Over the last five years or so, I've been trying to realize a dream: I want to engineer musical serendipity and make it the basis of my work as a composer. Rather than writing a score in which every detail of a piece is clearly indicated (and temporally fixed) in notation, I have been collecting materials I like-melodies, harmonies, texts, sound files-and exploring various ways of bringing them together spontaneously at the moment of performance, ensuring that no two realizations will be the same. This sort of process is typically called "generative," and it has a long, if slightly obscure, history in 20th century experimental art.
In divine the rest I try to generate serendipity through the interweaving of human expression and computer operations. NOW Ensemble plays minimally notated pitch material-single notes, dyads, chords, but no rhythms-according to a variety of strategies I describe in prose instructions. Their playing responds to the ever-changing sonic environments the computer creates through the manipulation of field recording samples, drones, fleeting sine-tone melodies, and fragments of spoken text. The texts themselves are generative: I recorded hundreds of nouns, verbs, adjectives, and adverbs (most culled from an essay by Miguel de Unamuno) that the computer assembles into extemporaneous utterances according to certain grammatical rules. The multivalent title divine the rest was itself derived from this vocabulary.
The strange, ambiguous texts combine with the electronic material and NOW's sensitive performance to construct a soundworld seemingly governed by the peculiar logic of dreams. With its indifference to motivic development and narrative arc, one moment in divine the rest melts into another like so many episodes during REM sleep. The experience of listening to this music will be, I hope, something like remembering music heard in a dream.