Alan Edward Williams

Composer/ cyfansoddwr

Collaborative composition with Paynter: a digital tablet-based platform

1 Comment

In 1990, when I was an undergraduate in Edinburgh, we had a visit from the Polish composer Witold Lutoslawski. Some of the students had been working with schoolchildren on a community music project with players from the Scottish Chamber Orchestra, creating music collaboratively based on one of his pieces, under the inspirational leadership of Nigel Osborne. “Very interesting”, said Lutoslawski, “but of course, this is improvisation, not composition”.


Lutoslawski was well known for techniques (described as “aleatoric” after the Latin for dice) which allowed musicians a certain degree of freedom within fixed parameters to play material at a pitch or time of their choosing, so it’s interesting he drew this clear line between composition and improvisation (or “not composition”). What he meant, I think, was that even where some freedom within given parameters is allowed, the composer should have sole authority over the structure of the music, and over the parameters set.


While Nigel Osborne was – and is – driven by a deep belief in the restorative power of music to inspire communities and to heal division, as he would go on to show in Bosnia and elsewhere, more generally the imperative to involve communities and schools in community arts projects related to the professional activity of an orchestra springs partly from a democratic principle – driven by public funders– and partly from an urgent need to develop new audiences for classical music, and probably the belief that the two things are one and the same.


Until relatively recently, the structure of community music projects tended to be hierarchical, with the composer at the top, whose work is performed by the professional ensemble; some players from the ensemble go into a community setting, usually led by a professional community musician or composer-animateur, to collaboratively develop a piece or an artwork based on the professional piece, and that community piece or artwork is then shown or played in an event which is close to, but not identical with, the main ‘professional’ event. The concert audience may form part of the community group; and they may  see the work produced  by the community group – but this experience is peripheral to the professional performance event

comm arts trad

In 2009, when my piece “Wonder: A Scientific Oratorio” was premiered, then BBC Philharmonic projects director Martin Maris led an amazing project with multiple animateurs to create artwork and soundscapes with schools from around Salford based on the ideas explored in the oratorio. The original plan was that music created by the community groups would feature in the same concert, but in the end, it was shown in a nearby space which audiences to the main event could go to if they wanted.


But what if you could reverse that hierarchy, and make a project in which the final performed work came as a result of the community arts activity, rather than a spin-off? Could we create a structure and practice in which ideas were sourced from the community group and ended up in a professional performance, rather than the other way round?


Paynter, developed by Dr Adam Hart as part of his PhD research, enabled that to happen.  Paynter is a tablet-based app, which allows the user to draw lines, blobs and position icons generating sounds in a drawing-type background, and for sounds generated to be played back. It is designed so the user can use sounds and notes to tell a story. Its design allows certain compositional procedures, such as motivic repetition, transposition and imitation to be used; it also allows sounds recorded from the environment, or from an ensemble to be manipulated in certain ways.

Paynter on tablet

Adam and I went into St Peter and St John Primary School in Salford on five visits in March-May 2019. On our first visit we were accompanied by four professional musicians  provided by the BBC Philharmonic orchestra: Gary Farr (trumpet), Gemma Bass (violin), Kathryn Williams (flute) and Elinor Gow (cello). All the musicians were accomplished improvisers and animateurs in their own right, and we were conscious that the tablet-based app that Adam had developed might prove to be surplus to requirements. But we persisted, because we wanted to see if the app could be used to encourage the participants to think in specifically compositional ways.


In the first session, the musicians introduced their instruments, and developed an improvised piece to 2 stories that the school had already been working on – one about the Titanic, and one about a cat and a bird. The whole session was recorded, and the sounds generated from this improvisation, included lots of extended techniques were assigned to icons in the Paynter app.


In three further visits, the composer-technologists (Adam and I) worked on the narratives in two classes, one year 3 and one year 5. We got the kids to divide up the story into scenes which could be relayed using sounds and musical ideas using the Paynter app on the tablets. The kids created and finalised graphic scores on the app, and I then transcribed these into traditional notation.

Cat in glue

“The cat gets stuck in the glue”

cat in glue score



I made great efforts to stick as closely to the graphic scores as I could: I was aware that all of the musicians could probably have played the pieces straight off the graphic scores. But Paynter’s design created a fixed relationship between the sounds created in the project and the graphic score – it would always play the score the same way. This clearly is the opposite to the standard use of graphic scores with human musicians, where they are used to initiate a more flexible relationship between the “work” and the sounding outcome.


But that fixed relationship meant that I could transcribe with some confidence onto the ensemble of four musicians. Some challenges remained. Although much of the material created in the group was either pitched notes (like MIDI) or recordings triggered by icons of music created in the first session by the instrumentalists themselves, much was also sound effects (how do you make the sound of leaves rustling on a trumpet?). Most problematically, sounds created at pitch on the original ensemble of trumpet, violin, flute and cello, when shifted to the bottom of the canvas, were transformed. So a simple trumpet note became a ship’s horn, for example.


When we played the pieces back with the ensemble, the school participants also created mouth and percussion noises alongside the ensemble, on cue from the composer-animateur (me). One noise in particular was popular in the Titanic scene – one student was able to make very effective dolphin chattering noises with his own voice. Another member of the ensemble, BBC Philharmonic cellist Eli Gow, used a harmonic glissando to make a seagull sound, which most groups used. Because the sources for musical ideas were varied, sometimes used in mediated form, a flowchart diagram for ideas is much more complex, and less hierarchically organised than the conventional model for community arts projects. It might look something like this:

coll comp structure


The aim here is to show the multiple relationships occurring in the project, in which no one person has overall authority over the final outcome.


The final phase of the project was the use of some of the ideas of the collaboratively composed score in a longer piece by me, The Rivet’s Tale. This performance had already been agreed, but the piece had to be written after the school project phase had finished, leaving me with little over three weeks to write a piece of approximately 9 minutes for 13 musicians. While the collaboratively written pieces – lasting 2 minutes each, and for only 4 players were relatively quick to transcribe (I did both of these transcriptions in one long composing day), it proved difficult to simply expand these short pieces into a longer piece. The material created for the shorter pieces just wouldn’t work like that. I eventually abandoned the thought of “simply” scaling up the collaboratively composed pieces, and chose only the Titanic story, weaving a more complex texture out of carefully chosen themes that would work in the new context.


As a project, we showed that the Paynter app could encourage children of Key Stage 1 and 2 ages to think compositionally. The use of graphic scores generated by the kids themselves which are then transcribed as carefully as possible meant that the process of collaborative composition is much more transparent. We didn’t test the proposition that by composing a piece collaboratively “bottom up”, the participants would have more of a sense of ownership over the collaboratively composed piece, or gain more understanding of the “professionally” composed piece, and future projects need to build that in.


I am left feeling unsure about Lutoslawski’s distinction between composition and improvisation. I use improvisation a lot to begin a piece, developing material by playing a lot of the time. When I am improvising in a jazz context, I’m often thinking and planning in a motivic way. By anchoring much of the activity on the app as “play”  – which is by its nature collaborative – the process could be viewed as a kind of improvisation. But it’s also very much like my process of composition. Perhaps this simply points to technology having a significant effect on the nature of the creative process, and that this has altered significantly since 1990 when Lutoslawski made his remark.

One thought on “Collaborative composition with Paynter: a digital tablet-based platform

  1. Pingback: Collaborative Composition with Paynter | Adam Matthew Hart

Leave a Reply

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

You are commenting using your 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