Composing Solos – or How to Cheat

I have often found it a very useful exercise to sit down and precompose a few choruses of soloing over a tune that’s captured my imagination at the time.

It’s pretty obvious to say, really, but improvising is ultimately nothing more or less than instantaneous composition. Since we don’t have the luxury of time when improvising, we have to rely on what our subconscious has learned during practice sessions, as well as using our ears, muscles, emotions, liver (?) to guide us. By composing solos, we’re actually practising the process of soloing. In sloooow motion.


So what can we gain by sitting down with our instrument, manuscript paper and a whole afternoon to produce 32, 64 or however many bars of music?

(Incidentally some would argue it’s best not to use pencil and paper and do it in your head. I agree up to a point – but the thing is that writing it out allows you to plot out ideas over many choruses and also means that you have a record of it for later. Plus you’re improving your reading and writing skills – never a bad thing.)

By precomposing solos, we are of course exploring the possibilities at complete leisure and discovering our own store of favourite ideas, some of which we go on to use quite deliberately. These ideas become a part of our musical personality, and every great jazz musician has them.

Nevertheless, I think the best way to apply this approach is to “fire and forget”. Compose the solo, practise it, make alterations and improvements as you go along. Then keep it on file for later reference – I’ll often revisit my precomposed things when I feel my playing is getting stale. Incidentally, your stock of precomposed things can be very useful when it comes to writing arrangements, countermelodies, solis, etc. Bits of them often appear later on in your own compositions.

But don’t try to learn solos and reproduce them note-for-note on gigs. The idea is to practice the process of composing a solo, rather than learning great big gobbets of music to trot out later.

Don’t be concerned that this means composing solos is a waste of your valuable time. You’ll find, if you take this “fire and forget” approach, that certain ideas you precompose will appear more or less spontaneously in your playing anyway – often altered in interesting and unexpected ways. There’s no need to force it (in fact, forcing things is a very good way to guarantee that things go wrong in performance!)

Rossini was famously bad on deadlines. Impressarios used to confine him and hire runners to convey each page to copyists as he threw them out the window.

Famously bad on deadlines: Impressarios used to confine Rossini and hire runners to convey the music to copyists, and thence to the opera house, as he threw each page out the window. He also, shall we say, recycled ideas – a lot.

The other reason not to precompose with the aim of reproducing the solo verbatim is to take the pressure off yourself. It’s much easier to compose music in the spirit of interested exploration than feeling that you have to come up with a work of art against the clock. Deadline pressure may have worked for Rossini, but give yourself a break. Anyway, there was a man who had a really big store of precomposed ideas…


Having said all that, I am certainly not the first person ever to sit down with a tune ahead of a gig or recording session and come up with a few general ideas to plug in later – particularly at the tricky or unfamiliar bits.

Actually, I’ll sometimes set part of my brain to considering a few ideas for my piano solo while comping the sax. It’s also possible to disguise trying out a few concepts in the comping.

If you listen to the more recent reissues of classic jazz albums you’ll hear a surprising amount of duplicated material between alternate takes. A lot of Charlie Parker’s solos, for instance, definitely seem somehow pre-planned, although he does use his “homework” flexibly. A lot of the greats have admitted the existence of “practised improvisation” (as Bill Evans’ bassist Chuck Israel has described it).


To a certain extent, composing a solo is the ultimate in slow practice. By taking as long we like to come up with a good solo, we are spending considerable time consciously applying the principles of good solo construction. Every time we do this, our subconscious mind is sitting at the back of the class, gazing out of the window (as is its habit) – it’s listening, but not attentively. The subconscious is catlike – it can pick up habits, but it can’t really be trained. The idea is to drip-feed our subconscious with good principles.


By the way, this process has a lot in common with transcription. In particular, you have the luxury of being non-linear. When transcribing, we aim not to get bogged down on a passage that is tricky, hard to hear or just not very interesting. Similarly, when composing a solo, we can just leave certain bars blank – to be filled in later or just to be improvised over when we’re playing through the solo.

How do you do a jigsaw? Build the frame then look for interesting bits.

How do you do a jigsaw? Build the frame then look for interesting bits.

I recommend writing out chords above empty bars for the whole solo before you begin (writing four bars to a line). This allows you a bird’s eye view of the tune’s structure.

Of course, the basic approach is to proceed sequentially, by writing a phrase then developing, extending, transposing it, altering the rhythm, inverting it, etc.

A more interesting approach is to start by writing some things in at preplanned high and low points – you can then distil more general motifs from what you write at these points and work up to them. You can actually go so far as to fill in an entire extended solo patchwork-fashion.

How useful is this approach in preparing for real life improvising? You’d be genuinely surprised how the non-linear subconscious can work towards a future point in the solo.


All the following concepts are used instinctively by good players. Getting to know how to use them by composing solos is a good way to sink them into your subconscious mind.

  • Overall shape. Tension and release, climax points, playing across the obvious phrase breaks and section breaks.
  • Phrase lengths. Satisfactory balance between long and short phrases, or development by starting with a short statement, then gradually lengthening it.
  • Mature use of silence. Most beginners don’t leave nearly enough space.
  • Development of simple motifs. Usually the most flexible structures to work with are quite short: a melodic motif of no more than five or six notes, a rhythm with no more than five or six “hits”. These can then be adapted flexibly to provide an underpinning of structure to the solo.
  • Sequencing scalar patterns across changing chords for a continuous line.
  • Balance between held notes and lines, high and low, soft and loud, quiet and busy.
  • Quotes from the tune, or playing close to the melody (and quotes from other tunes and solos).
  • Varying different triad qualities from each chord-scale – this is particularly valuable in a modal context.
  • Use of melodic “cells” – a triad plus any other note (eg 1235, 1345, 3567, etc)
  • Varied and balanced use of scales, arpeggios, guide tones, fourths, pentatonics – all enriched and extended by using passing tones and bebop-style enclosure figures.
  • Ad-hoc reharmonisations: tritone substitutions, pedal point, modalising the harmony by treating part or all of a section as just being, for instance, “in Eb” rather than chasing the changes.
  • Use of blues-scale gestures or just “blue” thirds.
  • Use of upper-structure triads.
  • Distinctive figures from melodic minor scales over many different chord types.
  • Playing call-and-response with yourself.
  • Playing inside, then outside, then back again. Use of sequences or chains of outside triads are useful to maintain a sense of structure against the underlying harmony.
  • Repeated rhythmic vamps.
  • Overlaid time feels – implied half-time, double-time, superimposed threes on twos or fours, etc.
  • Effects, atonality, noise.

There are two basic ways to generate ideas (two sides of the same coin, actually):

  1. You can start from the ear. Just play around over a section of the tune until you hit on something you like the sound of. Then analyse it to find out what’s going on, from a theoretical standpoint. It is the shape of the line, the combination of chord and non-chord tones, the rhythm or a combination of all of these? Once you have a clear idea of which concepts are at work you can then consider how to apply, extend or counterbalance them.
  2. You can start from theory. Maybe you choose a certain melodic cell or descending arpeggio and try it out. Does it sound okay, too dry, too theoretical? Okay, then embellish it – add a pickup, change the rhythm, insert passing tones, invert it, etc. Then consider how you might want to extend, or contradict what you’ve come up with.

The process of composing solos should be interesting and stimulating. The word “play” has two meanings. Enjoy yourself!

Tagged with: , , , ,
Posted in a) Soloing Scales & Modes
6 comments on “Composing Solos – or How to Cheat
  1. Adam Cole says:

    I love this. I’ve been wrestling with the “shoulds” and “shouldn’ts” of precomposing for years, and you’ve solved the dilemma…write it but don’t play it verbatim. I think Nat Cole was famous for precomposing, wasn’t he?

    • Jason says:

      Don’t know about Nat Cole, but I know quite a few singer-pianists who play pretty much set arrangements and solos (even if they’re good improvisers). Makes sense really – if your focus is on a vocal performance, it’s understandable to take a more classical approach and largely “automate” the piano part.

  2. I’m a classical flutist, but my favorite music to see live is jazz! I’m pretty good at Baroque improve, but don’t have the right language in my fingers to do much with jazz. I took an excellent one-on-one class in grad school – Jazz Improv for Classical Performers. The professor was a pianist and got me started by having me compose solos in the jazz style (several different ones per piece), then play them with him in the lesson. After a few weeks, he made me “write” one on the spot. It worked! I haven’t kept up with it, but it was a great way to get a classical player to improvise!

    • Jason says:

      Hi Trish,

      I do know what you mean. I came the classical route myself, and the first jazz forays were perplexing. There’s the harmony, just play what you like on top. Yeah, but what? Anything you like. Yeah, but what?!

      Formal analysis and composition is a bit neglected in the classical field, or at least regarded as a specialism that you should aspire to only after you’ve become an accomplished performer. Shame, really. For one thing, the best sightreaders usually apply a fair dollop of, shall we say, inspired generalisation… And an insight into WHY a composer might have written what they did can really bring to life the way you perform WHAT they wrote.

      I had a big concert piece performed a few years ago, and the orchestral parts were completed in a crashing hurry (as always). During the breaks in rehearsal I slunk round the desks and chatted about the parts, encouraging them to approximate within the framework if necessary (but don’t tell the conductor). They really took to it.

      Thanks for dropping by,

  3. Hannah Wilson says:

    You did a great job! This post sound really nice.
    I saw something like a couple of weeks ago, something similar to this, but you did
    in-depth research, and your article appears to be more persuasive
    than others. I am amazed by the arguments you provided as well as the style of
    your post. I like when articles are both informative and interesting, when even boring facts are presented
    in an interactive manner. Well, it’s surely about your post.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Books for Sale
...appetising young books for sale... Pents book is recommended reading on Gary Burton's Berklee course.

%d bloggers like this: