I’ve been fortunate to spend so much of my life enjoying music, both as a listener and as a student of the art. But along with all the pleasure music has brought me, it has also been the locus of my deepest frustration. Since the age of 15 or so, I’ve wanted to be a composer, but for a slew of reasons, some significant and others seemingly trivial in retrospect, I spent many years stuck in my efforts to compose. Friends never knew just how stuck I was, because I rarely spoke of it, not wanting to face the inevitable request: “So, let’s hear some music you’ve written” – and also because I’m passionate about other things – computer science, photography, words, guitar – and most of those things have been easier to discuss than my aspirations in composing.  My other interests matter deeply to me, but in some sense composing has always mattered the most and been what I actualized the least. In late 2014 that changed, as I wrote myself a Christmas present of sorts, an album of carol arrangements, and began work on an album of canons. I changed from someone who wanted to compose into someone who does compose. While it felt like the transition happened suddenly, in fact I’d been building up to it for years. My earlier practice in improvisation had been my first experience of bringing new music into the world, but although I’d recorded a couple of satisfying improvisations in previous years, I had never managed to write anything down that felt like a fully composed piece, until 2014; and, at the beginning of the year, I didn’t know when or if that would happen.

For much of my life I had felt alone in my particular blockage, because frustrated composers aren’t very visible as a group. The novelist with writer’s block is so common as to be a stereotype, but when was the last time you came across someone who actively yearned to write classical music but couldn’t? You’re much more likely to encounter a successful composer, whose story may have begun with a prodigious talent nurtured from an early age, or even more likely, you’ll encounter a musician who has given up entirely on composing, deciding it just isn’t for them. Those of us who held onto our early dream of composing, but got lost as we tried to realize it, are not always vocal about our experiences.

At some point I’d like to share more of my personal journey towards composing but here I would just like to offer a few practices and strategies that have helped me along the way. These are the tips that I’d go back and give to my earlier self, the hopeful but stalled composer, if I could. These tips are not really about “how to compose” but rather how to develop yourself so you’ll be able to compose.

Links: my first albums, Carol Colors, and Canons


Learn to connect with your voice by singing against a drone.

When I was little kid I fancied myself as an opera singer and used to belt out atonal aleatoric arias with abandon, but somewhere around second or third grade I became very shy about my voice. In school chorus classes I always felt there was something I just wasn’t getting. At some point when it came time for the teacher to assign students to the excellent chorus or the crappy chorus I was put in the crappy chorus, and that was the beginning of my conviction that I couldn’t sing. In subsequent years I grew terrified of singing, because I knew I might sing out of tune, and I came to think that singing out of tune would reveal some kind of deep flaw in my musicality. I was protective about my “musicality” because I felt music was my greatest source of joy; I was scared of there being major things about this “source of joy” that I didn’t grasp and might never grasp.

I always had a keen sense of pitch, in that I could discern what sounded good and what didn’t, but I didn’t have enough of a clue about how my voice worked to be able to stay on pitch – or to reliably return to pitch when I strayed from it – and unlike people who didn’t really care about sounding bad, I cared. Maybe that’s my most important piece of advice to anyone reading this – you’ve got to stop caring about sounding bad. (Practice making some really horrible sounds every day.)

As I became more skilled as a guitarist, and more explicit about my dream of composing, I came to think of singing as my Achilles heel. Voice seemed to me as the essence, the underpinning of all music, and I thought that if I couldn’t hum a tune comfortably I couldn’t be a “real” musician. I never said this to anyone, of course; it was one of those tacit convictions that eats away at you, and, because it’s not out in the open, it never has a chance to be examined or challenged. In retrospect I see my conviction about the necessity of singing as false – of course it’s possible to be a great musician who happens not to sing. But I do believe that developing a connection to your voice is one of the most valuable things you can do in deepening your understanding of music, and luckily there’s a simple way to do it.

Let me preface this by saying you should take some voice lessons with a good teacher even if you don’t plan on becoming a singer. When I finally got up the courage to take voice lessons as an adult I found them surprisingly fun. Singing felt to me much more physically relaxing than playing guitar, because good singing is all about good breathing, and when you breathe well, you relax. Voice lessons can even be quite goofy at times as the process involves making lots of weird and sometimes unmusical noises to discover what your voice can really do. And while there are some voice teachers who focus only on professional or pre-professional singers there are many others who will gladly accept, and know how to work with, a novice who’s a bit terrified of singing. You don’t need to be confident in your voice, or to have been told you have a “promising voice” before signing up for lessons.

Lessons aside, there is a wonderful practice you can do on your own that is often ignored by Western voice teachers and can be of great benefit to musicians at any level. The idea is just to sing a steady pitch against a drone. A drone is any instrument that keeps announcing a single note – so it could be a guitar or piano where a note like C is played over and over again – but ideally it is a continuous-pitch instrument like the cello or the Indian tanpura that can create the impression of a single, never-ending note. All you have to do is listen to the drone and match it with your voice. Try to align your voice with the drone and whenever you sense that you are straying from the drone pitch, try to return to it. If possible, use a recording of a drone or buy an electronic tanpura box, or a tanpura app, and keep it playing as long and as often as possible. When you’re washing the dishes, put the drone on and sing along with it, just one note. Instead of putting the radio on in the morning while you’re making coffee, play the drone and sing against it, just one note.

If this sounds like a remedial exercise, know that it is both remedial and highly advanced. In Indian classical music the tonic pitch announced by the drone is called “Sa” and even many senior, master musicians in the Indian traditions make a point of practicing Sa on a daily basis. Most Western musicians have ample experience matching pitch but few have the experience of sustaining the same pitch for minutes or hours on end. Once you think you can sing a continuous, steady pitch, staying aligned with the drone for seconds or minutes, listen a bit closer, and whatever level you’re at, you’ll probably become aware of ways you can tune in even more intimately with that one pitch. And once you’re really comfortable singing just one note, the rest of singing is so much easier.

Links: me singing the Hindustani Rag Marwa while playing tanpura, me practicing Western classical singing in a voice class after my first few years of lessons


Learn that the piano is not always correct.

It’s common to use the piano as a reference when checking out a musical idea or trying to gain your bearings in singing. Many musicians have the experience that when you’re off pitch, referring to the piano brings you back to pitch, and so it comes to seem that the piano is the arbiter of musical truth. But it’s important to understand that most pianos are tuned to an approximation of equal-temperament, which is a somewhat artificial system designed to facilitate the reuse of the same twelve pitches to support multiple tonal centers with identical intervallic structures, and to support modulation between them. One of the intervals that “suffers” most in equal temperament is the major third. You might have the experience that while singing a major third above a given note, you zero in on an interval that sounds pure and sweet, but then when you go to check your pitch against the piano’s pitch, the piano’s version is higher (e.g. you’re singing E against the piano’s C, but then when you check the piano’s E you find that it’s higher than the E you were singing). In a sense, you’re right and the piano’s wrong, because you’re singing an acoustically pure interval while the piano’s version represents the compromise of equal-temperament.

Some musicians who become sensitive to intonational issues start thinking of equal-temperament as an evil system that has destroyed musical beauty, but I don’t feel that way. I’m happy to write and listen to music that depends on equal-temperament, but I think it’s important to understand what equal-temperament is, and to do that you need to understand what the so called “pure” intervals of just-intonation are. Unless it happens to be your passion, I’d recommend not becoming obsessed with all the mathematics of tuning theory – just focus on the basics. Become acquainted with the sound of a pure major third (where the higher pitch’s frequency makes a 5/4 ratio with the lower pitch) and go on to explore some of the other intervals of just-intonation. Understand the harmonic series. Even if you don’t use alternate intonation systems in your music, it’s worthwhile to know about the physical basis of sound and to recognize that each tuning system has its pros and cons, so the question of what’s really “in tune” is actually a relative question – what’s “in tune” in one system is “out of tune” in another.

Links: my posts on musical tuning


Learn to write bass lines.

I went through a standard harmony course in college and have read dozens of harmony textbooks since then. After dispensing with musical fundamentals, most harmony texts have you dive in and write chords under a melody. But in my view, chords, even “simple” triads, are very complicated things. As soon as you write a full chord under a melody note you’ve committed to one interpretation of that particular note. The commitment is “sticky” in a sense, because creating good voice-leading between chords is tedious work, and when you want to change a chord it usually involves not just changing one note but changing multiple notes in the surrounding area and having to deal with all sorts of voice-leading constraints. Chords are great but they’re slow to work with. On the other hand, if you simply write a bass note under the melody note, you’ve made enough of an interpretive commitment that the music can start taking shape, but it’s not a “sticky” commitment – bass lines are easy to revise because you can often get away with just changing one note and not making major adjustments in the surrounding area.

I think the study of harmony should begin not with chords, but with writing bass lines under melodies. Don’t think of the bass line as a mere first step in creating a full harmonization – treat the bass line as a work of art in itself and really spend some time on it. Take some familiar tunes, like Christmas carols (suggestion of David Berkman in the Jazz Harmony Book) or some Pete Seeger tunes (suggestion of Hindustani vocalist Warren Senders) and try writing bass lines underneath them. To do this really well, you’ll need a combination of the knowledge that comes from first species counterpoint along with the knowledge that comes from a jazz musician’s study of walking bass lines. So read up on counterpoint and read up on jazz bass lines. Even in the jazz world, it seems the knowledge of how to construct bass lines is left to bassists and not always studied by guitarists and saxophonists and singers, and that makes no sense, since bass lines are fundamental in all music.

You’ll find that if you take a tune and write a mediocre bass line underneath it, the pair will seem to be lacking something, and you’ll want to flesh out the harmonies with full chords, but if you get the bass line just right, then the pairing of bass line and melody can be fully satisfying in itself, and you don’t need anything more. There’s something really magical about a good bass line. Writing bass lines is an art, but it’s an art that’s learnable, and it’s probably the quickest path you can take to finishing your first composition: compose a simple tune, or “steal” a tune from somewhere and modify it, and then write a bass line for it. There you go – there’s your first piece!


Learn to make musical outlines.

Music can get very complicated very fast. One of the challenges in writing music is that you might write a complicated and beautiful opening passage and then not know how to continue it. The worst thing is to find yourself with an array of options with no idea which of many seemingly promising possibilities will lead to a good result. That’s why it’s best to start with an outline.

What is a musical outline, exactly? The nature of your outline will depend on the kind of piece you’re writing, so let’s say all you’re trying to do is write a tune. You could start by creating a sequence of whole notes – that’s your outline. Once you’re happy with the shape of the sequence, try embellishing it: first try adding passing tones in half notes to connect notes in your outline that are situated a third apart. Try filling in large skips in the outline with scale passages. Then try varying the rhythm. Keep experimenting like this and you might be amazed by how suddenly a “new” tune can emerge from what seemed like a plain succession of whole notes.

Let me given another example from a form that I’ve been working with: canons. If you listen to one of my canons you might hear an ornate and elaborate melody stated in one voice and copied in another, and if you know the constraints of the canon form you might find it miraculous that I can fit such complex melodies into such a restrictive form, but it’s not a miracle, it’s a process. My secret is that I don’t start with complex melodies. I begin my canons as outlines, usually in whole or half notes, and the complex melodies emerge via a process of elaboration. Get a good outline, and when you elaborate it you’ll be sure to have a good piece. A lot of times when I feel something’s not right about my elaborations, I go back and listen to the simple outline, and usually that gives me an idea of what to do next; if I didn’t have an outline to return to, it would be harder to regain my bearings.


Get comfortable with a realization tool.

I’ve always been in awe of composers who can conceive of intricate pieces in their inner ear and then write them down without referring to an instrument. As with singing I used to think of this as something I’d never be able to do, yet something I’d have to be able to do in order to be a real composer. It’s undeniably valuable to develop your aural imagination, and I continue to develop mine, but the fact is I can’t write the kind of music I want to write in my mind alone: I need to hear it as I’m working. Sometimes I’ll play and listen to a short passage hundreds of times before I feel I’ve gotten it right. Where did I first get the idea that composing in the mind’s ear is the “right” or “true” way to compose? I’m not sure. While there are some teachers out there who go beyond advocating for the value of aural imagination and actually suggest it’s the only good way to compose, no one ever told me that.  Being able to compose “in my mind” was a self-imposed requirement. Maybe some of it came from hearing stories about how Mozart imagined symphonies faster than he could write them down, and how Beethoven was able to continue composing into his deafness, and I figured that the capacity for inner hearing must be what defines a great composer. Hindemith expresses this sentiment in A Composer’s World:

We all know the impression of a very heavy flash of lightning in the night. Within a second’s time we see a broad landscape, not only in its general outlines but with every detail. Although we could never describe each single component of the picture, we feel that not even the smallest leaf of grass escapes our attention. We experience a view, immensely comprehensible and at the same time immensely detailed, that we never could have under normal daylight conditions, and perhaps not during the night either, if our senses and nerves were not strained by the extraordinary suddenness of the event.  Compositions must be conceived the same way. If we cannot, in the flash of a single moment, see a composition in its absolute entirety, with every pertinent detail in its proper place, we are not genuine creators.

Hindemith’s standard for being a “genuine creator” is, in my view, unnecessarily limiting. The real question is: Do you want to create? If the answer is yes, the next question is: Are you ready and willing to use all means available to you to create?

At some point I decided to abandon my hangup about what I thought was the “right” or “true” way to compose and realized that I cared more about composing than about doing it in any particular way, and I decided to use all means available to me to do it. I would suggest that while you should always cultivate your inner musical imagination, you should never hesitate to use an instrument and you should not consider it a handicap if indeed you need to use one.

Working ideas out on an instrument is a good way to ground your composing in the physical reality of sound and in the constraints of a particular mechanism for producing sound, and constraints inspire creative solutions. On the other hand, it can be frustrating if the music you want to write exceeds the skill you’ve developed on any particular instrument, and you find yourself struggling more with the instrument than with the score itself. Even if you’re highly skilled on an instrument you might find that you can’t experiment with dozens of versions of a difficult passage without dedicating hours to practicing each one (sometimes a difference of one note means an entirely different fingering or bowing pattern), and the struggle in playing through each possibility brings your creative efforts to a halt. That’s why notation software (like Finale, Sibelius, Notion) is a blessing. With notation software you can try hundreds of possible variations of a given phrase and hear how they sound without having to practice each one on your instrument – it feels too good to be true, but it’s true. I used to have a romantic image of what composing was about, and the idea of sitting in front of a computer working with notation software did not match that image – it seemed artificial and unromantic. But I got over that hangup, learned to be comfortable working with Finale, and that’s when I finally felt my compositional imagination had been set free. Notation software was not available to the great composers of the past, and they managed create some of the most wonderful music we have without it, but the fact that it exists today is a great advantage to anyone who wants to compose and it’s one we should enjoy.


Practice structured and unstructured improvisation.

One of the challenges of studying music seriously is that you may become very particular, even judgmental about sound and you may lose the ability to listen to music without simultaneously criticizing it. A big step in my musical journey happened when I discovered the beauty of Himalayan singing bowls – bronze bowls that let you evoke a wide range of sounds all by rubbing a wooden mallet (sometimes covered with leather) along the perimeter. These bowls produce sounds that are as stunning as might emerge from the finest of any human-conceived instrument, but you don’t need a huge amount of skill to play them – you can pick up a singing bowl and in a few hours or days of practice, already you’re unleashing some of the most amazing sounds imaginable. The catch is that you can’t really play melodies or conventionally structured compositions on a single bowl – all you can do is make sound. It doesn’t make sense to try to excel at playing a singing bowl or to get it perfect. From singing bowls I learned the idea of exploring sound as sound – not thinking of sound as organized into discrete notes, but rather thinking of it as a continuous, unfolding experience that need not require simultaneous evaluation.

After spending lots of time playing singing bowls I began experimenting with “free improvisation” on guitar where my goal was not to make music per se, but just to produce a sequence of interesting or expressive sounds. Having studied classical guitar for many years with an aim towards “perfect” execution, this was a really big change for me, and I found that when the mood and setting were right, my free, unstructured, improvisations could turn into compelling music. For me what defines free improvisation is that you don’t try to direct the improvisation with your conscious mind, which is to say you don’t try to plan and predict how your next gestures are going to sound; rather you just do something to make a sound, then you listen to the outcome, and then you do something else to respond to it. As long as the responses are always connected to the thing that happened before them, music made in this way can sound cohesive and even planned although it isn’t planned.

But one of the limitations of this kind of improvisation is that it’s hard to do “on command” – so much of it depends on the mood and the moment – and because you’re not thinking about things with your conscious mind as you improvise this way, it can be very hard to remember and recreate what you did. For that reason I got interested in also learning to improvise in a structured way as is found in many styles of jazz, where “spontaneous creation” happens in the context of melodic motifs and chord progressions that you’ve practiced beforehand. I think the most basic aspect of jazz study is learning to elaborate simple melodic ideas over chord changes. So to get started in this practice what you need to do is get a hold of some backing tracks where a chord progression is repeated again and again. Try soloing over the changes. Don’t worry about it sounding like jazz. Just see if you can create some melodic lines that make sense over the chords. Backing tracks abound – anyone can find some decent ones and start this practice any time. My first attempt at doing this was comical because, while the backing track projected an aura of coolness and nonchalance, I found myself completely incapable of keeping up with it and wondered how such calm-sounding material could be so stressful to work with… but, it gets less stressful and more fun as you practice.

Links: some of my experiments in improvisation


Learn to recognize your musical responses.

I’ve said that the ability to conceive complex scores in your mind, while wonderful to posses, is not a requirement for creating good music. But what then is a requirement – what are the skills that all composers should have? In my experience, two of the most important things are being in touch with your musical responses, and knowing your options. To be in touch with your musical responses means that, when you hear something, you not only experience the sound but you notice the details of your own reaction to it. In listening to a passage, you notice what works and doesn’t work for your ear. Is there a place where a melodic line starts losing its connectedness in your ear – where it stops sounding fluid and coherent to you? Is there a place where the rhythm becomes slightly boring for you? Specifically where does it happen? If you’re not sure at first, do you know how to track it down, by playing lines individually, or listening to the score at different tempi, or stripping detail out of the score until the problematic points come to the fore? Knowing how you feel about a passage, and then being able to pinpoint the specific features of the written music that make you feel that way, is an invaluable skill because it lets you take the next step toward improving what’s written.

Of course, in order to take that step, you need to know your options, which is to say you need to understand what kinds of experiments with the score are likely to be productive, what kinds of modifications to the music are likely to address the problem at hand. Knowing music theory – not just having read theory textbooks but having really grappled with the subject to synthesize your own understanding of it – is indispensable since it lets you see your options clearly. (Instead of saying “That G-flat sounds funny, let me try changing it to a G-natural” you can say “That G-flat is the flat-5 of a minor-7-flat-5 chord, and maybe I don’t want a minor-7-flat-5 chord there.”)

As for getting in touch with your musical responses, I recommend periodically returning to a very simple exercise. Sit down with the piano or any instrument of choice and play a random sequence of a few notes. Let’s say you play C, E, F#. Now, how does that sequence sound to you? What does it mean? What does it remind you of? How would you continue it if you could add one more note? How would you answer it with another sequence of three notes? How do those notes sound when played together as a chord? What chords would those three notes sound good on top of? Keep doing this, keep asking these questions, and try to become a connoisseur of simple sequences of notes – collect them and know how you feel about them.


Learn to love cadences, standard chord progressions, and repetition.

In the years when I was stuck at composing, one of my self-imposed obstacles was that I felt the only music “worth” writing was music that sounded new, music that was original and somehow distinct from anything I had heard before. We live in a time when artists are often praised for being “revolutionary” and rarely praised for being excellent craftsmen in traditional forms; often praised for being “creative” or “original” and rarely praised for producing work that gives great pleasure without boasting of its own radical innovations. I found it very freeing when I finally decided to try a different approach. I sat down and aimed to write music that would sound familiar, music that would make me think I had heard it before. In time I found that the surest way to create something that sounded “new” without simply being random was to start with something familiar and then tweak it in subtle ways. At the same time, the surest way to stifle creativity is to be afraid of the familiar and the commonplace. My view is that the best thing you can do as an artist is to experiment; but, to require that the outcome of an experiment should be something new is to conduct a biased experiment that may not be helpful for your creative purpose.

Learn to love simple, straightforward, tonal cadences; even if you don’t include them in your finished product, learn to work with them, and see how a simple cadence affects your interpretation of all the music that has come before. Play around with simple variations on the I-V-I chord progression, and don’t think of it as a trivial exercise, but rather as an opportunity to harness one of the most powerful musical resources there is. And, when you’ve crafted a short passage, see if it can be profitably repeated.


Write etudes.

Whenever you’re intrigued or confused by a musical idea, try writing an etude that explores it. The idea behind an etude is that it can be simpler, shorter, and more mechanical or repetitive than you might expect from a full piece, but it should still be musical. If you’re stuck working on a small passage within a larger piece, see if you can turn that small passage into an etude of its own. If you’re studying another composer’s score or reading a music theory text and you come across a harmonic progression you don’t recognize, write an etude that explores only that progression. Feel free to leave your etude in its barest form, but know how quickly a “toy piece” can become a “real piece” once you add some ornaments and variations.

When I made the transition from someone who dreamed of composing to someone who composes, I did it by taking a month and spending all of my free time sitting in front of a score editor writing etudes. At first I had to force myself to do it but eventually I became addicted to it in a good way. My first etudes were really just little melodic fragments, sequences of quarter notes extending for a few bars: I spent hours and hours just seeing what I could do with a couple of notes from the major scale arranged in a sequence with no rhythmic variation. I moved from there to seeing what I could do with a few accidentals added here and there. And from there I began trying to craft sequences that modulated in a convincing way, or that employed simple rhythmic contrasts. And from there I started writing arpeggiated sequences that repeat a harmonic progression through the circle of fifths or in some other cyclic fashion. And from there… well, I’m still doing it. ■

Comments ༄