ArtDoc - an Experimental Archive and a Tool for Artistic Research

Henrik Frisk, Docent and Associate Professor

What is it to document a musical work?

  • Record it?
  • Describe it?
  • Represent it?
  • Given the method:
    • what is lost?
    • what is gained?
  • A documentation is a transformation.

Methods for documentation

  • An archived score is relatively easy to represent accurately but is a poor representation of the actual music.
  • A recording of a performance is an accurate representation of the sound but a poor representation of the material performance.
  • How may a sustainable archive for artistic research and artistic practice be structured such that it avoids the risk that the archiving force hides important aspects of the data?

In artistic research the material basis of the artistic practice may be both data and result

  • Unsatisfying in this context: both aspects are difficult to represent.
  • It is necessary to critically examine the relations between artistic practice in music and its possible representations in various forms for archives

Work models

A work centered view

  • The work is the result and the result is the work. In music this can be the sounding trace, or sometimes just the score, or the media.
  • In this case a recording may be a decent documentation.

A process oriented view

  • The putting together of the material is an integral part of the identity of the work.
  • A recording of the result misses out on some of the important aspects.

Open works

  • The open work as a concept is part of an aesthetic movement since the 60s (See Umberto Eco, The Open Work)
  • The more radical version is the work-in-movement:

    ``It invites us to identify inside the category of `open' works a further, more restricted classification of works which can be defined as `works in movement', because they characteristically consist of unplanned or physically incomplete structural units.'' (Eco, 1968)

The problem

Repetition Repeats all Other Repetitions


Commisioned by and developed in collaboration with Swedish guitarist Stefan Östersjö.

Repetition Repeats all Other Repetitions

  • Has seen at least three distinct versions
  • Has been performed > 15 times in America, Europe and Asia
  • Has appeared on two different CDs
  • Yet, it has not found its definite form

Symphonie Diagonale

The work's identity?

  • no such thing as an ultimate performance, or rendering of the work
  • the work is a realizable potential
  • each performance makes possible a new work and this development is the true nature of the work's identity

Its documentation?

  • How can this incremental process be documented to allow the work-in-movement to continue to move and not just repeat itself?
  • Is an archive that archives in order to allow for change possible or even desirable?
  • Usually, the point of the archive is to preserve.

Documentation classes


The Archive - The connection class

<Class xmlns:xsi="">
    <description>A child class to Document describing 
    any kind of connection between two nodes within the system. 
    The id of the connected classes are the references
    <documentation>Documentation for this class</documentation>
        <connection-from type="Ref" 
                         desc="Connection from" 
                         edit="1" required="1"/>
        <connection-to type="Ref" 
                       desc="Connection to" 
                       edit="1" required="1"/>
        <connection-type type="select" 
                         desc="Type of connection" 
                         edit="1" required="1"/>



Documenting works composed for, or adapted to, the Klangkupolen at KMH in a sustainable manner:

A scratch on the surface

  • The question of how to best document the intense creative forces that digital technology allows for is without doubt one of the important ones in the upcoming decades.
  • The question of the archiving force of the archive needs to be taken into account.
  • The need to archive can be personal, institutional, political, economical, etc. These different needs are often not in sync.

Future work

  • Make a stable implementation
  • Test it in a number of different configurations and for different purposes
  • Much more…

Thank you!


Henrik Frisk