4 Different Ways to Start a Musical Sentence

+ Take your modern jazz piano and hip-hop beat making to new heights with Soundfly’s new course, Elijah Fox: Impressionist Piano & Production!

When you’re composing music, you kind of have to split yourself into a bunch of different personas. There’s “you, the artist” who contributes all the emotive content; “you, the craftsperson,” who gets anally retentive over just how you’re going to translate the emotive content in such a way that another human being can interpret it with their instrument; and “you, who just writes stuff down.” That last “you” is not particularly glamorous, but sometimes, especially in the early stages of a piece, that “you” is the “you” who has to drive the bus.

Both the artist and the craftsperson are vital to the composing process and are responsible for making your pieces unique, performable, and satisfying. At the very beginning stages of a piece, however, these unique personas can be a liability.

They’ll hover over your shoulder and say things like, “Oh gee, is opening with an F minor chord really the best way to convey the heartbreaking tragedy of the Great Bunny Massacre of 1906?” or, “Hmmm, I’m not so sure that’s exactly how you notate quintuplet 64th notes. Perhaps you should spend 30 useless minutes trying to come up with the exact phrase that enables Google’s algorithm to locate some obscure article about it. You know, just to be 100% sure.”

They’re like the eager-to-help-but-vexingly frustrating roommate who keeps telling you how you’re making pasta sauce wrong. If you don’t get them out of the kitchen while you’re cooking, they’ll take over and ruin everything, and then everyone is going to be eating peanut butter on toast for dinner.

At the early stages of composing, you need to just let the music happen. You have to be the “you, who just writes stuff down.” Some people call this listening to the muse, while other people call it following your gut. I call it getting out of your own damn way.

And it can be tricky to do. So, to help you out, here are four different tactics you can try to keep the fussy “artist you” and the busybody “craftsperson you” occupied for a couple of days while the stolid, dependable “you, who just writes stuff down” works.

A brief aside…

If you’re looking to learn how to “finish a musical sentence,” well we call that a cadence. And we just published a video about it on our YouTube channel.. It’s…pretty rad. Musicians, composers, and curiosity-junkies, check it out and subscribe if you like that sorta thing!

1. Limit Your Parameters

There’s no denying that composing is a difficult thing to do. There are a lot of different elements to think about. Sitting down and saying, “I’m going to compose a piece of music” is as useful as saying, “I’m going to write a novel.”

That’s all very well and good, but unless you have some idea of what the book is about and how you’re going to tell the story, then you’ll be sitting there for a very long time with not much to show for it.

You have to limit your parameters to take some of the choices away. By doing so, you’ll free yourself up to experiment more deeply within the leftover choices and hone in on a whole host of different elements. Just a handful are:


What pitches will you use in your piece? Is the piece major, minor, or something else entirely? Is it a mixture of different kinds of tonalities, or is there a clear harmonic progression? Stuff like that.


This covers a lot of stuff, including things like whether the piece is fast, slow, or changing; if it contains multiple time signatures; or if it’s rhythmically erratic or predictable.


What instruments are playing the piece? How can you blend them to create a new sound? Are they playing loudly or softly? Do the instruments perform in a conventional style, or will they be using extended techniques? Are they playing at the extremes of their registers or in the middle?


How are the instruments working together to create the overall aural texture? Are they playing together in unison, or are they all playing their own lines? Are they playing jagged lines that jump around a lot, or smooth lines that move gently?

There are endless questions to ask yourself, but if you start setting some rules, you’ll get somewhere quicker than by declaring, “Today, I am composing.” And once you get started, stay out of your way.

The goal is to get an idea down that you can then go back and edit as critically as you like. Sometimes you’ll nail it in the first round. Other times, it’ll take a few rounds to find the confidence to jot something down.

2. Improvise

Alright, maybe theory isn’t your strong point. Maybe limiting your parameters sounds about as creative as drawing a connect-the-dots picture. Perhaps you just like to dive in and get into it. Improvisation is probably for you.

Improvisation is a totally legitimate way to get started on a piece (and to continue, finish, and perform a piece for that matter). The advantage of improvisation is that you can quickly develop the broad strokes for how your piece might sound, then go back and concentrate on the finer details later. The disadvantage is that musicians tend to fall back into old patterns of playing their instruments when they improvise, as opposed to setting out with composition in mind.

That’s why it’s always a good idea to record your improvisation. In the heat of the moment, you might play something perfectly, and then totally forget what you did. When I do it, I play my MIDI keyboard into a DAW, which gives me the added flexibility of actually cutting bits of the recording out, looping them, and making minor adjustments to things to keep the experiment going.

One of my most successful pieces I started through improvisation was for four cellos, an instrument I play about as well as I can fly. I tried noodling around on the piano, but a piano doesn’t have the smoothness of the cello, which is the sound I wanted to explore.

So, I tried something else instead. I improvised by singing into my DAW. By singing, I was able to recreate more closely what a cello could do, and therefore, get a better feel for what it might sound like. Here’s my short, rough singing experiment:

And now, take a listen to the first minute of the piece I wrote, “Traveler” (played here by Ashley Bathgate on cello and recorded by Yoav Shemesh):

They’re pretty similar, right? There’s no way I would have come up with something like that if I’d simply limited my parameters or tried improvising on the piano.

+ Learn production, composition, songwriting, theory, arranging, mixing, and more; whenever you want and wherever you are. Subscribe for full access!

3. Leave It Up to Chance

One of the most influential musical experimenters of the 20th century was John Cage. Cage made many useful contributions to advancing contemporary composition, but perhaps the one he’s most famous for is aleatoric music, also known as indeterminate or chance music.

Like the word “chance” implies, Cage used randomized processes that aimed to remove himself from the compositional equation. Think about it — even when you improvise, it’s still you playing it, with all your habits and influences, so it’s not entirely random.

Cage would do things like hold a piece of blank manuscript paper up to the light and fill in musical notes where he saw imperfections on the page. He would use the ancient Chinese fortune-telling technique of I-Ching, which involves throwing a handful of yarrow stalks in the air and interpreting the shapes they create when they land, to inform his compositional decisions.

Anybody can use chance processes as a way to get some ideas down on paper. Before you get started, come up with a system. Here are three I just came up with:

  • Roll two dice; each number you roll corresponds to a pitch.
  • Write down a bunch of small motifs on paper, then randomly draw them out of a hat to string them together.
  • Pin some manuscript paper to a dartboard, and throw darts at it.

While Cage was sometimes purist about his random music (to mixed reactions), I suggest that you be a little more pragmatic and try to smooth down the gibberish you’ll likely end up with.

Once you’ve done the random thing, go back and and summon “artist you” and “craftsperson you” to get the piece into some kind of sensible shape. Now you’re starting with music that you’ve already begun, rather than trying to do it all again from scratch.

4. Use an Objective Process

This isn’t much different from Cage’s idea of leaving it up to chance (although the more hardcore composers who followed such processes back when they were in vogue would have a conniption if you dared suggest such a thing).

The idea is to set up some kind of mathematical or procedural operation that “decides” certain aspects of the piece for you, systematizing the compositional process. This removes that pesky “artist you” from getting in and messing everything up with all that emotion.

One of the most well-known objective process is serialism, which hit it big between the World Wars. There’s also its more extreme offshoot, total serialism, which ruled the art-music world with an iron fist for about 10 to 15 years after World War II.

Serialism (or dodecaphony or 12-tone music) was a process created by an Austrian composer named Arnold Schoenberg in the early 20th century. Basically, the way it works is you make what’s called a 12-tone row. Once you use a pitch, such as D, or F, or B♭, you can’t use it again until you use all 11 other pitches.

Once you’ve made your tone row, you do all kinds of weird mathematical things to it, like reversing it, inverting it, transposing it, and eventually, you end up with a whole bunch of different tone rows as the basis for your music.

tone row

If you’re sharp, you might already see that this system doesn’t completely remove the composer from the process of composing. The composer still needs to decide how the tone row will sound and how the tone row will be used in the piece. In fact, really all it does is try to stop the composer from writing a piece that sounds “major” or “minor.” Everything else can be as conventional as you please.

This brings me to the total serialists, a group of composers who wrote in the years after World War II. They admired the serialists (well, one in particular, Anton Webern, one of Schoenberg’s pupils), but felt they didn’t go far enough. So they invented total serialism.

Where serialists just applied processes to the pitches in their pieces, the total serialists applied objective processes to decide how to use virtually every element they could think of in their pieces, from rhythmic units, to dynamics, to register, to articulations.

Today, total serialism has fallen out of fashion, but for a while it was the only way to compose if you wanted to be taken seriously. Which is ironic, because total serialism is sometimes indistinguishable from completely random music.

This attitude kind of blew up in the total serialists’ faces though, because the music is sometimes quite… difficult to listen to. Unsurprisingly, it tended to alienate audiences. In fact, I would suggest that the reason the general population is wary of new music concerts is because of some deep, underlying fear that they’re going to be stuck in a room with a pianist going “bling blong blam” for two hours.

+ Read more on Flypaper: Was Arnold Schoenberg actually the godfather of techno music?

This is not to say that you can’t use some of these ideas. I still encourage my students to try 12-tone music if they’re stuck getting started. Here’s how:

  • Write out a tone row, then see what happens when you make a melody out of it.
  • Stack the notes to create chords and harmonies, and write a melody to go over the top.
  • Select little sections of it, and play it over and over to create an ostinato (a repeated pattern).
  • Or use it as a bass line.

In fact, if you want to see just how effective serialism can be, then check out this absolutely awesome video by Vi Hart. It’s so great. I wish I made it!

Next Steps

So now what? What do you do once you’ve started your musical sentence?

That’s when it’s time for “artist you” and “craftsperson you” to get to work. Be critical. Just because you’ve spent an hour working on something doesn’t mean it’s good to go. Ask yourself: Do I love it? If you don’t, then figure out why you don’t love it. Then tweak it. Don’t throw it away just yet.

Sometimes just changing one or two little details in an original sketch will transform it from cacophonous rubbish to something beautiful and unpredictable. And the great thing about composing is that you don’t have to come up with a genius idea every single time. You only have to do that every now and again. Once you have a rock-solid idea that you love and that inspires you, you’ll be surprised how quickly other ideas start to flow.

Don’t stop here!

Continue learning with hundreds of lessons on songwriting, mixing, recording and production, composing, beat making, and more on Soundfly, with artist-led courses by KimbraCom TruiseJlinRyan Lott, and the acclaimed Kiefer: Keys, Chords, & Beats.

Com Truise: Mid-Fi Synthwave Slow-Motion Funk

Join our Mailing List

We offer creative courses, articles, podcast episodes, and one-on-one mentorship for curious musicians. Stay up to date!


Laura Mvula’s Lyrics Have No Business Being This Good.

Resistance, resilience, realness. Laura Mvula uses her lyrics to weave complex stories told with accented, melodic, and tonal perfection.


Is “If I Fell” The Beatles’ Hippest Pre-Revolver Tune?

What’s going on in this subtly sophisticated pop hit? Turns out… a lot! Let’s break down this romantic ballad off of “A Hard Day’s Night.”


The Modal Mixture at the Heart of Elijah Fox’s “City In the Sky”

Elijah breaks down the parallel modal patterns that make up his track, “City in the Sky,” which imbue a “rocking back and forth” quality.