signatures (patterns larger than one beat) from disintegration during recombinancy.
EMI is a software program that, although not intelligent, has produced aesthetically convincing new music. Intelligence seeks survival by the exercise of power over a surrounding environment. In composition, intelligence equals decision making.
Every composition results from the selection of a finite set of constraints to operate on selected materials; even the most intuitive decision remains itself a decision, and consequently, a product of constraints.
Much of the fear and uneasiness surrounding EMI’s acceptance and appreciation, lies on the unwillingness to understand it’s purpose and design. EMI is a composer’s tool. It can be said to “compose”, not because it has the intelligence, the motive or the free will to do so, but because it performs the functions David Cope, the composer, programmed.
