Opus 17a + EVOL

I recently had the honour of working on something for EVOL, the computer music project by Roc Jiménez de Cisneros and Stephen Sharp. This particular project involved a re-creation of Opus 17a by Hanne Darboven, and my task was to analyse the music from an MP3 and identify the notes so that EVOL could play it.

Hanne Darboven was a conceptual artist who worked with processes that could be described as generative because they were based on rules for manipulating numbers and making patterns. Darboven often used calendar dates re-arranged according to rules, displayed visually in grids drawn on paper that were arranged in larger grids. Opus 17a is derived from one such calendar-based artwork (Wunschkonzert, 1984), in which the numbers are transcribed into notes. The result is an hour-long piece for solo double bass.

At first, Opus 17a might sound ‘random’, but there are different patterns at different scales that become identifiable with repeated listening. My approach to re-constructing this piece from the recording ended up using these perceivable patterns. Initially, I’d hoped to be able to identify the rules that Darboven used and to re-create the piece by coding it into a computer program, but this wasn’t possible because we couldn’t find enough information about her process. Roc managed to find the music for the piece Wende 80 (‘Turning Point’), which had been performed by Trevor Dunn and Eyvind Kang. Although this provided some clues to the process, it was still impossible to reverse-engineer the music of Opus 17a to figure out the numbers, and the rules that had been used to map the numbers to notes.

So with the idea of programming defeated, I ran the MP3 through a plugin that would analyse the pitch and timing and output MIDI. The result had lots of mistakes, and was quite pleasing in a way. I’d used this inaccurate conversion process before – in the piece Mirror in the Mirror (on the album Symmetry-Breaking, 2011), which was based on analysis of Spiegel im Spiegel by Arvo Pärt. In that case, the errors were central to the piece, but EVOL’s Opus 17a required a good, clean transcription. Tidying up this messy MIDI output was laborious but it seemed like the only viable option. This meant removing all the obviously wrong notes, re-aligning the remaining ones in time, and gradually removing any remaining clutter and filling-in the gaps by listening to the original and comparing it with the MIDI version. Working in this way relied more on identifying the patterns by ear, then also beginning to recognise them visually as they were constructed in the MIDI editor. The output of the MIDI conversion became less useful as the new piece, which was set to a fixed tempo grid, drifted away from the original recording in which the performer modulates the tempo. It would have taken too long to do the whole piece this way, so it was restricted to the first 12 minutes, which is just over 1,500 notes. This 12-minute portion is cut off at a section similar to that at the very end of the piece.

Listening to Opus 17a in the process of re-constructing it – bit by bit, and seeing the pattern of notes build up in the MIDI editor – brought its musical structures into focus. The process of analysis and re-synthesis revealed that it has a semi-regular tempo based mostly on 4 beats and sometimes 2. There are lots of 4-note arpeggios that gradually ascend in pitch. Often there are ascending arpeggios interleaved with static or ‘drone’ notes. It is based on a fixed scale (F Lydian) with a range of just over 2 octaves, from E1 to F3. The first 64 notes of the piece are based on this pattern of ascending arpeggios:

Opus17a-MIDI-1This image reveals how the 4-note arpeggios are related: First the middle two notes are raised, then the first and last notes are raised. The first 4 arpeggios are: {{F, A, C, F}, {F, B, D, F}, {G, B, D, G}, {G, C, E, G}}. The root note F often appears to be the lowest note in the piece, as in this section, but occasionally a note below this (E) is sounded. Numbering the lowest F as 1, the sequence for the first 16 notes is: {{1, 3, 5, 8,}, {1, 4, 6, 8,}, {2, 4, 6, 9,}, {2, 5, 7, 9}}. With this numbering scheme, the low E would be numbered 0, and the highest pitch (F3) would be 15. It’s unlikely that Darboven used this method, because it’s not easy to see how the numbers 0 to 15 would map to the calendar numbers in the visual version of this generative artwork. It means that even though we can’t know Darboven’s generative process, this analysis does shed light on the generated structure. In the section pictured above, this process of shifting arpeggios continues until the pattern is just a step away from one octave above where it started. The next 64 notes in the piece look like this:


That section shows some of the variations on the main 4-note arpeggio, including a 2-note variation at bar 19. Starting near bar 28 is another type of pattern with ascending notes alternating with a steady pitch. Sometimes these ascend from the steady pitch, or descend from it, or pass through it from above or below. The whole piece has some symmetry, with the densest clusters of higher pitches appearing in groups at the start and end of the piece, and fewer in between.

EVOL’s début performance of Opus 17a opened the Unsound festival in New York, along with Oren Ambarchi who performed the epic Knots, from the album Audience of One. A review in the New York Times described how

Mr. Jiménez de Cisneros unleashed a prismatic extended tone at excessive volume. He shattered that core sound into jagged rhythmic clusters, each lacerating note accompanied with a piercing strobe light flash.

Roc’s performance was recorded but the quality of currently available clips isn’t very good, so I’ve removed the link until a decent one is available. It’s interesting to note that the variation in tempo that had been introduced in the original instrumental performance, then straightened out in the analysis and re-synthesis process, is now taken to extremes in this computer music.

This entry was posted in Art, Audio, Complexity, Music, Visual and tagged , , , , , , , , , , , , . Bookmark the permalink.

4 Responses to Opus 17a + EVOL

  1. Pingback: 2015 music | Aesthetic Complexity

  2. Hein Haak says:

    I was reading with great interest your article as a scientist and cello player.
    Is the score of opus 17 a, b and 18 a,b available in any way?
    I would like to play it!
    Hein Haak

  3. Guy says:

    Thank you, Hein. Unfortunately I don’t have scores for the opuses. But I do have scores of Turning Point / Wende 80, which I would be happy to share if that would be of interest to you?

    • Hein Haak says:

      “Dear Guy, I would be very interested in the scores of Wende 80 and Turning Point! Perhaps you know that Wende is currently exibited in the Kröller-Müller museum in the Netherlands. Besides that, to analyse, and play the score could shed some light on this somewhat misterious Music. Hope the hear from you, All the best, Hein Haak

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

This site uses Akismet to reduce spam. Learn how your comment data is processed.