Wednesday, 9 August 2017

The analyst’s toolbox

Making sense of musical information

In recent blogs we have considered the matter of complexity and simplicity in music. Most of us as well aware of the planning that goes into making a coherent pathway through a composition. Many of us have seen the Beethoven sketchbooks and his furious deletions in an effort to find the best of the possible routes to be taken to progress logically through a composition. A coherent pathway is for many listeners essential to the enjoyment of music, and coherence comes through repetition, sequences, confined tonal frameworks, adherence to convention and similar signposts. Without the guides that make a pathway a motorway many listeners give up on the music they hear, but some of us prefer the less well walked routes and a few relish the opportunity to take a risk and endeavour to encounter different landscapes.

Evolution has developed us into creatures that establish pathways through our daily experiences to guarantee our survival. In our time the evolutionary process empowers us by finding patterns of activity in finance, science, literature and the arts, and how we hear music. In order to clarify this let us begin with an extreme viewpoint and examine how we react to randomness. As randomness can mean different things to different people let us use the definition

The quality or state of lacking a pattern or principle of organization; unpredictability.

Randomness is the antithesis of the quality which for many is the essence of art and music, order. Cage’s introduction of indeterminate procedures marks for many the point where sound and music part their ways and all that was familiar becomes alien. However, if we examine our responses to the unpredictable without bias some fascinating aspects come to light.  

Humans are not well designed to accept randomness, we impose by instinct order onto random information and we filter that information to impose a narrative on our experience. One recent example of this revealed itself to the listeners of MP3 music on devices that offered random selection, the problem being that listeners considered the results insufficiently random. The selections were considered to be too closely linked. For an experienced listener one might argue that there may be limits imposed by the listener’s own collection of music, but whatever the restrictions because randomness lacks pattern clusters will occur. The problem for the technicians was how to identify sufficient differences to satisfy the listener’s notion of ‘newness’. The solution to the problem was to shuffle the entire content of the player so that each piece was played once and not played again until all others had their turn. If the owner had 10 different works he or she should have the full range during a run or car journey, but if, like me, you have a large collection you are likely to switch off your player and in this case the shuffle starts again.

So powerful is the characteristic of imposing order on events that it creates life-destroying characteristics, the “gambler’s fallacy”, the notion that a string of losses (denied expectation) means that a win has to occur. Translating this to the MP3 player is like saying “I should have the Tchaikovsky next” when there are several hundred composers on the device.

We are caught in an ancient dilemma, predictable patterns are desirable but we find rapid change both stimulating and worrying, we have to accept that there may be a tiger in the tall grass, as there is in Rousseau’s wonderful painting “Surpris!”

What happens when (if we chose to do so) listen to haphazard, unpredictable successions of sounds? The answer is that at times we will sense continuity which we will cluster into an event and at other times discontinuity, the alteration of these frames will impose over time a sense of order. If restrictions are placed on the origins or nature of the sounds the clusters will be more frequent, and the restrictions may be one of any of the musical parameters.

In the blog on simplicity diagram I marked a point at which musical matter (pitch, rhythm etc.) 'evolves' into information.

How does information arise, how do we recognise it, is it just a process of chunking different materials so that matter evolves into a more distinct form? The following table sets out to categorise the actions we engage in when encountering information, whether organised or not. To demonstrate its use let’s use the opening pages of Ligeti’s Etude No 4 for the piano, “Fanfares”.

Making sense of information

R.H. / L.H. activity
Main design
Rhythm / ostinato 3+2+3
Part to whole
Changing rhythm of dyads (constrained by ostinato)
Progressive change
Expansion of original dyad phrases (imbrication)*
Changes of register
Accents on dyads, suggested bar line downbeat
Bartok dances

*The imbrication or layering of the four lengths are shown by / marks:

First appearance                 

a 3+2+3+3 
b 3+3+2+3 
c 2+3+3+2
d 3+2+3+3 

Second appearance

a 3+2+3+3 
b 3+3+2+3 
c 2+3+3+2

b     3+3+2+3

a         3+2+3+3

Third appearance                b, c, d / c and then a / c / b / a

Though these combinations of 4 length units will not be noticed by the listener they are of some interest to the analyst at this point in that they may predict later events (or not).

We should notice:

A recurrence of particular clusters of information
Particular types of change
A pattern that reveals the rate of change
Some evidence in the opening pages.
Cyclical events
Stability or capriciousness to clusters of events
Lengths of longer phrases between LH / RH exchanges small scale changes

I found this particular approach useful when approaching the Feldman work for Bass Clarinet and Percussion discussed in a recent blog “Space and the composer’s toolbox” and with “Cartridge Music” by John Cage “What I do, I do not wish blamed on Zen”.

The difference between indeterminate music and organised music is that the regularity of what we perceive to be a cluster of events is going to be far higher in the latter. The process of articulation in performance plays a significant part in our perception of these clusters; articulation is part of the toolbox for both composer and performer, though in performance they may disagree with each other! Mahler’s use of markings on his symphonic scores should produce greater similarity, but experience shows there is great diversity. Articulation is so important in the process of creating information from text that it deserves a blog of its own.

I shall close with this observation on information:

A piece of information is considered valueless if, after receiving it, things remain unchanged.


  1. I used to live in front of the fire station. Every day at 11:00 AM those guys would come out and fire up the SAW. They would run it super hard and loud. They have to test the equipment every day. I didn't like it. I didn't like it one bit. Can't complain because they are the fire department and when you need them you really need them. There were also motorcycles all in a gang. All the motorcycles together makes me feel uncomfortable.