3. Composition and Improvisation

Composing for an ensemble of free improvisers.

A second paradox:

Aren’t composition and improvisation diametrically opposed?

”I would like to define improvisation as denoting the spontaneous element in musical performance, which either takes place within some kind of implicit or explicit framework or (as in ”free improvisation”) creates and transforms that framework as it proceeds. I would define composition as any kind of musical creative process or the results thereof. Therefore, within this scheme improvisation is a method of composition, no more and no less.”  (Barrett 2014: 107)

In the quote above, Richard Barrett goes on to say that his definition of improvisation as one way of composing, makes it clear that the two ways of creating music are in no way in opposition. Thus, composition can mean ”making music” and improvisation is a method for making music, in a spontaneous, real-time way.

Then, if ”composition” means ”music-making” and ”improvisation” means ”spontaneous music-making”, what is a useful word for the other main method of composition: ”Planning and notating how to make music in advance and have it executed at another point in time (possibly by musicians)”?

”Predetermined musical structuring or material” feels like a useful definition for me.

This material, which is usually notated in some way, is normally more or less similar from performance to performance, whereas free improvisations can, maybe even should, be very different.

I agree with Barrett’s definition, but there are some fundamental differences between what we normally associate with composition (predetermined structuring or material) and improvisation:

Improvisation is an ongoing dialogue, and is usually based on communication from the very moment it starts, with other improvisers and the audience. Composing music on paper is usually a solitary process until just before it is performed. There may be communication with the players and the composer in advance, and also when rehearsing the music, but the main form of communication is verbal or literary. In improvisation one communicates via musical sounds.

Composing predetermined music is also a process which takes place out of time, and subsequently the composer can work with time in a different way than the improviser: The improviser works with spontaneity and intuition in real-time, and can not change something which has been played. The composer, on the other hand, can plan the length and timing of every event in a musical structure if she wants to, change her mind and move sections around as many times as she’d like.

In my ensemble, and within me, there is a constant dialectic between the two ways of composing. They influence each other and when I can find a balance between the two, sometimes it results in successful music.


Barrett also mentions improvisation within an implicit or explicit preexisting framework in his article (Barrett 2014). An example is a jazz standard, where a rhythmic and harmonic framework is improvised over, somewhere on the continuum between the completely improvised and the completely predetermined.  This notion of the ”referent” is something I will return to later in this text, and has been a useful tool for me in my work with the ensemble.

A second Continuum : Between the Improvised and the Predetermined.

In the article ”Generating and Organizing Variety in the Arts” (Eno 1981), Brian Eno describes a ”scale of orientation” or continuum from right to left between composition ”tending to subdue variety in performance”, i.e. predetermined music, for example prerecorded electronic music on the extreme right, and composition ”tending to encourage variety in performance”, i.e free improvisation, on the extreme left. In between we find all the facets from classical music, various folk musics to jazz, free jazz etc.  Even free improvisation is rarely completely invented on the spot, but influenced by a host of different factors.

I imagine my own work as a composer in this project as oscillating along the continuum between the completely improvised and the  predetermined. This movement back and forth, with varying degrees of determination, is noticeable in the way I have developed the music throughout the project period. It can also sometimes be seen within each piece, where some sections will be more predetermined and some less. Sometimes the pitches and rhythms can be notated, sometimes the music is described by text and sometimes completely improvised. The music moves back and forth along the continuum on a gliding scale, both within a section of a piece, a whole concert, and throughout the project period.

I have tried to discover how to balance my own material (the predetermined) with the material of the improvisors (the spontaneous), finding the optimal point on the continuum for me and the ensemble, where I can both express compositional ideas and we are able to improvise freely at the same time. The music should feel comfortable and inspiring for the improvisors playing the music and myself as the composer, while yielding ”successful” musical results. This optimal place for music making with an ensemble of improvisors is probably a utopian notion, but the search is a constant motivation for developing the music and methods. An ideal combination of what I would like to create and hear and how I would like to create it.

In a way, trying to find this possibly illusory/utopian point is a balancing act of ”everything”: My own tacit and explicit influences from many different sources, what kind of music I want to make both as an improvisor and a composer, the personal ”webs” of the musicians, letting their musical voices be heard, the social situation of improvisation in an ensemble and so on.

Algorithmic vs Heuristic

Brian Eno also describes the difference between what he calls ”algorithmic”  and ”heuristic” ways of organizing a composition (Eno 1981).

The ”algorithmic” way of organizing he describes as ”a rigidly ranked, skill-oriented structure moving sequentially through an environment assumed to be passive (static) toward a reso­lution already defined and specified.” For example a classical piece where everything is moving according to a clear logic.

”Heuristic” he defines as “a set of instructions for searching out an unknown goal by exploration, which continuously or repeatedly evaluates progress according to some known criterion.” This is a more common way of organizing experimental music, and it encourages autonomous behaviour and self-organization, as opposed to the central control system of the ”algorithmic” variety. This is related to the dynamic system metaphor which Borgo uses to describe free improvisation, and feels to me like a fruitful way of thinking about organizing improvisors along the continuum. Another perspective of the continuum emerges here, and Eno talks about how most music utilizes hybrid forms of these two ways of organization.

Now, in composition these days (Eno’s article is originally from 1974), ”Algorithmic composition” usually means “the process of using some formal process to make music with minimal human intervention” (Alpern, 1995), often in a random (stochastic) way where the outcome is not known beforehand. This would make it more related to a heuristic process:  Enabling a person to discover or learn something for themselves.

I still think his article illustrates the differences between two ways of organizing music in a succinct way, so I choose to use his definitions for the remainder of this section.

My own music has evolved towards having some  ”algorithmic” structure in the macro form, as I decide the temporal organization of the sections of each piece: Start, middle, end, transitions between sections. I do, however, also utilize sections of ”heuristic” composition, where conditions and some kind of musical identity is predetermined by me, but the outcome is unknown. When these different approaches occur is the result of my artistic choices, but eventually, all music is in linear: Sound sculpted in time.

There are also sections of free improvisation in my forms, where the only explicit condition which influences the music is instrumentation. There may also be a tacit influence from the preceding section and a feeling of what will work musically in the macro form and ”culture” of the ensemble. I try not to directly influence the actions of any of the improvisors in these sections, in the sense that I don’t tell the improvisors how to improvise. They will, however be influenced by their memory of what goes on before them in the macro form.

The notated music in the ensemble has gradually moved towards a less predetermined way of working during the research period. In the early days, the emphasis was on more strictly notated pieces with a clearer separate musical identity, but now the project has become more process and improvisation oriented.  I have felt that the improvisors needed to have more influence over the material, but I still wanted the benefits of predetermined composition in the large ensemble.

”Composing means building an instrument”  (Helmut Lachenmann 1986)

The thing that matters most in group music is the relationship between those taking part.

The closer the relationship, the greater the spiritual warmth it generates. And if the 

musicians manage to give wholly to each other and to the situation they’re in, then the 

sound of the music takes care of itself. Good and bad become simply a question of how 

much the musicians are giving.” 

(John Stevens 1968)

Composing the Ensemble

The first, and one of the most important compositional decisions to make while composing with free improvisers, is choosing the right musicians for the ensemble. In this context the choice of musicians is a decision beyond choosing the instrumentation for a piece, or even writing for an ensemble with a certain established approach or way of doing things.

Photo:Francesco Saggio

Instrumentation of Personalities

With free improvisers, I give away a lot of my control, and also place a great deal of trust in them. To invite someone to take part in this ensemble is an invitation to collective music-making, and to taking part in a closer relationship than that of a ”normal” ensemble playing music which is more predetermined. Mutual respect is a prerequisite, and it is also assumed to be present from the beginning, at least for me. For this reason, an important part of the process is to get to know the improvisers in advance, both personally and musically, and then attempt to make an ”instrumentation” of musical personalities. Musical and personal chemistry is very important, and essential to the success of the music. (In previous ensembles, I would take a much more dictatorial approach, having a clearer idea about how I wanted the musicians to play in advance).

The musicians have some shared aesthetics regarding form, choices of sounds, how to relate to each other and communicate musically. Some of them also have a shared history through working together in other situations. Yet: No two people are alike, and they also have different backgrounds, and varying musical interests and ideals.

Now, chemistry does not necessarily mean that everyone have to be best friends and have the same taste. Some musical friction can be healthy, and I think it can yield some interesting results as long as there is a mutual respect. The most original music in the ensemble, seems to me to emerge when the different aesthetics of the musicians clash and react to each other, using my predetermined music as catalyst. There needs to be enough compatibility to enable us to work together, yet enough friction to let the musical reactions occur.

Unfortunately, It’s hard to really know how everyone will function in the large ensemble, both musically and personally, before we try it. Some may present a different image of themselves before we start working, and not function in the way I had hoped.

Choosing Musicians

I started the process of choosing musicians by listening to recordings of them playing solo or together with others, trying to get an image of their aesthetics and attitudes,  and imagining how they would play in the ensemble. Listening to the recordings gave me an idea of what they are capable of, and also an idea of what kind of aesthetic area they might like to move in. If this coincided with my interests, I would use this in my planning of the compositions.

Another approach was to do a number of improvised sessions with them, both concerts and rehearsals. Some of them I already had relationships with over many years, but we hadn’t necessarily played free improvisation together regularly. Some were new acquaintances to me.

I wanted to feel how I worked with them musically, what their voice and skills were like, and how we interacted socially. Are they good readers? Good improvisers? Big egos?  It was important to me that the musicians had strong expressions, and could contribute to the music, as well as the ability to reflect on and discuss the project in the rehearsal situation.

Sonic History

To give a concrete example: The saxophone player Martin Küchen and I have worked together for 12 years with the Trespass Trio with Raymond Strid. We have good discussions about the music and are usually in agreement aesthetically. In the Trespass Trio we play mostly free jazz, with simple, open melodic tunes, but I knew that Martin also

Photo:Julia Spicina

has a more sound based, low dynamic expression, closer to European free improvisation and contemporary music. This is the case with myself and Raymond Strid as well, and, although the borders are not absolute, we seem to avoid typical material from the free jazz idiom when we play free improvisation: Rarely pizzicato bass and ride cymbal together, less ”jazz” influenced saxophone, metric rhythms etc. We all seem to compartmentalize a little when it comes to free jazz vs free improvisation (another example of how ”free improvisation” can be an idiomatic term?).

Since I knew I wanted to have both Martin and Raymond in the ensemble, I saw that we needed to use more of our ”improv” language, and less of our free jazz ways of playing together. Especially when we play as a trio, or as quartet with other instruments in the ensemble. The trio of saxophone, bass and drums has a sound which makes it seem more idiomatic, giving associations to jazz, and making it more challenging to make ”non-idiomatic” music.

Trespass Trio: ”Centers”

John Butcher says that certain instrument combinations have a much stronger ”sonic history” than others. (Butcher 2016) This term refers to what associations a certain instrumentation gives the players and listeners: In a ”jazz” group like the Trespass Trio it’s possible, by for example playing arco or preparations on the bass, prepared saxophone and untraditional percussion, to alleviate some of these associations. However, if the goal is to make ”non-idiomatic” music as a trio, it will give a lot more resistance in the process than for example a trio of electronics, sax and percussion. An electronics player and a sax player will probably approach improvised music from different starting points, with different improvisation traditions in the back of their minds. (Not to say that they can not learn from each other).

I do use this idea consciously for musical purposes in my music sometimes. In Li 2 (Kongsberg), there is for example an almost free jazz-like improvisation with two saxes, pizzicato bass and two drum kits playing in a very energetic way. I wanted this to contrast the more controlled, contemporary music-like, low dynamic of the rest of the piece.

Kongsberg 2016 excerpt (Quintet)

The same goes for more ”classical” instrumentations, like flute, strings and clarinet, for example, which can sometimes give associations to ”contemporary” music. The musicians will tend to move into this landscape when the instrumentation suggests it.

Here is an example of a recording of myself and Ole Henrik Moe from a duo rehearsal, in one of the first cycles of the project.

Acoustic Instruments

I early on made a choice to only include acoustic instruments, or acoustic instruments that were amplified only slightly, in the ensemble. One reason for this was practical: I wanted to be able to play concerts and conduct rehearsals with a minimum of technical requirements.

I also wanted the sound to be blendable dynamically, following John Stevens´  well known idea in the Spontaneous Music Ensemble, that all the musicians should adjust their playing volume to be able to hear everyone else at all times. With electronic instruments, this principle can sometimes be difficult to follow, because of amplification which makes dynamics harder to control, the difference in acoustic vibrations, and physical gestures which can be harder to interpret.

I will not tell musicians to change their instrumentation (for example play electric guitar in stead of acoustic guitar), but if I notice an interest in using (or not using) a certain different instrument, I can encourage it. This will also motivate the improviser, in their own process of discovery.

Like I said, composing a group of free improvisers, the instrumentation does not only mean a blending of instrumental colours. It also entails a blending of the musical personalities behind the instruments. The ensemble I‘ve put together was intended to be a mix of different ”voices”, with somewhat different musical backgrounds and aesthetic ideals. Yet not so different that it would be impossible to reconcile their differences when working together.

”..the interest for me in improvisation is making that kind of music which you couldn’t really imagine before you find yourself in the middle of it.” (John Butcher 2016)

Benefits of Composition and Improvisation

Throughout this project I have asked myself which musical ideas can work in both composition and improvisation. Or rather: What ideas, thought out in advance, can be of use to my ensemble of improvisers? I have, in a way, tried to compose some of that which I feel is difficult to improvise in a large ensemble:

Clarity of form, structure and temporal organization, identity of materials (that can be reproduced), instrumentations, large scale behaviours (and listening) and achieving equilibrium and coordination quickly.

I have also asked myself what the improvisers’ spontaneous musical ideas can provide for the music that predetermined composition can not:

The improvisers can interpret the predetermined ideas or material, making them sound better and more alive and subtle.

They can help me compose in a quicker way, where I do not need to re-notate everything to try out compositional ideas.

They also inspire my compositions through a constant feedback loop:

I might, for example, bring an idea to a rehearsal, which is not yet fully formed.

In an ensemble of 11 improvisers, there is an almost limitless creative potential, and I will always get new, and better, ideas after rehearsing with them. If I bring back one of these ideas in a more worked out form, it will probably be influenced by the musicians a second time, and possibly morph into something new. Their way of making my material their own becomes an important part of the feedback loop.

The musicians are also composing on their own, ”making that kind of music which you couldn’t really imagine before you find yourself in the middle of it”, improvising within their sound worlds, bringing their own materials to the music.

By ”material” I mean that which the music is built from: Musical ideas, sounds, pitches, rhythms, gestures etc.

The Voice

John Butcher, in his article ”Freedom and Sound – This time it’s personal”, gives the simple definition of the expression ”voice” as ”a useful shorthand term for an individual’s sound and ways of playing” (Butcher 2011).

The voice is an expression of the musician’s previous musical and personal experiences and taste, of her listening and expression of what she hears, of her ways of reacting to the overall sound, of her web of artistic practice.

I chose the word ”with”, as in ”music making with an ensemble of improvisers” as opposed to ”for improvisers”, for a reason:  I have begun to think about the voice of the improvisers, including my own, as the main materials to compose with.

Other Materials

The other materials, those which I use to compose predetermined material, have gone through some changes during the project, from an initial inspiration from Korean Gugak music, including a detour into Spectral music, and ending in a toolbox of approaches inspired from patterns found in Nature. In the following chapters, I will discuss this process, starting with my relationship to Korean traditional music, Gugak.

4. Gugak