BOOKSHELFSPEAKER
April 2023
pen on red rosin paper
Logic, Premiere
BOOKSHELFSPEAKER is a composition for piano and voice, generated from the books on one of my bookshelves. It explores the temporal experience of books, the unique timings and inflections of my speaking voice, and western music theory concepts of harmony and melody.
PERFORMANCE INSTRUCTIONS
The physical score is spatially tempo'd at 1.2 cm/s, with the exception of call-out sections which are tempo'd at 12 cm/s. Measure lengths can thus be retrieved by measuring the physical widths of the score. Right-hand notes within measures are of equal length, and left-hand chords are played at the beginning of each measure.
Vocal performers may choose to perform digitally, in which case they shall record themselves speaking the book titles and author names and then time-warp each syllable to the correct length WITHOUT warping pitch. Or, they may choose to perform live, in which case they shall speak each syllable to last the correct length of time.
Piano performers are advised perform digitally, in which case they shall input notes at correct lengths into a MIDI editor and then adjust note velocities to achieve their desired dynamic effects. Piano performers may also attempt to perform live, but they may find the faster sequences difficult to play accurately.
COMPOSITION PROCEDURE
The page-length of each vertical book is proportional to the length of its sequence in the composition, which reads at a pace of 45 pages per second. The sequences of horizontal books cover the length of all books below them.
For each book's compositional sequence, the left-hand chords are derived from the author's name and the right-hand melody is derived from the title of the book. These derivations follow a mapping of the 26 letters of the alphabet to 26 keys below middle C for the left hand, and 26 keys above middle C for the right hand.
The author's name is split up into syllables, with each syllable corresponding to a chord in the left hand. The letters in the syllables create the notes of the chords, using the mapping mentioned above. The durations of the chords are derived from durations of the syllables in the audio recording of myself saying the author's name, with the recording's duration stretched to fit the total book-sequence duration.
The melody is similarly constructed, with each letter of the book title corresponding to a single note of the melody. The durations of each syllable-melody is also derived from the audio recording of myself saying the title, stretched to fit the total book-sequence duration. The letter-notes in each syllable have equal duration within the syllable-measure.
