Lukas Foss told me once (by way of justifying his reorchestrating of parts of Beethoven’s Eroica prior to a Milwaukee Symphony Orchestra rehearsal he was conducting) that, as far as he was concerned, “we should always treat music as though the ink’s still wet.” Fascinating as the resulting performance was (it certainly had an electric spontaneity to it), Lukas was asking for an awful lot; his days as music director there were numbered.
As a composer, I was apalled when, twenty years ago, I showed up at the first “orchestral read” of a revival of one of my operas at a regional opera company, looked at the parts on the stands, and realized that the conductor had gone through them and—with great care—changed all my dynamics. I caused a fuss with the chap, who explained that he had limited rehearsal time, and that he was simply doing what he had to to make sure that my orchestrations worked with him on the podium—a variation on the old musician trope "play the clarinet you have in your hands, not the one you see in the store window." In other words, I learned over the years, it was I who was shocked to find gambling going on at Rick's. Twenty years later, settled in my seat in the theater to observe a wandelprobe of a revival of another of my operas, I wondered, throughout the first act, why I couldn’t hear the low piccolo doublings of the violins (a useful commercial pit orchestrator trick that subtly firms up the pitch and plumps the tone of a small section of strings) and the very high, Britten-esque passages for two piccolos (they make orchestral climaxes for a small orchestra sound a lot bigger). When the musicians took a break, I walked down to the rail and leaned over and asked the venerable maestro engaged for the revival quietly, “Where are my flutes?” He shot a look at the flutes, both of whom were swabbing out their instruments and all at once attentive. “There was a lot of low piccolo that can’t be heard, and a lot of very high piccolo that sounded shrill, so I had them play everything on the flute in the correct octave,” he replied. “Ah,” I said, “I understand. Thank you.” I made quick eye contact with the flutes as I turned away. One nodded almost imperceptibly. Subsequently, they played their parts exactly as written. I’m proud of that moment, because it is the way I believe a mature professional composer should behave.
Nevertheless, the older I get, the more I agree—when it comes to my own music, at least—with Lukas. I now look to Verdi and Puccini, who laboriously crafted new iterations of their operas for each major production, adding and subtracting arias, changing tessituræ, crafting—in the Italian fashion—roles specifically to the artists who would sing them. When I worked as a proofreader and copyist on Broadway I witnessed firsthand as songs were added and excised from scores by the shows’ creative teams at lightning speed. After all, the American music theater would be a lot poorer today if Stephen Sondheim hadn’t retreated to a hotel room in Boston during out of town tryouts for A Little Night Music and come up with Send in the Clowns.
Astonishing it was, back in the early 80s, to sit next to David Del Tredici in the shed at Tanglewood as the orchestra rehearsed one of his magnificent, sprawling Alice-inspired orchestral works, and to see (in green pen for Solti and Chicago; blue pen for Bernstein and the New York Philharmonic, red pen for Slatkin and St. Louis, and so forth) his small, though trenchant revisions as each score was run through its paces by a different set of players. Even more astonishing it was as a student in Philadelphia to examine Leopold Stokowski’s copy of Stravinsky’s Rite of Spring, and to see, in one color, his own orchestrational changes for performance in Philadelphia’s Academy of Music, and Stravinsky’s own, in another color, for another performance.
Nevertheless, when it comes to concert music—symphonies, string quartets, and so forth—there’s still a strong feeling amongst most composers that “the document”—that thing labored over in private for months and years by the solitary composer in her studio—is sacred, and that changes are made only with the greatest trepidation. Even I, as hard as I’ve worked to cure myself of this attitude, find it hard to revise my symphonic works. Orchestral rehearsal time is incredibly expensive—especially nowadays, when a twenty-minute long composition can receive thirty minutes’ worth of rehearsal before the first performance. When a player stops the rehearsal to ask a question, it costs money. Moreover, although the composer has (in principle, at least) all the authority AND the power when her music hits the music stands, every question diminishes her authority. The players cease trusting the dots and dashes on the page. They begin second-guessing things. The result is as inevitable as it is chaotic.
Consequently, the full scores of serious concert and operatic works attain an almost tombstone-like stolidity, crafted as they have been to withstand bad performances and facilitate great ones. I’ve conducted Aaron Copland’s Third Symphony with a community orchestra that struggled with it, and a regional orchestra for whom it represented no serious technical challenge. The transcendent glory (and I mean glory!) of its execution is that it came off with both.
What does a composer do, after the inspiration and composing is over, to protect her vision and to furnish to the players the most durable road map she can—one that, like Copland’s Third, will make a bad orchestra sing and a great orchestra burst into flames? I was asked this the other day by one of my adult pupils whose opera was being premiered at long last by a major company and wanted to know if I had a “work routine” I could share with him so that he wasn’t at the mercy of the generosity of the company’s orchestra librarians and musicians once his music hit the stands. I was surprised to admit that I didn’t have one. Sure, I had in my files “work routines” for use back when I was a proofreader during the 90s, but nothing more current that considered engraving software and contemporary practices. So, I jotted these thoughts about “ten passes” through the score for him, and share them now with you. They are by no means comprehensive, but they represent a starting point, and making yourself go through the score ten times to check for these things will make life better for everyone, including the audience.
First I go through the vocal parts and recheck the hyphenation of every word with a copy of the Oxford English Dictionary. Then, I go through again and check the punctuation. Singers and vocal coaches look to punctuation for an indication not just of what the words are trying to express, but where they can breathe. Finally, I check the prosody. In short, I set text for sense first, then, for sensibility. I avoid putting unaccented syllables on downbeats, since that isn’t the way people speak in real life. Well, William Shatner does, but he’s the magnificent exception.
I go through and check all the key changes if it is a tonal work. Engraving software tends to leave vestigial bits of code around double bars that confuse not just the midi triggers hidden in your score file, but they mess up the spacing. This leads to a pass through to double check the layout of the score pages. Most software defaults to putting too many bars on each system of music so that, when you must squeeze extra music in to facilitate a page turn you can. I deduct a measure each page to give more space to the music.
Then I pass through the clarinet part to see if I can’t make the keys easier by having the player switch to an instrument with a different transposition, like D or E-flat. I then do this for the trumpets. Finally, I’ll change heavily sharp keys to flat ones for the brass. Happy players perform better.
Then I go through and recheck all the slurs in the winds and brass. Slurs in the winds refer only to where the player should breathe, not to the shape of the phrase. Then I check all the bowings (yes, I do my own bowings) in the string parts. Again, bowings are NOT phrase marks. If there’s something tricky, I’ll pick up a violin and try it slowly; I’ll physicalize it so that my authority is actual, not hypothetical.
I pass through the entire orchestra to recheck articulations. Each composer develops her own personal glossary of what each accent means. I lean on Benjamin Britten (whose articulations are the best of anyone’s—they always sound without special explanation, even out in the audience) and Richard Strauss, both of whom spent a lot of time on the podium and were taught a lot by players about what they needed to see in order to give the composer what she wanted.
I then conduct through the score one more time, checking to make sure that the time signatures I’ve chosen match the beat patterns that the conductor will likely choose to keep things together most efficiently. Sure, they’ll choose their own patterns, but, as with bowings, one wants to establish a basis for a mutually-respectful dialogue.
I then check the “dove tails” in the score. These are the points when players “hand off” tunes to one another—whether from solo to solo or from one choir to another. These are frequently a little tattered. A seamless orchestral sound is something attained only through attention to this detail. One never gets any credit for having done it, but one can tell when a composer hasn’t. (Remember, craft is only really satisfying when it is good enough to conceal itself.)
I then go through and check the dynamics. I remind myself that the players have been trained since childhood to balance with one another. Second-guessing their training leads to the same chaos that a conductor “following” rather than “leading” the orchestra does. It’s like a sonic hall of mirrors, and it leads to disaster. If you want the winds to balance as a choir, just give them all the same dynamic and score it accordingly. Nothing else is needed. If you want the different sections of the orchestra to balance, look at the repertoire and you’ll see that they are marked the same dynamic; the composer’s choice of suitable ranges is what ensures the balance, along with the players’ training. Fine-tuning with all sorts of dynamics within the chord leads to stressed-out players and weird sounding tuttis.
After running through the percussion parts to make sure that I’ve given the players enough time to run from one instrument to another, I check the rising chromatic lines to make sure they are spelled in sharps, and the falling chromatic ones in flats. This is particularly important when the music is based on an octatonic (or any artificial) scale. The players see only one part in front of them. All those augmented seconds make sense intellectually when you see them in the full score, but they make a single line player's life harder. That said, a famous composer once asked me “Why do your chords ring and mine don’t?” I was compelled to answer that it was because I spelled mine correctly. After all, an A sharp is higher in pitch than a B-flat, and so forth. The other composer was not amused.
Finally, I go through and make certain that all the rests are “collapsed” into sensible groups. Double bars exist only as a reminder to the player to “look up” for information from the podium. Composers who’ve mainly played chamber music always divide up the beats too much in their orchestra pieces. Players need to see only where the stick is probably going to be in their peripheral vision—nothing else. Then, if I’m using Sibelius software, I’ll go through and “reset note spacing” to get rid of more digital kudzu, and then “lock score” and “freeze position” so that all my work isn’t lost.
I’d say that, if you do all that, then you’ve found about ten percent of what is likely to go wrong in rehearsal. Throw your hands up in the air and begin again, friend Sisyphus. The ink's still wet.
This essay has appeared in the Huffington Post. Click here to read it there.