Information theory, randomness, redundancy, coding, noise, chaos

Chomsky tried to apply information theory to linguistics. As far as I know, few have talked about artistic aesthetics in terms of information theory. Recently I’ve been reading about Alan Turing, and the development of information theory during the Second World War and up until the seventies, during the development of the computer. It made me start thinking about contemporary music in terms of information theory. I remember in the seventies, when I was studying at the University of Illinois, we would talk about how extremely organized serial music began sounding like completely improvised and random music, for example the so-called events that someone like John Cage used to create. That’s probably a result of the fact that serial music was not interested in conveying a message, but in conveying structure, so that if every relationship can have a meaning, then there’s no hierarchy, and where there’s no hierarchy, all relationships are equally interesting or uninteresting. In other words, everything is random.
The point being that redundancy in language is often necessary in order to convey meaning, because there’s often so much noise in the signal. For instance, if you’re on the telephone, there’s always a siren or an airplane passing, disturbing the conversation. The analogy is made to secret codes, in which a lot of noise is added to the message so that only the person who has the key to the code will be able to understand the message. One can also put it in terms of the most efficient manner of conveying a message. When telegrams were the most common means of long distance communication, a language was developed similar to what we now see with text messaging in order to convey what would be a natural language in a shorter language full of codes and abbreviations. In this case it was in order to save money, but it’s still a law of information theory. There’s a direct relationship between clarity and efficiency, and meaning and information. This can be applied to poetry and prose as well. Take for example the sentence: ‘The birds are singing cheerfully in the forest as the sun rises over the mountains’. I would estimate that about 50% of the words in that sentence are redundant. If I were to write ‘birds singing cheerfully in the forest, sun rising over mountains’, the meaning of the sentence is the same, but the syntax is closer to telegram style. If we were to go one step further, then we get ‘birds singing, forest, sun over mountains’. And suddenly the message is poetry. The question is then: what is the difference between prose and poetry? Prose attempts to convey a clear though redundant meaning, whereas poetry conveys a meaning that is open to one’s own interpretation and fantasy. We could go a step further and make a “secret message” out of the text: “brdsngforstsnomtns”. The complexity of the message has now increased so much that there is too much noise and not enough meaning.
Somewhere during the last century, artists and especially composers became wary of traditional language and wanted to create music in a language especially created for each new composition. For the audience, this often gave them the feeling of being clueless and lost in an unknown world.  Music became a “secret code”, only those who had the “key” to the code could find meaning in the music. The lesson would seem to be that if one wants to convey meaning in music, redundancy is a necessary element. In other words, the individual elements of music – melody, harmony, rhythm and timbre – need to support one another.
Melody is the carrier of meaning. The redundancy that will make this meaning clear is in the relationship to harmony, rhythm and timbre. The closer they come together, the clearer the meaning. But beauty lies somewhere in the middle. Between simplicity (prose) in which melody, harmony and rhythm completely support each other, and complexity (secret code). The balance between these two extremes needs to be rediscovered. There’s still plenty to be said.

2 thoughts on “Information theory, randomness, redundancy, coding, noise, chaos

  1. Hi Jeff! Interesting thoughts… You might like to look up some of Fred Lerdahl’s work – particularly his article “Cognitive Constraints on Compositional Systems”, which was published in CMR some 25 years ago.

    Redundancy can be a potent term in a discussion of these issues, but I happen to believe that such redundancy is already apparent in creation (eg. in sheer sound), should we choose to perceive it. As a composer, I can try to articulate or reveal some of the exuberance of the sensory world. Maybe that will make this choice easier. I don’t know.

    When I enjoy a piece of serial music – or any music for that matter – it feels less like successfully cracking a code than perhaps witnessing a natural occurrence, like observing a skein of birds. It’s more a touching phenomenon or percept than a successfully conveyed message. For a case in point, I find myself dedicated to the idea that a single sound can carry as much ‘meaning’ as a melody. I can’t be sure if this is a sensible idea, but I don’t see any other way to find out than by making and listening to music. (Probably most of these constitutive ideas are implicit in my mind.)

    But more broadly speaking, I think talk of ‘meaning’ only makes sense with reference to a community that bestows it. When we use the word, a certain solidarity of opinion seems to be implied. So if there are people out there with whom my music somehow resonates – as there happen to be – that’s good enough for me. I think that’s how these things work. So, in my view, looking to perception to anticipate aesthetic preference or success is to put the cart before the horse.

    • Hi Tylan,

      Thanks for your thoughts. I will certainly look into Fred Lerdahl’s writings. In regard to what you wrote, I think though that one has to make a distinction between the audience and the creator. As a creator, it is imperative that one create a structure (though this does not necessarily have to be conscious). An audience listens to music and is not necessarily interested in cracking the code, as you so elegantly stated.
      Of course there is meaning in a single sound. John Cage taught us all of this, as well as what you said about meaning being bestowed by community agreement. Perhaps what I was trying to pin down was that there seems to be a middle-ground somewhere in which some kind of intrinsic meaning becomes apparent, intrinsic in the sense that it is apparent to, well, a community. Which is also the point you are making. If you are appreciated by some people, then all is well. I was more interested in what exactly that sliding scale is though: between information and redundancy. Signal to noise ratio and meaning. It is a tough subject especially since there isn’t a meta-language to discuss meaning in language. Perhaps music history is that commentary, but that is another subject I’m afraid.

Leave a Reply

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

You are commenting using your 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 )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s