Some thoughts on recording braneworlds

Getting close to the finishing line for the PhD (submit on March 1), hence I’ve been a little more occupied with writing the dissertation than with writing blog posts. There are, however, a couple remaining posts to write before I submit. The first is about recording braneworlds.

The first thing to say is that because braneworlds is split into four groups each with their own clicktrack, it is very simple to record the work in individual groups and then piece it together. In fact, with the clicktracks all loaded into the music recording software, we could record everybody in the one file so that the compiling was already basically done. In general, groups did not need to hear each other for tuning melodic unisons or matching tones, rhythmic nuance, etc, so there were only occasional sections where different parts were fed into the headphones so that the musicians could hear what was going on in the rest of the recording.

This way of approaching things meant that the recording could be done group by group, region by region, ensuring that we got the best takes and enabling us to have maximum control over the levels of the respective groups and over the eq, spatialisation, etc. It could also lead to a kind of artificial feel, especially in the transition between regions, but that can be corrected in the editing/mixing stages.

Besides, the work has a kind of ‘artificial’ feeling about it anyway, and I don’t feel the recording needs to really attempt to represent at live performance at all, but to give another access point to the identity configurations established in the work. It’s not that the recording is necessarily a more ‘correct’ representation of the work either, since while it is true that many aspects are clearer in the recording, the performance also offers ways into the relationships not offered by the recording (visual associations, unified space, etc). I plan to spend some time in March mucking around with spatialisation, eq, and effects to see how much the relations between the parts can be highlighted in post-production.

But for now, listening to the rough mix that we just put together yesterday with Mark Smith, our sound engineer, there’s a few observations I’d like to make.

Firstly, the piece has a focus and clarity well beyond my earlier works in the folio. My feelings after the performance were a bit mixed. I thought that overall the work was a ‘success’, but that there were numerous sections that didn’t really work. With the extra rehearsal time, the musicians becoming familiar with the piece and the clicktrack, and the benefit of the studio, a lot of the stuff that I thought didn’t quite work (e.g. Regions 12, 13 and 14 in particular) actually work very well. So I’m pretty happy overall with the work. I’m still not 100% convinced of the major-minor triadic sequence that pervades the piece, since such familiar harmonies have a tendency to dominate the musical space, and since the alternative harmonic logic is not sufficiently transferable to other instruments, so it is even less capable of mounting a challenge to the hegemony of the chord sequence. But within the logic of the experimental system that is braneworlds, it works pretty. The chords are kind of funny too; melodramatic in a sort of Hollywood way.

There is some great ‘group polyphonic’ textures in the work, and in fact I wish now that there was more of this. That’s one thing I crave in the work and it never delivers quite enough of it: properly materially dissociated textures. For example, Region 4 has a really strong polyphony between Group I (flute), Group III (clarinet and piano) and Group IV (guitar and cello). Flute 1 plays a really simple, rhythmically stable line of held notes in the upper reg, and the instruments of Group IV play a punchy, lower register unison line of staccato notes, while Group III plays a flowing heterophony. There’s a kind of oblique motion going on too between Group I, which meanders around the high register, and Group IV, which descends from the middle register to the lower register across the section. This is a great moment in the work, but it passes by in 23 seconds.

Likewise the polyphony at Region 5 is really enjoyable. The two groups (II and IV) are both slowing down, but at different tempos and at different rates, and it gives this kind of sea-sick feeling and is constantly forcing the listener to pay attention. I also think Region 12 is much more interesting than I first thought. At this stage, the mix needs to be adjusted, and a little bit of editing as well, but I think the three different groups (I, II, and IV), very registrally distinct, offer a pretty wonderfully confusing rhythmic and textural moment.

One thing to be said about these textures is that much of their material is quite simple. It is just complex enough that the material of each group doesn’t become redundant, but just simple enough that the distinctness of the individual groups doesn’t disappear into a complex and undifferentiated swirl. The registral dimension is also super important to the polyphony of these sections. More of this in the work would be enjoyable… something for future works.

Something that works in the recording is the approach to common gestural shaping between ensembles that I did via my time plots (see this post). After the performance I was worried all that work might have been a waste of time. But with the tighter playing in the recording, these aspects really come out in a number of sections. In particular, Region 5, where the accented upward arpeggios in Group IV connect with stronger attacks in Group II; Region 6, where the metric structures and tempos between Groups I and II are totally different, but where the phrases begin and end at roughly the same point because I worked this out in the time plot; and Region 14, where the piano and flute have clear connecting gestures.

One thing about that is that often what is only a nearly synchronised attack between groups is perceived as a rhythmic unison because it is in such proximity and the brain decides that near enough is good enough. This is a good argument for working with slightly simpler ratios (rather than always prime numbers) in future works, since rhythmic unisons are going to be present anyway in the perceived work, they may as well be written into the work in a more controlled manner.

The one section that still doesn’t work, unfortunately, is Region 15: the flute and drums soli. My guess is that it seems unsuccessful because it is a rupture on the level of the sonic material (this is the first time this entire drum array has been used), and yet it doesn’t really pack the punch (in terms of density and complexity, or, for that matter, simplicity) that a rupture implies. It’s unclear how that could be fixed, but perhaps in future performances of the work I will revise this aspect.

Most of the rest of my criticisms of the work (particularly the fact that it is too sectional) still stand, or are even more obvious, in the recording. However, all in all, it makes for fun listening, and it’s a solid piece to end the PhD project on.