Schumann’s symphonies have traditionally been criticized for their amateurish orchestration. This criticism is valid. Schumann attempted to transfer the mechanics, techniques, and acoustical character of the piano to the orchestra, which works poorly. Transferring orchestral style sonority to the piano, on the other hand, paradoxically works reasonably well; consider Brahms’ First Sonata or Shostakovich’s op. 34 preludes, or Stravinsky’s “Serenade in A” for example.
The “Spring” symphony, composed in 1841 and published as Schumann’s op. 38, has many virtues. It is a lengthy work but one absolutely without “longeurs”, it abounds in contrast, it is by turns exuberant and lyrical, the melodic writing is consistently superior without compromising the sense of symphonic narrative and drive, and its organization and structure is flat out brilliant; consider the beautiful masterstrokes that frame the piece, the evocation of Schubert’s “Great” C Major symphony (whose manuscript was discovered by Schumann in Vienna) in the horns and trumpets fanfare at the beginning, which summons spring, and then is reiterated with perfect calculation at the beginning of the recapitulation in the first movement, and the lovely and unexpected birdsong cadenza that elegantly and touchingly launches the recapitulation in the finale. This warm, youthful, and luminous work is a treasure.
But the orchestration is problematical.
This may come as a shock to the uninitiated, but conductors alter orchestration all the time, and not just in Schumann. Yes, in Beethoven and Schubert as well, if not usually in Mozart and Mendelssohn. Just go to a rehearsal. In no time you’ll see a conductor ask a clarinet to double a passage for bassoons and horns, or divide the double basses so only half play a scampering figure, or tell the flute to play something an octave lower, or tell the disengaged last member of the second violins to take his finger out of his nose, pack up his fiddle, and go home. Just kidding about that last one! But seriously, this stuff happens all the time, and needs to happen. Only idiots and children think you can serve the music by slavish devotion to the printed text. And composers who are also conductors probably make the most changes. Look at Mahler, for example, who changed so much in the Schumann symphonies that he has his own version of the works. Mahler also made changes again and again in his own works.
Here is an example of Schumann’s bad orchestration. The beginning of the second movement features a lovely melody in octaves for the first violins, a sustained, syncopated accompanimental texture for the second violins and violas, and a bass line in contrary motion to the melody in the cellos and basses. Here are the problems:
1. Octave doubling of melodies is great on the piano, but ineffective in the violins, especially in the relatively less expressive middle of the violin register, where this melody lies. The octave doubling adds little weight and makes the melody a tiny bit out of tune, which is sometimes welcome, as it warms up the sound, but not here, where it is essential that the first violins, who are all on their divisi lonesome, hold their own against all of the second violins, violas, cellos, and basses.
2. The second violins and violas are trying to duplicate the effect created by a pianist’s right foot on the sustain pedal. Maybe Schumann congratulated himself on finding such a subtle rythmic expedient to represent the piano’s pedal, but he shouldn’t have been trying to represent the piano’s pedal at all, the orchestra has plenty of ways of its own to create sympathetic vibrations. Also, and most damaging, the ensemble is gonna be a huge problem. Even the finest players in the best orchestras are going to be tentative here, they are not going to be as comfortable finding the beat and coming in properly, with well co-ordinated ensemble as they could be. And all for nothing. Plus, this passage is likely to consume valuable rehearsal time.
3. All the cellos and basses playing a kind of mirror image of the first violins (who are not in a brilliant register, or in a great violin key [the key here is E-flat Major]) are more than a match for the first violin section; there is a danger of a bottom heavy sound, and in any case, the melody should predominate, as this is a homophonic texture.
Okay, Tough Guy, your solution?
1. Have all the first violins play the top octave of the melody, and half of the second violins and half of the violas play the bottom octave of the melody.
2. Have the syncopated accompaniment played by the left over second violins and violas, and re-notate their parts so you don’t have two sixteenth notes tied together commencing on off-beats. Use overlaps rather than rythmic unisons, so you get the sustain, but don’t have ensemble problems. You might consider putting mutes on the players executing this passage. That depends, it’s hard to know if it’s necessary or desirable before trying it out in the hall.
3. Keep the cellos as is, but make the basses (who double the cellos an octave lower) punctuate rather than double the cello line. In the first measure they could play an eighth note E-flat, in the second an eighth note A-flat, in the third they could take the whole value of the D, as the melody is now in a more exposed register, etc.
Conductors have to mess with stuff like this all the time, it’s a tough job, I promise you. When done right. Some conductors might skate past the whole problem. I have not heard Mahler’s solution to this passage, but I understand that his versions have been recorded. Perhaps someone who has heard what he does can enlighten me…I’m sure his solution to the passage discussed is better than mine.