Five things I learned writing my first piece for orchestra
This article was published on my previous blog on July 21 2022.
During the peak of the pandemic when I had little else to do, I decided to devote most of my all-too-plentiful time and what little energy I had to writing a piece for orchestra, something I'd wanted to do since I began composing back in high school. I'd had a couple ideas for orchestral works over the years, and I decided to finally commit one to manuscript. The other major idea I had has transformed over the years into something completely different, bearing almost no relationship with the original composition, with the original musical ideas becoming a modest song without words for violin and piano.
The piece I did compose, Emergence, is an orchestral prelude. I originally conceived of this movement as the first of a five-movement tone poem. I elected to focus on just the prelude: it was the movement that could stand on its own, and it was the one in which I had the clearest musical ideas. The other movements were hazier and had little to go off of except really basic melodic ideas and a general "vibe."
At the end of the arduous process of writing this piece, meeting a deadline for a call for scores, I stepped back to look at what I made... and I didn't like it. I have barely thought about the piece since then, and after all that hard work to write a piece I didn't like, I snapped pretty quickly into composing my first album of rock/jazz tunes, playing to my existing strengths as a musician rather than to my artistic ambitions. (I plan to talk about my similar struggles in composing the album in a future article.)
Through this writing process, I learned a lot about myself, my musical tendencies, along with some hard lessons. And I am going to share those with you in hopes that you learn something useful through my own artistic struggles.
I - Don't Go Overboard
It can be easy to fall into the trap of using every single technique in the book to emulate the great orchestrators that we look up to. In my composition classes, for instance, we looked at some wild scores by Lutoslawski, Penderecki, Ligeti and Ives who really stretched the orchestra to the limits of what they were capable of. On my own I was listening to and reading these gorgeous and complex scores by Lindberg, Dutilleux, Saariaho, Takemitsu, Adès and Norman, and while there wasn't the same focus on experimental notation, aleatory and extended techniques, their scores were still sprawling and awe-inspiring.
Trying to do that as a young composer, however, has many problems. For one, those orchestrators wrote with decades of experience I don’t have, so they know what works in an orchestral setting and what doesn't. Many extended techniques you hear in the solo repertoire are too quiet to even be heard in a concert hall, at that's without considering the difficulties of those techniques. Plus, these older more experienced composers can get performances by some of the best players in the world. Me, I’m in my twenties, I’ve never had any orchestral music performed, and most realistically I can get a performance from a local college orchestra or smaller independent ensemble, not a premiere by the New York Philharmonic.
The ensemble may not be used to playing works with experimental notation, difficult solo parts and less common instruments. Major orchestras will usually have access to the largest capabilities of the ensemble: quadruple winds, massive brass sections, full string sections and any percussion instrument conceivable along with players on call to fill in the occasional organ, cimbalom or guitar part. College orchestras may have a fraction of this capability, and smaller orchestras composed of community players or part-time musicians rather than the full-time union orchestra musicians also have time and monetary limitations. (On the other hand, smaller groups may be more willing to experiment in ways that large orchestras with dozens of concerts a year cannot; it all depends on who is running the ensemble, what resources are available and the sensibilities of the players.) That, given how precious rehearsal time can be, means that it's always best to keep things simple and make the process for the musicians as seamless as possible.
These are just a few of the unfortunate realities young orchestra composers face. The orchestra is an old institution, and it takes thousands and thousands of dollars and dozens of people to make even the smallest performances happen. Anything you can do to make this process smoother and less difficult will be immensely helpful.
A Case Study In Going Overboard - String Divisi Nightmares
To give one specific example of how I went overboard with this first score, let's talk about string divisi. I decided that the main texture would be a blooming string canon, expanding out through the sections from the center outwards. As such, I divided each string instrument into groups (four groups of 3-4 for the violins, three groups of 4 for the violas, and five groups of 2 for the cellos, omitting the basses). I imagine any experienced string player reading this just whispered "oh, oh no..." upon reading this, so let me explain why this was such a bad idea.
The more divided the string section gets, the harder it becomes for smaller groups of instruments to be heard or stay in tune. There’s a reason a good-sized orchestra will have nearly thirty violins: that's how many you need to have even a chance of competing with a single trumpet playing fortissimo. Also, having that many violins not only amplifies the group but also creates the pleasant chorusing effect we expect from the ensemble. The few instruments in such a group, the harder it is for them to play in tune. Any musician with experience performing in an ensemble will tell you it is far easier to play in tune with two dozen people than with just two. The more you divide the string section, the more you open yourself up to intonation problems.
This isn't even mentioning the problem of seating. For this particular gesture I envisioned the sound radiating outwards from the center of the orchestra out towards the edges, so I started with the back of the violas and second violins before adding in the cellos and first violins. But not every orchestra seats the strings like this. Some European orchestras put the seconds across from the firsts with the violas and cellos in the middle. Even my local orchestra puts the violas outside the cellos! Rather than having to re-orchestrate the passage for each performance or asking an orchestra to change its seating for my one piece, its probably better to ditch the idea altogether.
Besides, given the limitations mentioned above smaller ensembles like college and independent groups are less likely to have full string sections, so it helps to go small and save yourself any headaches. Keep divisi to a2, soloists, or maybe some specific effects dividing up one section. No need to do what I did.
Similarly, I would keep the instrumentation minimal. Double woodwinds, 4221 brass, timpani and a relatively small percussion section, maybe a harp, and however many strings you can muster is about the best you can hope for from the ensembles who would realistically play your first piece. Best to compose with that in mind.
I'll come back to this point later, but this absurd divisi also had the effect of making the notation and writing process far more complex than it needed to be, which had an immense psychological effect on me while I was writing the piece. Writing for orchestra can become immensely difficult and complicated, and it's very easy to get lost in the woods. Anything you can do to simplify the process will help. And speaking of...
II - Use a Short Score
Writing for orchestra can be extremely overwhelming, given how many staves there are, so it can really help to cut that down as much as possible. A very common way for orchestrators to start the process of writing is to begin from a short score, or a smaller score that acts as a template for the fully-orchestrated composition. Very few composers write directly onto an orchestral score (and I both envy and pity those who can), so even the best of the best have to start somewhere. Ravel, for instance, orchestrated a lot of his piano music, using that as the basis, and if it's good enough for Ravel it should be good enough for you.
Starting from a short score is super helpful, since it allows you to focus in on the most basic musical ideas, what is the most essential for each moment. I like doing a two piano, four stave short score, with one lining out the melodies and bass-lines and the other indicating accompaniment or textural figures that would clutter the space if reduced to only two staves.
Another more detailed way of writing a short score is to write a grand staff for each section of the orchestra: one treble and bass clef each for woodwinds, brass, percussion and strings. This makes moving from short score to full orchestration very smooth and simple and works really well if you have particular instrument groups in mind for certain phrases and gestures.
The broader point is to find a workflow that you find helpful. Some like to compose by hand and use notation software like Sibelius, Finale, Musescore or Dorico purely for engraving. This system has its benefits such as allowing you to write things without regards for what the cumbersome notation software will do to it. The brain-to-pencil connection is far stronger than the brain-to-keyboard connection. (If you are curious as to which one of those programs I think is the best, I think they are all terrible in their own ways. I use Finale just because it's what I'm used to, and it's quite powerful despite being extremely user-unfriendly, but I would not recommend it to new composers.)
After working on my album and getting to know my DAW better, I've found a couple unique ways that I work that helped speed up my workflow significantly. I'll talk about those later, and that's just what works for me, but I would still find some way of simplifying musical ideas to make the ideas as simple as possible.
III - Score Study is Your Friend
While I was writing Emergence, I was also score studying a lot. I had already been score studying and taking private lessons before the pandemic, but when my schedule opened up I had all the time in the world to score study. I would spend hours every morning cracking open scores, listening to new pieces, watching score videos on Youtube, and scouring IMSLP, Issuu and publisher's websites looking for everything I could find.
Digging super deep into each score got me into the mindset of asking various questions about why this was done over that and what effect that has on the music. It also helps to have models: if there’s a particular effect you are going for, look at a score to see how they did it.
For score study, usually the advice is to look to the classics–there are countless lists out there for various pieces that one should study, and there is merit to that, to reaching beyond your own tastes. But I also think it’s most important to study music you like first and foremost, since that will give you the best template for how the music you like was written.
It also helps to know some things about how to read a score. It’s so easy to get wrapped up in the details that you fail to see the big picture sometimes. Try looking at a passage and ask, what isn’t on the page? Why would the tuba be doubling the basses in this other part, but not this one? Also, there is no linear correlation between pages and lengths of time. There are pages of score that can take minutes, others that breeze by in seconds. There are also a lot of details a conductor may notice that a score reader obsessed over details may not pay attention to, such as tempo changes and moments where the music can stretch and contract.
Knowing what has and hasn't been done before is extremely helpful. But you also have to ask yourself: is the reason this hasn't been done before that it's a good idea no one has had yet, or that it's a terrible idea that no one should even try? This is something one gets to see from studying dozens of scores, noticing trends, common techniques and effective strategies. Given the difficulties of getting a piece performed, it may be wisest to avoid anything too ostentatious and complicated. To return to my divisi problems, I may have noticed some impressive divisi scattered throughout the scores I studied, but looking back they were often only divisi a 2, or using only a smaller part of the section (one player from each section, three players per section, one player per desk, etc).
So where do you find scores? As I mentioned above, local libraries can be great sources for all sorts of contemporary and classic music. At the time I wasn't going to the library for obvious reasons, but libraries can also have a great selection of scores, especially those in major metropolitan areas and at colleges with big music programs. The Multnomah County Central library, for instance, has a huge music library taking up a full room, with shelves upon shelves of orchestral scores alone.
You can also buy the damn things, though more recent pieces can get quite expensive. At the time I had the money and figured it was an investment in my musical education, so I did buy a few fancy scores from far-off publishers that had to ship across the Atlantic to get here. For those looking for cheaper options, most classic works are available from Dover in relatively inexpensive paperbacks.
If you don't mind using PDFs or other online resources, there's a lot out there. There are a ton of score-reading Youtube channels, one of the most incredible resources out there for budding composers and music fans. Unfortunately the abysmal Youtube copyright policy tends to hit these channels hard so there's no guarantee the piece you want will be available, or that what was there last week will be there next week.
Many music publishers have previews of their scores available on their websites, usually implemented through some in-browser PDF reader or a watermarked download. While a bit inconvenient, they do offer them for free and often boast a wide collection of contemporary music. You can also find some similar collections on Issuu, a file-sharing site.
IMSLP is a massive database of public domain scores, where you can find score for almost anything written prior to about the 1910s. This begins to get into dubious legality, since things that may be public domain in one country may not be in another. There are also issues with file sharing and privacy, which can cause problems despite not being particularly rigidly enforced. While I support the open and free access to information for all, given how difficult it can be for a composer to make money, as well as how easy it is to find basically anything legally, I strongly discourage any sort of digital sheet music piracy.
There are also a lot of younger and lesser-known composers who choose to self-publish, and you can find their pieces on their websites for pretty reasonable prices. And, from anecdotal experience, sometimes composers will send you their scores if you email them and ask nicely--though in this case it helps if you know them personally.
Also, don't limit yourself to solely studying orchestral music. You can learn a lot from all kinds of scores. Solo and chamber music literature, for instance, can teach you about idiomatic use of each instrument. You can also find equally incredible orchestration in concerti and opera scores (in fact, one of my favorite orchestral scores, Dutilleux's L'arbre des songes, is a violin concerto). Since the turn of the century there has also been a massive growth in music for new ensemble arrangements such as the Pierrot ensemble (violin, cello, flute, clarinet and piano), chamber orchestra (basically one of each instrument in the traditional orchestra), and percussion ensemble. While not technically "orchestration," these are still great scores to study. Some of my favorites for the percussion ensembles are Les Noces by my personal bête noire Stravinsky and the stunning Sur Incises by Pierre Boulez for three harps, three pianos and three percussionists. And you can't forget vocal scores as well!
As you can see, there are immense amounts of resources out there for composers who wish to score study. It would be foolish to pass up such an opportunity.
IV - Use Repetition and Blank Space
Repeats and repetitions are your friends. You’ll hear them all the time in music and there’s a reason why they are so important, especially in live music: It helps more important ideas stick out. Audiences will hear this piece once, most likely, so it helps to create a strong impression from that one performance, whether it's via a catchy melody, a dramatic moment or gesture, some unconventional use of an instrument, or anything else. If you regularly attend orchestra concerts, this will be a lot easier wrap your head around since you have a clearer idea of what the orchestra actually sounds like in-person versus over MIDI playback.
With the orchestra, audiences are expecting to be sitting there for a while, so there’s no need to condense everything down into a brief three-minute expression. You aren’t writing pop tunes. (This isn’t a dig at pop music at all, just a point about recognizing the medium you are working within.) These don’t have to be verbatim repeats, mind you. Returning to an older musical idea re-orchestrated is a common technique throughout orchestra music, as well as varying the harmony or texture. The verbatim repeat can sometimes feel like the recapitulation of a sonata, which is a bit old-fashioned, so keep that in mind as well.
It also doesn’t hurt to let an idea sit there for a while. There are countless pieces that don't shift textures or musical ideas all that often. Filling a page up with too many ideas can distract the listener from the most important central idea, in favor of impressing some imaginary audience of critical score readers (like yourself!). I often repeat to myself as a mantra when I get too bogged down in the compositional process is, "I don't know what this is, but it sounds good. I'll let the analysts figure that one out." Not everything in a score has to be explained perfectly via music theory, and ultimately that isn't the point. Music is sound in space, not a series of abstract harmonic and rhythmic structures.
One thing I realized when writing this score is that I am experiencing this piece as a series of dots and lines on a massive sheet of paper that is a facsimile or close approximation of what I hear in my head. Most people will experience this piece as pure sound, and won't particularly know or care what the score looks like. Sure, an experienced score reader can recognize a score's brilliance from how it looks on the page, but that's not a skill many people have. And really it doesn't matter that much. There are countless scores I studied that look utterly unremarkable on the page, but sounded great in performance, alongside scores that look immensely more complex than what one hears. Both of these deepen my appreciation for the music in different ways.
My point is that it's vital to think about what a score will sound like as its final product. Don't be afraid to leave blank space or have passages that look bare. Sometimes the music needs to be simple to convey what it needs to. Your job as a composer is to give detailed instructions to performers for what notes to play and how to play them. They will do the rest and make the piece sing. Let them do their jobs.
V - Get to the Double Bar
It's very easy to be overly critical of your own work. As I mentioned above, it's not hard to get caught into a loop of re-writing the same passage over and over again trying to get it perfect, before handing anything else that needs to be done. Sometimes, you just have to get something onto the page. I like to think of it like sculpture: you begin with a large slab of clay or marble, its daunting emptiness staring right back at you. Slowly you begin to work away at it, first with big dramatic incisions and distortions that become more refined over time. It does feel great to obsess over the one perfect move, that one precise detail that works perfectly, but maybe obsessing over that one perfect move when there are empty, unfinished spots may be a waste of time.
When you get to the end, it may not be perfect. In fact, I guarantee you it won't be perfect. But you have something you can look back on and refine, adding details, scrapping parts that were half-baked, re-writing things, adding repetitions, excising unnecessary clutter, etcetera. No one is going to judge you based on the quality of your first draft; it's a first draft. Even the best writers have editors. This is where your own critical skills come in: you have to look at what you wrote and judge it based on your own tastes. It's not a hit against your own ego, because you are the one judging it. The only question that matters is, "do I like this?" If so, that's great! If not, then you have some more work to do, but you have a clearer idea of what needs to be done.
I found it so deflating when I put in all this work, all to a piece I didn’t like. But it was a learning process, and I can go back to Emergence in the future with clearer intentions and a cleaner approach to the orchestration. I got to the double bar, finished the damn thing, and now I can look back on it with a critical eye to see what I can do better. I can also show a completed score and parts to musicians and ask questions like, "is this oboe part too high?" or "is this enough time to change mallets?" or "would this sound better with mutes or sul tasto?" which is immensely helpful as well.
I may not have written the piece I had in my head, or even come close, but I learned a ton through this process, as you can tell, and I hope you learned something from me sharing my thoughts on my failures.
