Is it wise for a composer — even a great composer — to return to old compositions years later, to revise and update them?
It’s a question posed by Rachmaninoff’s Piano Concerto no. 4, which was sketched in 1914, completed in 1926, and then revised and published in a heavily edited edition in 1941. Some have stood by the composer’s final decisions, upholding and performing the ’41 version (which Robert Russell Bennett helped Rachmaninoff to complete), but others are convinced Rachmaninoff got the Piano Concerto no. 4 right the first time. The original ’26 version of the piece has only been available in print since ’00, so time will tell whether an historical consensus is ever reached on which of the three versions of the piece is best (there’s also a 1928 version, rendered by Rachmaninoff’s first round of cuts).
In A Year in Classical Music: 1926 vol. 12, I draw a comparison to George Lucas second-guessing and revising the original Star Wars trilogy in the late ’90s and early ’00s, when he made the three prequels. Star Wars devotees were scandalized by the changes: the screen cluttered by legions of CGI-added aliens, Greedo made to shoot first at Han Solo, and so on. The overwhelming consensus has been that Lucas’ changes to those classic films were misguided (to put it mildly). Doesn’t that indicate that it contradicts something in the creative process to revisit, second-guess, and revise old pieces?
Tchaikovsky greatly improved upon the original 1870 version of his Romeo and Juliet overture with revisions in 1872 and especially 1880 (the final version we know and love today), but at the time of the original composition he was still a young composer who had not fully mastered his craft. In 1947 Prokofiev extensively revised his Symphony no. 4 of 1929, so much so that the latter piece is a new composition, not so much a revision of the original. Bruckner, with his neurotic self-doubt and sensitivity to criticism, revised his symphonies many times, but this has generally been seen as indecisive fussing over trivial details, not as important creative work.
So I’m not sure I can think of an instance in which second-guessing an old composition has produced a superior piece. (Ives may be the one important exception, as he spent the last half of his career tinkering with old works — but there again, it’s an exceptional case as Ives was a pioneering and experimental writer.) It seems that a work of art must be the product of a single progression of thought, developed and refined within a relatively short time, if it’s to be a coherent representation of the composer’s creative process. Beethoven didn’t revise his first two symphonies to reflect the developments in his creative process that characterize the Third; he allowed the two early works to remain as artifacts of their time and place, and presented the Third in the context of a new period of style and technique. My suspicion is that an artist — even a great artist — will be unable to return to the time of a previous stage in the development of his creative process and successfully work with the styles and techniques of that past time, even though they had once been his own. If he tries he’ll only be doing music theory, not creative composition.