Why we improvise in classical music

This past week I was teaching a ten year old clarinet student. We have been working together for roughly 6 months and we are starting to get to know each other. We’re finding out where her strengths and weaknesses are. What concepts are easy to grasp and which ones are far more difficult. She is a concrete thinker. She likes boundaries, rules, techniques, and absolutes. With that in mind, I often find myself trying to teach the abstract and the more artistic sides of music in her lessons. I find that I am often asking her how she feels about the music we listen to in lessons. I ask her what she’s hearing in her own playing or in a recording and what those sounds mean to her. I know that she will easily grasp and put into practice everything I say about embouchure or tongue placement. She will use lots of air when I ask her to, and she will learn her key signatures with just a few weeks studying.



This weeks lesson was focussed on listening and putting that listening into practice through an abstract exercise. We began by listening to Eric Whitacre’s Lux Aurumque. I wanted to choose something that had sounds she was not used to so we listened to the choral version of Lux. It has a good mix of both sparse single or two line sections as well as colourful and dense chordal passages. We listened to it once and I asked her to describe what she heard. She answered, in predictable character, with absolutes. It was quiet. It was high and low. There were people singing but no instruments. Sometimes there was only one person singing and sometimes there were many.



    As a second step, I gave her a blank sheet of 11x17 paper and a pencil. I asked her to listen again and draw whatever she felt. Whatever shapes, lines, designs, or patterns she felt represented the music. We listened. When the piece finished, she showed me the paper. At the top left corner she had started a straight horizontal line. This line carried on with no bumps or deviations down the page like lines on a sheet of looseleaf. I asked her to point on the paper as we listened again. To show me what she drew for each section of the music. Of course, she couldn’t point to anything specific based on the music because there was no change in her line.



    We listened again. This time, I asked her to listen closely to the music. How it changed. What types of different sounds happened and when. Then, based on that, add to her line any shapes and squiggles that worked for her. Very tentatively, she drew in some small squares, a few triangles, and one circle about the size of a pea. From here, we grabbed clarinets. I put her newly written score in front of us and told her we were going to play what she had written. This of course baffled her. For the first time she was being asked to create; to improvise and interpret a concrete from an abstract. We began to play and as might have been expected, she played one note, held, for as long as she could. No variation. No reflection of the piece, or her drawing in her playing. This clarinet version of interpretive dance was confusing to her. The circuit was, somewhere along the line, missing a connection.



    So, I took a step back. I added back in a couple boundaries. I asked her to name and then play all the notes she knew. As she is a very technical learner, she knew two full octaves from low F-F as well as a few sharps and flats. For someone only 6 months into playing, that is fantastic especially given that several of those notes are in the higher register usually reserved for the second full year of playing. We wrote down on the bottom of her score all the notes she knew. We took a moment to play a mimic-ing game. I would play something slow and low, and she would mimic. I would play something frenetic and high, and she would mimic. I asked her to play her score using many of the notes she knew, and she played 2-3 notes for long sustained sections with little to no movement or activity.



    Almost four years ago, I was living in Ottawa running a new music focussed production company. We would program new works from composers every concert and perform with a variety of ensembles from choirs to orchestras, trios to soloists. For a concert based on opera mad scenes, we commissioned a wonderful composer from Washington DC named Ryan Keebaugh. We commissioned him to write a completely graphic score based on five texts we sent from five different operas. He sent us a set of 5 movements to be played by full orchestra. We had assembled an orchestra of players from the University of Ottawa masters program, members of the Ottawa Symphony, and freelance players who all play with some fantastic groups. We put the graphic scores in front of them and began. The sound that was created came in two forms. Either the players had no framework in their heads to begin to play so they began playing variations on excerpts from the operas they knew the score was based, or; they sat listening and waiting for more instruction or boundaries from me, their conductor.



These players who had several degrees, had been playing as professionals for many years, and who had a wide breadth of musical understanding were finding it difficult and in some cases impossible to create music without a strict and overarching set of rules guiding their playing. Obviously those musicians with jazz and improvisation training are comfortable creating and feeling music for themselves and interpreting any number of auditory and notational cues to make the music you hear. Likewise, there are new music ensembles that focus entirely on creating music from hand signals, graphic scores, and any variety of other improvisation tools. For some reason though, this training and understanding of music is limited to a very niche musician. There are any number of pieces for school band that feature aleatoric or improvised sections. It has even become fairly standard in the wind world to include these features in even high level compositions. Somewhere along the line though, the intent of why we play this music is getting lost. Why do we improvise?



Eventually, through much discussion and experimentation in playing, the ensemble was able to connect to the work and feel the emotions and sounds they wanted to present. They began feeling what each other performer was playing and interacting with their fellow musicians to create a truly beautiful, complex, and chaotic sound. There were phrases, ebbs and flows, and moments of joyfulness and deep sadness, madness and clarity. However, It took far less time for us to rehearse and put together a full Hindemith Symphony than it did for us to learn to improvise 20 minutes of music based on a graphic score.



In both cases however, I am left with the same questions. Why is it difficult to improvise? In both of these cases, there are no boundaries - it should be the easiest form of music making. No rules to follow and no wrong notes. Is it because for most of a classical musicians life we are given very detailed instructions for every possible element of the music? Is it because that type of improvisation training is not explored in our classical education correctly? Is this type of music is not prevalent and therefore little known to players and listeners?



I always wonder then if an equivalent challenge presents itself in other artforms? I would hazard a guess that if I asked a painter or dancer who has been working for one year to paint or dance a representation of “angry” that they would be happy with that challenge. If I ask a player of one year to play “angry” for some reason that presents a much harder challenge. Is it because improvisation and creation from an abstract thought are inherent in other art forms? Is it because dancers, painters, and writers are intrinsically connected with the creation of their art from an abstract thought? Have we been coddled as musicians to the point where anything we play needs to be very well defined, presented to us in a form we expect and have been taught?  



The next question should probably be, why does improvisation in classical music matter when clearly it is more efficient, standardized, clear, and reproducible to play from a very heavily notated set of instructions? We as classical musicians are not trying to be jazz musicians. We have scores that have hundreds of years of performance practice attached to them. If we are not playing new music then we know that whatever we are presented with will be standardized and we will not have to create the music from scratch. It will be there on the page for us.



A week ago, the same week that my young clarinetist was having an existential struggle with hearing and playing, I was listening to CBC radio play a young Canadian piano prodigy play the Goldberg Variations. They played with technical skill, accuracy, and with musicality. However, it was not their performance. The ornaments were Glenn Gould’s. The tempos were Barenboim’s. There was nothing that made this performance their own. What makes a performance like Gould’s 1951 astounding is the individuality with which he played. The phrasing and understanding of counterpoint were all his. He had taken his abstract ideas about music, his self-discovered thoughts on phrase and line, and imposed them on the concrete score of Bach’s.



So finally, my clarinet student and I came to the end of our lesson. We left the drawing and playing exercise to the side and took out a piece of music that she has been working on. A wonderful piece from the red Standard of Excellence book, for those of you who know the system! She played it through the way she has for the last couple weeks. Fairly fast. Two bar phrases. Articulated as it was on the page. Then, I asked her to make two changes to how she played it. Tempo, articulation, dynamic, repetitions, or anything else. She played half the tempo, legato, and with longer four bar phrases. She had created a unique and individual understanding of music that was not based on what was written on the page. She had played something that she thought about and created through improvisation rather than simply pushing the right buttons at the right time. It will take a long time for her to consistently think in that way and it will take many more experiments to broaden her understanding of what she’s playing.



I think though that this is a good lesson for all of us no matter the stage of our musicing. What real-world applications can we learn from improvisation to inform how we approach classical music? How can our confidence in listening and creating in the moment actually change how we interpret music, phrase, and direction? We can be better at connecting those nebulous and abstract ideas about interpretation and improvisation to real-world and concrete musical moments. We can teach better listening skills through improvisation that don’t need to be based only in jazz or new music. That is why we improvise in classical music. 

Jacob Caines