Timbre and ‘spectral’ colour – just so we’re clear…

I thought it might be wise to actually outline what the point of this blog is, as I’ve actually launched into it without a great deal of fanfare – I also noticed that my last blog received some attention, especially among composers and music buffs, with a few notions about harmonics and music intervals that aroused some interest, though probably posed more questions than provided definitive answers. One supposes that’s the nature of blogs – just thinking out loud and seeing what the world makes of it.

If I haven’t already mentioned it, I’m completing an honours year at the University of Western Sydney, specialising in music – this blog serves the dual purpose of both consolidating my thoughts about my research and also serving as a record for it. This may be an unconventional way to work, but I figure it can’t hurt to put it out there, and it’s been gratifying to receive a response or two from people who are as obviously into music as myself – thanks to John for his feedback on my thoughts about composing in fourths and fifths.

The basic tenet of my research is looking into the importance of timbre, and what I refer to as ‘spectral colour’, in music, especially as a compositional device. Although timbre can be easily defined as the quality which gives an instrument or indeed any sound-making device it’s distinctive ‘character’ (i.e. what makes a violin sound like a violin and not a trumpet, for example), ‘spectral colour’ tends to mean different things to different people. For serious music types, it indicates a school of musical thought that emerged at the Institut de Recherche et Coordination Acoustique/Musique (IRCAM) in Paris in the 1960s and 70s, usually associated with Gerard Grisey and Tristan Murail. For the uninitiated, Grisey and Murail took as their musical inspiration ‘sound evolving in time’, which appears to be kind of self-evident, until you realise they literally took the information from Fast Fourier Transforms (spectral analysis) of particular soundwaves, and then applied that across an entire orchestra. All well and good – a little bit like extreme macro photography in sound. And granted, some of this stuff is kind of interesting – take Grisey’s Partiels (1975) for example:

In case you missed it, Partiels is based on a FFT analysis of a low E1 (41.2 Hz) note on a trombone. If you can listen to the entire piece (and this is only part 1, incidentally), you’ll eventually hear all the high order ‘harmonics’ appear, all those screechy notes high up, mostly in the upper woodwind section of the orchestra. And there was plenty more where this came from – Tristan Murail’s Gondwana, Grisey’s Transitoires, and later, pieces by Iannis Xenakis, Iancu Dumitrescu and other names I can’t pronounce. But the one thing that stuck in my mind from listening to most of these pieces, was how downright scary and unpleasant many of them sounded, and how little they related to the idea I had in my head to what ‘spectral music’ was. I applaud the extraordinary effort that went into writing these pieces, and the analysis that I’ve read on some of them thus far is impressive, if exceedingly complex (try Francois Rosé’s essay Introduction to the Pitch Organisation of French Spectral Music for a bit of light reading). But I can’t say I enjoyed listening to many of them, and what I interpret as spectral music is a very different kettle of tuna altogether.

Whereas the French spectral school and its proponents seem to prescribe to reproducing the harmonic series to varying degrees by seemingly very complex and elaborate means, what I interpret as spectral colour heads in almost the opposite direction. When I hear the buzz of harmonics high up on say, an Indian tanpura or sitar, that is what I mean when I say spectral; when I hit a key on a piano with the sustain pedal held down, and strings in integer ratios of the first vibrate in sympathy, I consider this spectral; when I grab my trusty Epiphone Casino electric guitar and encourage it to ring and feedback when plugged into an amplifier, that to me is spectral; when I overdub two or more vocal lines of the same note on a multitrack recorder and they slip slightly out of phase revealing higher order harmonics, this is what I mean by spectral. These are all methods, to me, of revealing the harmonic spectrum, or tone colour, in all it’s glory. And this essentially is the gist of what my research is all about – using any or all of these techniques and foregrounding them sufficiently to use as a compositional device. The trick will be whether I can make it interesting enough…

Timbre, of course, is all the constituent harmonics and other bits of audio grit that make up a sounds overall ‘character’, which is of course interesting and relevant – if you wish to take it apart like that. But it’s a bit like the science experiment in high school, when you had to  dissect a mouse, and take out it’s heart, and it’s spleen, and it’s liver, just to see how it worked. It was a perfectly healthy, happy (and alive) mouse, squeaking happily (or perhaps not) and now it’s just a pile of organs. I do understand that initially we have to do this to find these things out… but do we have to keep doing it, again and again? This is how I feel about some of the French ‘spectral’ music – it’s a bit like watching a car crash in extreme slow motion…

I think I’ll just play with the mouse and enjoy it for it’s own company. More soon…

Harmonics, fourths and fifths

I’m having several different ideas of where to go with this honours year spectral music project at present, and my thoughts have been a little scrambled. From where I was coming from originally, and trying to use a compositional method that related everything back to the harmonic series, I’ve been feeling a little boxed in – to compose completely in this style throws up two challenges:

• as pointed out in a recent paper, ‘The European Folk Scale’ (Hirt, 2013), if I’m to truly use the series as a basis for composition, notes ideally need to follow the order of harmonics as they evolve, i.e. in the key of C, it would follow C1, C2, G2, C3, E3, G3, Bb4, C4, D4, E4 etc. Any transposition down or up an octave is pointless, because there was no ‘octave’ equivalence on early instruments, one had to work with natural overtones inherent in the instrument.
• if I’m to be true to this method, I would really need to start writing in just intonation; apart from being particularly limited (as a guitarist and keyboard player, tuning to just intonation is more than a little awkward, especially on piano) and essentially limits you to only one key.

In a way I feel I’m confining myself, yet a core part of my reasoning behind this idea of using the harmonic series is to get back to some basic harmonic ‘truths’. When I first started investigating using the harmonic series as a compositional method, it was pointed out that I should look to the French Spectral School of composition, as the main proponents, Gerard Grisey and Tristan Murail, were chiefly concerned with representing the often very detailed information contained within Fourier transforms (a spectral analysis) of musical instrument tones. The results were, to my mind, interesting as they represented, in great detail, the partials evident within say, a trombone note (e.g. Partiels). Although initially very striking, I generally found these pieces both difficult to listen to (the upper partials being represented in the woodwind section of the orchestra became increasingly dissonant) and exceedingly complex: an analysis by François Rose, Introduction to the Pitch Organisation of French Spectral Music, was enough to convince me that this was not a path a wished to pursue. It was getting further and further away from where I wanted to go, which was more a preoccupation with tone colour and the compositional possibilities offered by the harmonic series, rather than a need to replicate it.

One thought that occurred was that I didn’t have to necessarily represent every harmonic. Some time back I decided that everything past, say, the 12th harmonic I could disregard, as the overtones were becoming so tightly packed that their usefulness became questionable, not to mention fairly unpleasant to the ears – these were also beginning to get out of the range of what could be physically ‘played’ (on guitar, it’s very hard to obtain natural harmonics past the 12th partial anyway). In this way, it occurred to me that it was perhaps more the phenomena of the series I was interested in, and my focus shifted to timbre, and tone colour. It also seemed to be particular tone colours I was interested in, in particular metallic sounds – electric guitars and string instruments, pianos, chimes, bells, even xylophones and marimbas. It was also the ring of overtones that got me, the sound one gets with open fifths and fourths, open ninth chords, which are stacked intervals of a fifth anyway, and… perhaps it was the intervals themselves contributing to this phenomena?

I tried some recording experiments with open guitar tuning back on March 5, which were interesting – apart from experimenting with slide rubbed back and forth across the strings in the manner of a violin bow (which brought about some natural harmonics) I also tuned the guitar to open fifths, in the manner of an Indian classical drone instrument such as a tanpura – although the original intention was to exploit the use of slide to find harmonic tones naturally, I gradually moved to using the slackened G string (now tuned to an E3) as a highly nuanced melodic string, with the other strings acting as sympathetic drones. Not unusually, this sounded somewhat like a sitar, and also not unnaturally considering the tuning, I started to come up with melodic lines that emulated sitar ragas, especially with the use of a flattened 5th and microtonal bends. I noted that if I abstained from using the 4th degree of the scale and stayed with notes contained within the harmonic series in order of their appearance, the resultant sound was very much like Indian classical music. Which was not exactly what I had in mind, but led to another idea…

On a hunch, today I re-listened to a number of pieces that I felt encapsulated to some extent the quality I was after. Aaron Copland’s Fanfare for the Common Man and Debussy’s La Cathédrale Engloutie, both struck me as exemplifying what I was actually hearing in my head, without knowing exactly why. Once I examined the scores for both, it started to become more apparent. Both pieces make extensive use of open 4th and 5ths, often consecutively. This gives, to my ears anyway, a kind of bell-like, ringing quality, and emphasises overtones without having to actually play them. I noticed this with my Indian-esque noodlings too; the buzz of overtones was quite apparent with the sympathetic open strings picking up on what was played on the ‘melody’ string. But the difference of course with both the Copland and Debussy pieces, were neither were chained to one key: both modulated several times, and both strikingly so, with some faraway excursions to distant keys. Evidently, it was the quality of the intervallic nature of the pieces that was attractive to me; once this particular penny had dropped, it became quite apparent in music where I had perceived this quality before: in the Georgian male vocal tradition, in some Bulgarian music, in fact a whole range of popular music which had started this investigation in the first place – the vocal harmonies of the Beatles and those that came after them, and in fact much of the Indian-influenced ‘raga-rock’ and psychedelia that appeared in the mid 1960s. Many of these musics employ the use of ‘stacked’ 4th and 5th intervals to produce a particular ringing kind of harmony, often against a common ‘drone’ tone which carries through a chord sequence – the result are chords that will often contain suspensions and a mix of sixths and ninths, often without the third degree of the scale, which lends them a floating, unresolved quality.

I suspect in some ways I have known this all along, and have in fact experimented with similar tonalities previously (see Grass, Desert, Earth and Air). But it has taken this long to confirm my thoughts, and I hadn’t realised the importance of the intervallic nature of the fifths and fourths – it also implies pentatonic modes (travelling by fifths/fourths eventually produces a pentatonic scale) and the cycle of fifths itself. So in a way, I’m tapping into something quite fundamental and possibly rudimentary… but it feels like I’m finally on the right path.

I will press on with a few more audio experiments… more later