This week, a colleague was reading a unit plan I was working on, and noticed the phrase “practice improvisation.” She immediately pointed out to me that improvisation can’t be practiced, and to put those two words together makes an oxymoron. I was not convinced and still am not convinced that this is so. What is true is that once a musician starts practicing something he or she has already improvised, working on it to play that particular idea better, then that is no longer improvisation, and the student at that point is practicing a melodic line or chord change. But the artistic sub-process of improvisation can be practiced just as composing can be practiced. Practicing is the act of doing something, evaluating the results, deciding on what change/improvement to make, and then doing the thing again with the change/improvement, and repeating the process until the thing is done well enough to meet a performance standard.This  three part cycle of perform, evaluate, plan can certainly be applied to improvisation.

Before a person can improvise, he or she must first have a musical idea in mind. We are accustomed to having thoughts in the form of words, pictures, and feelings, but having thoughts in music is less familiar to most of our students, and perhaps to us as well. I consider myself a good improviser even though I am not a jazz musician.This is because there is nearly always music running through my mind. By this I do not mean that I often have a melody “stuck in my head,” but instead that I am frequently just humming away at something. If someone asks me what I’m humming, I often have no idea, or else I am aware that I am just “making stuff up.” But I’m not really creating music from scratch, I’m playing with musical ideas I’ve heard before, or that sound similar to musical ideas I’ve heard before. I’m singing things that come  to mind out of a wealth of musical patterns, and motifs that I have heard and remembered over years of listening to, singing, and playing music.

When I was in high school, I enjoyed going down to the music wing during my study halls, and just sitting at the piano and making up little bits of counterpoint. I loved the inventions and fugues of J.S. Bach, and wanted to imitate his music. I was aware then and am certain now that nothing I ever improvised came remotely close to actually being mistaken for a Bach fugue, but it didn’t matter to me. I enjoyed thinking up some little phrase, and then playing it with my other hand while trying to keep some kind of counterpoint going. My point in telling you this is that in both cases, while humming and while making up counterpoint on the piano, I was practicing improvisation. It wasn’t even an intentional thing, where I decided I was going to practice improving, but that was in fact what I was doing. I was putting my store of learned bits of music to use to generate new bits of music, and stringing them together into a cohesive melody that resembled music of the culture within which I had lived and with which I had become familiar through years of listening, humming, moving, and playing.

It is possible to turn practicing improvisation into something else. This happens when music theory is allowed to drive the choices a musician makes in the guise of improvising. If I am consciously selecting notes based on whether or not they belong in this chord or that chord, and I am not calling to mind patterns I have already learned through listening, then I am not really improvising. The reason is that if I am just choosing chord tones as I go along, I don’t really know what motif or phrase I’m going to play ahead of time. I’m truely making it up as I go along, and creating something that has little if any musical meaning to me, because I cannot group the notes into phrases until after I have already played them and only if I can remember what I played. It may sound like improvisation to a listener, but the action I am taking as the musician is unmusical and fragmented. It is as if I am reading by only seeing one word at a time, and with no prior knowledge of what the material I am reading is about. Every word is a surprise, and cannot possibly be read with fluency or expression because I cannot know how any word fits into the context of the sentence until I have already read it; until it has been revealed to me at the moment of reciting it. We don’t think that way. We need a complete thought to find meaning, whether it is in language or music. So improvisation must be the performing of complete musical thoughts, at least a completed motif, with the context known before it is played or sung. Chord changes provide the context that gives the motif or phrase meaning. Their function cannot be corrupted into selector of notes on the fly.

Music theory is of great value in understanding and studying music that has already been performed or written down. There is much to learn from studying transcriptions of solos of great improvisers. We can see how one idea generated another, and learn a great deal about the improvisers musical mind by studying these. But this is apart from actual improvising. This is learning about music, which is a different matter from learning to make music. As with developing music literacy in general, theory can never be allowed to drive practice, but rather theory must be used to explain practice. Out of the stores of a person’s musical memory will come improvisations informed by common cultural practice.