My approach to orchestration is traditional, in that while I don’t think of it as merely arranging pre-existing material, I also don’t think of color, of sheer orchestral sound, as primary. Most of the standard repertoire—even The Rite of Spring, in spite of its astounding use of orchestral color—sounds good on the piano, and this kind of orchestration can to some extent be considered separately from the music. Up until the second half of the twentieth century, when color became the main focus for many composers, it was only the rare piece, such as the third of Schoenberg‘s Five Pieces for
Orchestra, op .16 or the beginning of Das Rheingold in which orchestration and composition were totally inseparable and a piano rendition therefore meaningless. Still, orchestration should always be an integral part of a composition, not just something applied to already written music.
The currently popular approach to orchestration calls for as much color as possible, as many different instruments as possible, often as much noise as possible, and above all as much percussion as possible. I hasten to add that occasionally this can lead to remarkable results, such as Thomas Ades‘s powerful Asyla. Nevertheless, I prefer a different approach, believing like Robert Frost that “In art a little bit of anything goes a long way.”
(This applies especially to harp and percussion.) And in writing my concertos for violin, viola, and cello, more important for me than orchestral color was keeping the orchestra out of the soloist’s way, in order to avoid balance problems.
Certain books have influenced my thinking about instrumentation. From Cecil Forsyth‘s Orchestration, I learned the importance of choosing instruments based on their special emotional quality, easier to sense than to explain, while Adam Carse‘s The History of Orchestration impressed upon me the importance of pure orchestral colors. (On the other hand, the opening of the Prelude to the First Act of Parsifal and many parts of La Mer make me green with envy at their imaginative use of blended colors.) A wonderful short book of conversations with Sibelius introduced me to the idea that the one thing the orchestra lacks is the piano’s pedal, so that the pedal has to be composed into the orchestra. I thought of this a lot while orchestrating the beginning of my Symphony No. 1, in which low chords are held by certain instruments while being repeated by others, and the beginning of the slow movement of my Violin Concerto, in which the double basses hold harmonics on top of which the other voices move.
When I told Yehudi Wyner how much I disliked Schoenberg‘s garish orchestration of Brahms‘s Piano Quartet in g-minor, which makes a great work sound trashy, he pointed out that this undermines the common criticism of Brahms’s own writing for orchestra as dull and colorless. My favorite orchestrators are Debussy, Stravinsky, Sibelius, Mahler, and perhaps Tchaikovsky, all of whom had an extraordinary sense for both color and clarity. Nevertheless, whatever the occasional faults in Beethoven, for me his is the greatest orchestral music, and listening to it, one hardly thinks about orchestration at all.