UPDATE: I’ve added a fifth skill that I’ve come to realise is very important, this is the art of mapping parameters.
I havn’t formally studied music in any way. But if I’d done some form of undergraduate course, this is what I would have liked to learn. I’ve learnt these things the hard way. Over years of making mistakes. I’ll start by discussing the points that I’ve felt I’ve spent the most time learning.
- Spectral balance. When designing sounds, the first and foremost thing perceived is how balanced the component frequencies are. Put simply, one might say: “make sure you’ve got a good mix of highs and lows in it”. This is the kind of thing taught at audio engineering schools (under the label of ‘mastering’) but appears to be unconsidered in teaching ‘computer’ music. I think this is a legacy mentality. Traditional instruments cannot change their timbre, so the spectral balance is left to the audio engineer. In electroacoustic composition, spectral balance is front and foremost. But this is not taught!When I hear sounds created by young artists working in electroacoustic music, I often think that the work could have doubled in quality and interest if someone had just said: “you need a bit more lows in that one” … or … “the high frequencies are a bit harsh, if you bring in some mid-frequencies that will take the edge off without hiding the texture you have created”.I would have loved to attend a course called Designing Sound. The first assignment would be to produce a sound piece with a restricted set of sounds supplied. The assignment would be marked on how balanced the final work sounds. Its not about composition, its about listenability.
The second assignment would involve creating a sound piece from 1 uncomfortable-to-listen to source sound. The challenge would be to apply effects and equalisation such that the final sound ‘sounds’ great.
A third assignment would be to ‘mix’ or ‘master’ a track, so that it sounds good on 3 totally different sets of speakers. This is extremely difficult but really forces the composer to realise that bad spectral balance can destroy a piece.
The last assignment would be to design a sound for an alarm clock. Or some such thing.
- Compositional forms. Its not cheating to use an existing and established compositional form! The great composers used forms. Forms are pre-determined structures that have been proven as vehicles for musical ideas. I always thought that a composer started with nothing and built musical empires out of their creative genius and thin air. Not so.Even (or perhaps *especially*) electroacoustic composers should start by studying compositional forms and their structures. From fugues to 12-bar-blues to serialism. I say this because there are no compositional forms (except perhaps musique concrete) attached to electroacoustic music, and so new composers are left with no quick-and-easy-access context to host their musical ideas.The assignments would comprise of 1 work created for each structure taught. The final assignment would be *choose any form you like* or *invent one* (and document it).
- Audio Synthesis. I remember being flabbergasted, years ago, when I discovered that I could control the exact position of a loudspeaker’s membrane by sending the sound-card a number between 1 and -1 … where 1 made the membrane sit forwards and -1 made it sit backwards into its box.Made me realise that sound is just air pressure. And a speaker just pushes and pulls the air in front of it by moving a membrane very quickly.Audio synthesis involves manipulating numbers between the range -1 and 1, in the shape of waves.
I spent years trying all sorts of strange things in PureData (an audio processing environment), with typically atrocious results, before realising that sound is pressure waves and there was not much point manipulating the numbers without considering them as numbers-that-describe-waves. Manipulate the waves, not the numbers!
There are many things that the audio software environments let you do, but which don’t make any sense from an audio perspective. You gotta know what those are if you want to manipulate sound.
- Improvisation and structure. I played the tenor saxophone for over 10 years. Mostly interested in Jazz to start. But I played a lot of free Jazz and realised years later that doing free Jazz is very very difficult. Its very easy to play free Jazz, but very very difficult to do it well. There’s a reason for that.Most electroacoustic improvisations (that I have heard) are free. And not done very well.Free improvisation has no structure, and so is subject to the ravages of random hazards like an undigested lunch. The audience are not interested in the expression of indigestion.
It requires enormous amounts of experience and practice to be able to pull off an entirely free improvisation. Its not a good place to start. It *is* a good place to end.
I think it is important to learn how to define and invent structures that act as contexts for improvisation. Especially for electroacoustic music, which has no or few structures associated with its culture and practice. Electroacoustic music has no standard chord-patterns, or bar-counts or well known tracks. The performer must learn to define musical constructs which could be based an any of electroacoustic music’s parameters …. including timbre.
Exercises to support this would involve creating several structural constructs for the one improvised piece, and then attempting to understand which worked better, in which way, and why.