In the session with L.M. and S.Ö., the immediate impression for a viewer could be that of a complete swapping of the agent's respective roles: Who is the composer and who is the interpreter in the first video clips? S.Ö. is writing music while L.M. is passively listening. L.M. suggests the addition of a fermata in the variation that S.Ö. has just notated. Our claim is that, even though we approach a situation in which the relative positions happen to be at their respective extremes, what we see is still within the boundaries of artistic practice both for composers and performers. The observed interplay is an example of how the roles of composer and performer in themselves overlap, and can even seemingly be interchanged in this way.
In essence this is a matter of ontology: On the one hand, what makes up the musical work, and on the other hand, what does performance interpretation amount to? Just as was suggested in Section 1.1 S.Ö.'s actions in the video clips involves a strong element of construction. What performers do is making versions of works, and these versions are in a sense the performer's co-creation of the composer's piece.3 But what then defines the composer's work independent of its performances? Following the line of thought of Stephen Davies, it is the work-identifying instructions that delineates the work [Davies, 2001]. For a musical work these instructions are usually regarded to be some of the fundamental aspects of the notation. However in Section 3.1 we observed how the work assumed a number of different representations. In that case, though the notation will be fixed and may eventually be said to constitute the `work-identifying instructions' we will argue that the work does exist even prior to its notation since, on the part of the composer, it makes up the reference against which the negotiations are held. Furthermore in the case of a piece for instrument and electronics, much of the identity of the work is also specified in the computer programming and in the electronic sounds. This is important to bear in mind while studying the session with L.M. and S.Ö.. The point of departure is a melody that L.M. has derived from a previous tape composition. The musical material that evolved from L.M.'s transcription appears in a context where real-time processing and pre-prepared tape material contributes strongly to the identity of the music.
If we accept the idea that L.M. and S.Ö. are acting as composer and performer respectively and consider how their actions can be divided between the poeitic and esthesic fields, also taking into account the discussion in Section 3.1, we may draw some important conclusions from the empiric studies:
The flexibility that we can observe in the interaction between the two agents in the video clip is remarkable. Complete misunderstandings and miscommunication does not halt the process nor does it appear to lead to false conclusions; it is only at a close examination of the flow of events in the video that we can observe the misinterpretations. In the end some of the misunderstandings, such as the idea of adding several fermatas (line 59 in the graph, see figure 8) worked their way into the final version of the score.
The quote from Molino in Section 2.1 can now be read in the light of the performed analysis. Although the misunderstandings can be regarded as 'noise' when analyzed from the point of view of information theory, in the collaboration between L.M. and S.Ö. it rather seems to be an integral part of the artistic process. It shows how the classical notion of the 'creative misunderstanding' really can play an important role in artistic work.
The way the computer part in ``Viken'' is set up, S.Ö. has a pedal that controls the synchronization between himself and the computer. This method of resolving that particular issue in mixed media music is not uncommon. It relates to the notion of synchronization as purely a technical issue; a unidirectional stream of communication. Though the occasional pressing of a pedal does not resolve the critical issue of rhythmic alignment and musical timing on the micro level, it does keep the musics of the two parties aligned in the larger structural meter. That is; when it comes to synchronization of preprepared elctro-acoustic material with acoustic instruments what is achieved is a series of meeting points and more seldom an integrated flow of events. Now, in the context of the composer/performer interaction (see the analysis in Figure 8) there is a striking lack of synchronicity between the different actions. There is an evident and independent flow of the initiative, of the constructive and of the interpretative input between the two agents.
Would it be possible to use the knowledge gained from the analysis of the video in the design of an interactive interface for a mixed media piece to be performed live? Before drawing any conclusions it must be stressed that the session with L.M. and S.Ö. is obviously not performed under the same conditions as are required for a performance of a piece of mixed media music. When it comes to real time electronic processing and synthesis the processes quite naturally translate themselves into the language of esthesic and poietic. In general - and somewhat simplified - we can assert that processing of acoustic sound input is an interpretative action and the generation of new sonic material is a constructive process belonging in the poietic domain. The actual program, and the code and the run-time instructions that constitutes it, can be analyzed on the neutral level. Finally, as has already been asserted, the program or computer part may affirm important aspects of the work-identifying instructions that a graphic representation of the computer part may not harbour.