The Birdfansyer's Delight or Choice — Idiomatic Quantisation — CAC-project by Dagfinn Koch

Context

In the pursuit of new ways of expression, composers and instrument makers have, over the centuries, explored untraditional playing techniques to produce original sound combinations. The symbolic notation has been central to the development of Western art music. It provides the possibility to record existing musical sequences, construct musical material in contact with the paper and, in particular, serve as instructions for performers to reproduce music. The reproduction is physical, and in the pursuit of “nie erhörte Klänge” (Schönberg), the idea the composer has had has been difficult to produce. Composers have blamed unwilling musicians, while musicians have blamed incompetent composers. I am a viola player myself and have a long experience as a chamber and orchestral musician, primarily at a semi-professional level. I do not know of many composers who are competent instrumentalists.

I would like to map traditional and less traditional techniques to test a given note material against playability using artificial intelligence. Elvio Cipollone has touched upon something somewhat similar in his work with OM-Virtuoso, as described in an article in The OM Composer’s Book 2. But he hasn’t further developed the idea, and it was limited to one work and one instrument, the clarinet. Apart from this, as far as I know, there hasn’t been much research done in this area. The research group at IRCAM Centre Pompidou is interested in the idea, which I think confirms the lack of research. Symbolic notation and quantisation are key areas of their research in musical representation.

Important detour

On the Baroque transverse flute with one key, one can achieve a consistent tone across the entire register only in the keys of G major and D major. I recorded Professor Hans Olav Gorset playing all 41 pitches without adjustments. All tones are recorded for both classification and mapping in a software sampler.

I’ve mapped the instruments' tonal possibilities and fingerings to use machine learning to create datasets for algorithms realised from scratch in Open Music, producing “scales” that result in a consistent tone, the opposite, or transformations between these. As it turned out, one can understand the necessity that Theobald Böhm (1794-1881) developed a system of keywork and a fingering system for the flute.  

Software: Open Music (IRCAM), Kontakt (Native Instruments) and Visual Studio Code (AI code editor). 

Artistic method/process

My project, in the long run, is about mapping the span from traditional to non-traditional (extended techniques) playing techniques for clarinet, guitar, glass harmonica and viola. I wish to develop a library (a plugin) for the computer-assisted composition software Open Music from IRCAM. Using the library, one can assess the playability of musical material. The library can also suggest idiomatic (quantisation) solutions and solutions that challenge given conventions. The time dimension is very important. How long does it take to move from one tone to the next, from one chord to the next, or from one technique to another must be mapped. One can imagine this expressed as percentages. For example, the library may be asked to produce results in which 80 per cent of the musical material falls within traditional idiomatic practice. In addition, the library can serve as a model for creating electroacoustic music by providing a virtual instrument based on the characteristics of an acoustic instrument. (The Diophone Studio software from IRCAM can do this at a micro level.)

The development of the directory (library) will be one of the three pillars in my work. (The other two are a historical-philosophical reflection around the body, machine, and instrument and the third of the mentioned seven short compositions.) I have been using computer-assisted compositional software for 28 years.

Results 

The project will lead to the development of a library for the Open Music software, to be distributed and maintained by IRCAM. I imagine a concert at the end of the program period, in both Oslo and Paris. As mentioned, the written reflection is thought to be a historical-philosophical reflection around the body, machine, and instrument. The project is open to articles and presentations under the patronage of the Norwegian Music School and IRCAM. What I may arrive at could also serve as a starting point for discussions about the use of the material, in particular, and the idea of form, in general, in composition.

I want to show how important it is for composers to learn how instruments work. Yes, how important it is to be a competent practitioner yourself in order to become a good composer. My impression is that younger composers at least have a lesser understanding of this. They are used to having a computer as a user interface and don’t get the real physical experience an instrument provides.

On the other hand, I see that it may be possible to excite the younger generation about acoustic instruments through their interest in computers, even though it is too late to become capable instrumentalists. But one can contribute to a change of attitude, as computers are a perfect tool for working with formalised music. Hopefully, I will be able to enthuse those who are sceptical about computer-assisted composition by highlighting its benefits. A computer does not exclude the human aspect as long as one is willing to understand that an instrumentalist isn’t primarily interested in the work, but rather in what the physical instruction notation can provide in expression possibilities. Virtuosity doesn’t have to mean mastering something “unplayable”. It may also mean the pleasure of playing something that lies well for the instrument (including extended techniques), which in turn contributes “to the instrument revealing its spirit.” (Bis das Instrument seinen Geist offenbart. Klaus K. Hübler)

Translated from Norwegian by Malin Kjelsrud and Dagfinn Koch