next up previous contents index
Next: Discussion Up: Negotiating the Musical Work. Previous: Method   Contents   Index


Empirical study

Harp piece


The work discussed here is a work in progress. It was commissioned by the Mexican harpist Mercedes Gomez in 2004 for harp and computer. The melodic material used for the piece consists of a small six note fragment that is transposed and repeated creating a tone row of potentially infinite length. Due to the cromaticism of the series the material isn't very idiomatic for the harp. It was chosen because of its inherent structure and its relation to the original vision of the music. But also because of a genuine interest for the process of negotiation between the fundamental building blocks of the composition and the idiomatics of the instrument for which the composition is intended. The process of adapting this specific material for the harp will inevitably involve constructive interpretation, performed by the composer on his own musical idea. In the case of the bars discussed below, much of these negotiations were discussions between Frisk and the performer. It is important to bear in mind that to some extent the notion of what is ``idiomatic'' and ``playable'' is relative to the technical approach and instrumental modes of expression of a certain performer. Naturally, many of the solutions and suggestions below relate very specifically to this collaboration and cannot be generalized as such. A certain amount of dialogue between idiomatics and musical ideas is likely to occur in any kind of compositional work but the conditions under which this work in progress has evolved makes the internal as well as the external negotiations stand out.


The following is a discussion on four notations of the same two bars of music. The first (Figure 1) is the original idea transcribed as closely as is meaningful into the atomized rhythmic structure of western notation. The basic musical idea at this spot is to have the same variation of the tone row in three individual parts, separated by octaves, each one following its own unique and precisely notated rallentando.
Figure 1: First transcription of the idea into notation.
Image harpVersion1

In the second example (Figure 2) we find the first attempt at transcribing the musical idea for the harp. Pedal changes are not notated. At the indicated tempo, these two bars are still virtually unplayable on the harp. The F flat to F natural pedal change at the end of the first bar is a technical problem as is the G flat to G natural on the second eighth note of the second bar. After working on this passage with the harpist a version in the lines of Figure 3 was suggested.

Figure 2: First transcription for the harp.
Image harpVersion2

The third example (Figure 3) is rhythmically less complex. With a few written indications the effect of the slowing down of the music could be approximated. The pedal changes are resolved by means of pedal glisses. However the independent parts and their individual rallentandi cannot be traced in the image that this notation produces, which also means that a neutral analysis of this version will reveal little of the original intentions.

Figure 3: A transcription rhythmically less complex.
Image harpVersion3

In the final example some of the rhythms have been simplified by use of grace notes and, as in the previous example, some of the pedal changes have been changed into pedal glissandos. This contributes to making the idiomatics of the instrument a part of the counterpoint and a balance, acceptable by both the composer and the performer between ``authenticity'' (to the original idea) and the playability of the excerpt, has been reached.

Figure 4: A final version. More idiomatic than the example in Figure 1, but closer to the original idea than Figure 3
Image harpVersion4


Considering the esthesic processes in the evolving notation of this passage in the light of the discussion in Section 1.1, it might be worthwhile to consider the significance of Frisk's conceptual vision of the work and how the negotiations between this musical matter and the idiomatic constraints of the harp lead up to a version of the work in musical notation. And further, how the presence of the performer in this discussion provides new impulses for the piece, specifically its notation and rhythmical articulation. The original vision is the trace that constitutes the source for the interpretative, i.e. esthesic, actions leading to the different notations. Though the notation is altered significantly through the four variations the core of the original musical vision remains the same throughout the different varitations; i.e. on the poietic level the negotiations did not alter the music. However, in a neutral analysis of these four excerpts we might be tempted to suggest that the process lead to four different kind of musics. In any event, we find that the study shows the recursive nature of the interplay between constructive (poietic) and interpretative (esthesic) processes both in the communication between the composer and the performer, but also as an internal process in which the composer himself is negotiating between the original vision and its representation in musical notation.



Love Mangs' (L.M.) ``Viken'' for guitar, banjo, e-bow and electronics (2004-05) was commissioned for Stefan Östersjö (S.Ö.) by the Swedish Arts Grants Committee. The project had several explicit intentions, apart from the mere production of a work for guitar and electronics. One was to use real time processing as the main source of electronic sound, the other was to explore the boundaries between composing and performing; between the performance interpretation of a work and how different kinds of fixity can be established in a work. This should be taken into account when studying the video documentation of this process. It would be a mistake to regard the material as documentation of a typical collaboration between a composer and a performer, as both L.M. and S.Ö. were well aware of the underlying intention to explore the possibilities for improvisation and other interactive ways for composer and performer to approach the process of creating a score-based work with electronics. While studying the video it is also of importance to remember that both parties involved are aware of their process being documented. However, the session is taking place less than two months prior to the scheduled premiere which implies that both parties are strongly focused on the task of getting the piece together.

The transcription has played an important methodological role in our analysis and can be found at
and all references in this text to the video refers to sections of the transcription. The video clip is edited; sections with little or no action are simply removed but the order of events are not altered. A compressed version (QuickTime movie) of the edited video can be found at For reference, an unedited version of the same passage can be found at The video was recorded during a session in the composer's studio on September 17, 2005.

Material worked out prior to the documented session.

It is of importance to the analysis of the communication processes in the video to understand the material that L.M. and S.Ö. had at the outset of the session. L.M. had derived a melody from a filtered electronic sound clip, which originally wasn't intended for ``Viken''. As the process of composing ``Viken'' evolved he wanted to include the sound file as well as the melodic material derived from it in the work. Almost any kind of notation will inevitably be a reduction of the material that is the object for notation. Already when L.M. decided to make a transcription of the sound clip he subjectively chose elements to emphasize and elements to exclude; thus making an interpretation of his own material. He is working in the esthesic domain on the trace left by a work performed in the poietic domain.

Figure 5: Love Mangs first notation of the melody derived from the sound file.
Image viken

What is interesting with the way L.M. has carried out the transcription is that he doesn't even try to establish a connection between the sound clip and its expressive qualities in the notation. Instead he has extracted an ordered set of discrete pitches that establishes a clear tonality (see figure 5). We can say that he re-constructed a musical motif independent from its source. In the context of his working on ``Viken'', what he heard in the sound clip was the melody. An action performed in the poietic domain as a result of working with the material in the esthesic domain but with 'knowledge of the poietics of the work' as Nattiez would put it, the work in this case, not being the context of the sound clip but the poietics of ``Viken''.

Analysis of the video.

Figure 6: Material 8B from the final score of ``Viken''.
Image viken8b

The agreed purpose of the session documented in the video, was to work out variations on the melody transcribed by L.M.. His intention was to use this melodic material in the piece.

In the first scene S.Ö. has just played an improvisation on the melody and on L.M.'s suggestion he is notating the new variation (see figure 6). S.Ö. is active in the poietic domain, constructing new material for the piece. He turns to L.M. for feedback, but at this point L.M. appears remarkably indifferent. This is illustrated by the arrow going from the new variation box in the poietic field on S.Ö.'s side of the graph pointing down towards L.M.'s side in figure 8. There is a lack of communication between L.M. and S.Ö. (illustrated by the dotted arrow going upwards from the restless, passive box) as L.M. does not respond to S.Ö.'s invitation to discuss the new variation. L.M. seems to have accepted the new material as it was played initially and instead takes the initiative (illustrated by the initiative axis going from S.Ö.'s side to L.M.'s), adopting an interpretative approach on S.Ö.'s variation. L.M. is now active in the esthesic field, suggesting to S.Ö. the addition of a fermata in the variation (line 24) represented by the fermata box in the graph. Now there is apparent noise in the communication (represented by the dotted arrow going from the fermata box to the new variation box): L.M. appears to be unclear of where in the notation the fermata should be. This in turn leads to a misunderstanding by S.Ö. (line 59), taking L.M.'s suggestion to mean several fermatas (dotted arrow from the several box to the fermata box). Our interpretation of the dialog and the interaction here is that it takes L.M. a while to find the right spot in the notation (by S.Ö.). He seems to point at different spots in the score but in fact he is seeking for the end of the phrase which is where he meant for the fermata to be. Eventually L.M. points it out and for the first time a clear communication takes place, illustrated by the two arrows in the graph going in both directions (line 68 in the transcription). In this segment both L.M. and S.Ö. are acting in the esthesic domain, L.M. in his interpretation of S.Ö.'s notation and S.Ö. attempting to try out L.M.'s suggestion. S.Ö.'s initial misunderstanding of the fermatas seems to lead to the next initiative taken by S.Ö. and is illustrated by the initiative axis going from L.M. to S.Ö. at 75. Again in the esthesic field, S.Ö. suggests that long fermatas could be added to the last notes of the phrase. At first L.M. doesn't get the idea at all (line 78, dotted arrow going from the long fermatas box to the what? box) but eventually approves of the suggestion (line 85, solid arrow going in the same direction).2

This is followed by what seems to be an attempt on L.M.'s part to enter the creative discussion or to reclaim the artistic initiative. The response from S.Ö. is not related to what L.M. says (the dotted arrow going from the 4th string box over to S.Ö.'s side in the graph at line 94). The passage ends with S.Ö. playing the whole phrase again giving L.M. a look at the end (glance box at line 111) without getting a noticeable response (dashed left bound arrow). It is obvious that the communication in both directions is very noisy - this passage is filled with unanswered questions and misunderstandings.

In the next clip S.Ö. is writing down the variation in more detail, inserting L.M.'s idea of the fermata as a normative inscription (line 119). In that sense the initiative is on L.M.'s side, in spite of the fact that S.Ö. is the physically active part with the writing. At this point S.Ö. is not artistically involved, basically just making a note of L.M.'s interpretative idea. L.M. then develops his idea of the fermatas and their significance in this passage, still active in the esthesic domain. The way we analyze this follows a model in which the difference between creative actions in the esthesic and poietic domains is a difference in what class of material the creative act refers to. Nattiez defines these as the psycho-sociology of creation and psycho-sociology of perception respectively [Nattiez, 1990]. L.M.'s discussion of the fermata emanates from his perception of S.Ö.'s improvised new variation at the very beginning of the video clip and is therefore to be regarded as an esthesic process.

Figure 7: Material 15 from the final score for ``Viken''.
Image viken15

The idea of inserting several fermatas, which in the beginning was a misunderstanding on S.Ö.'s side, is now completely accepted and incorporated in the music as it is envisioned by L.M.. However, just as in the previous passage, S.Ö. doesn't respond to L.M.'s remarks. Instead he starts playing the phrase from the beginning (line 139) and, at the time he reaches the end of the phrase, introduces new material in the form of an extended arpeggio (line 140). S.Ö. regains the initiative and moves into the poietic domain. The communication at the moment when the new material is discovered is immediate and distinct; S.Ö. gives L.M. a glance and L.M.'s humming reply is evidently positive (at line 145). At line 150 S.Ö. takes an interpretative approach, commenting on the sound of the new arpeggio. The clear communication at this spot is underlined by the fact, that for the first time in the video clip, L.M.'s attention turns to S.Ö. and the instrument and away from the music stand.

Figure 8: A graph of the interaction between Stefan Östersjö and Love Mangs in the session analyzed in Section 3.2. The scale in the center axis refers to line numbers in the transcription of the video.
Image timeline_horiz

At this moment S.Ö. starts trying out a new context for the arpeggio which evolved from the previous variation but is of a different character. He plays with the minor seconds that since the introduction of the idea at line 140 have been leading up to the arpeggio and attempts to merge the new arpeggio with a series of chords from L.M.'s material notated prior to the session. At this moment S.Ö. starts to summarize the achievements so far during the day. He starts playing the version of the melody with harmonics. L.M. interrupts him by asking him to "notate the last thing you did!". The remark indicates that L.M. has decided to include the new chord progression in his conception of the piece (see figure 7) and thus his actions move into the poietic field. This leads to a discussion on how to define the passage in terms of musical notation. L.M. suggests that it doesn't have to be all that defined ("Just notate it as a draft", line 201). S.Ö. suggests a strategy for the notation of the phrase which L.M. finds satisfactory. In this last sequence L.M. is organizing the material and performs a typical 'compositional' action still in the poietic domain.

next up previous contents index
Next: Discussion Up: Negotiating the Musical Work. Previous: Method   Contents   Index
Henrik Frisk, Stefan Ostersjo