Revising has long been a common practice among composers. J. S. Bach made changes as a matter of course when making a new copy of a work. Chopin revised habitually, even marking changes on published versions of his compositions.
Several composers whose music I love reworked some of their compositions over periods lasting years. Berlioz, for example, based his Damnation of Faust on a work he had composed 17 years earlier. Tchaikovsky completed the final version of Romeo and Juliet 20 years after the premiere of the first. Stravinsky made revisions for various reasons, including numerous changes and additions of detail in his orchestrations. Mahler revised most of his symphonies extensively.
Such decisions to revise usually stem from a persistent pursuit of eloquence.
After I began writing librettos and operas, I increasingly approached my work with a dramatist’s frame of mind. The more experience I gained working from within a theatrical perspective, the more I wished that I had composed and orchestrated a few of my earlier works with more élan—with more vivid exteriorizing of fervor.
In a few cases, when I looked at an earlier score, I was surprised to find myself spontaneously revising passages in my head—mentally hearing them performed differently than the way I had originally imagined them. When this happened, I sometimes decided to revise the earlier score so that the notation matched the music the way I was now hearing it in my more unbridled imagination.
In some cases a few small revisions triggered others, and these changes then cascaded into an extensive reworking of the score. More rarely I decided to treat the original score like a sketch or prototype, and in that case the reworking might include structural changes, a new governing concept, and perhaps even a new title connected to that concept.
When I have decided to make revisions, the impetus has most often been a desire for greater expressivity, including such things as more telling textural details, or a more fascinating sense of physical motion.