Last time I was in Tokyo I went to an excellent installation by visual artist Shiro Takatani and composer Ryuichi Sakamoto at the icc Gallery and as I just stumbled across a couple of videos on youtube I thought I would reminisce a little. Regardless of whether you understand spoken Japanese or not, check this out:
I visited the installation twice & each time stayed for an hour or more… incase you can’t tell from the video, the moving images are being projected into nine tanks filled with water & dry ice, and as the fog drifts the apparent depth of the image changes… Visually it was very hypnotic, but it was also quite unreal to walk into a darkened gallery with people lieing on the floor, quietly watching & listening this complex evolving environment…
But for me, even more mezmerising was Ryuichi Sakamoto’s soundtrack. Suspended with each of the 9 tanks were pairs of speakers, each of which produced a soundtrack that at times was unique, at other times related to sounds from speakers nearby, but was always evolving. If I spent a total of 3 hours in there listening, I would not have heard a single sound repeat and it intrigued the hell out of me as to how the soundtrack was composed/generated & reproduced…
Luckily a few weeks later I was in a bookstore in Osaka & in the music book section came across a very interesting book documenting the installation as well as including interviews with the two artists.. I’ll quote some of what Ryuichi Sakamoto has to say about his creative process relative to the project;
Secret of Sound
– And how did the structure of the audio aspect develop?
RS: ‘The first thing is that I didn’t want to bring it back to a linear vector. Which of course means that it’s alright to have it completely random, but at some point you want to have certain parts of the music matching certain image sequences. So when you create groupings of sounds and images you want them to end with certain images, and you wind up creating minimal guidelines, things to be avoided. In total, we’re dealing with about 400 sound fragments from the opera (LIFE was originally conceived & performed as an opera) Some of them have some level of musical coherence, and some are resounding crashing sounds. Just setting them off at random would be boring. Especially because I do like taking risks, this kind of became the objective. I don’t like ‘experimental music’ that’s only point is experimentation. So when i started looking at what might make interesting groupings, and this is slightly embarrassing to admit, but I started rereading Deleuze and Guattari’s A Thousand Plateaus. I don’t remember the chapter, but there was a brief mention of the classification of insects as corresponding to the classification of musical instruments. Eureka I thought. Natural history, natural sciences, taxonomy. This is the way to organise the work!’
– And just what kind of taxonomy was it?
RS: ‘In A Thousand Plateaus it’s the way that insects produce sound; ‘criers’ and ‘scrapers’ are quite different. If you follow this line of reasoning you see musical instruments which are ‘struck’, ‘scraped’, ‘blown’ or what have you. So this is how I began creating my own musical taxonomy. Once I had my major domains, I developed kingdoms, phylum, classes and so on. In the end I arrived at some 30 groups. With nine aquariums it meant that they should be organised into nine variations. Well considering that they are in stereo this would be actually 18. Anyway, there are nine playback systems. One thing that was interesting is that usually you listen to stereo as one thing, but with 9 aquariums created individually, even when you’re hearing the same sound, it doesn’t sound the same. There will be delays and other artefacts of the installation. Of course, there are examples in music, like Toru Takemitsu dividing up the orchestra into two sections for performances, and of course many others in post-war music, but this is the first experiment I’ve heard of using nine. Anyway it was really interesting working in this way.’
– Is this something you were able to simulate in your mind?
RS: ‘These days there are very convenient software tools like Max/MSP that allow you to place nine different sources in stereo headphones, and which allow for random sampling, and these enabled me to test everything first. I tried first playing each of the nine playback systems with seperate sound groups, or dividing them up into arrangements of threes, for example. I had imagined that mixing them up would yield more interesting results, but when I listened to the playback I found that the best results were when a given group was played syncronized or when the times and locations were offset. I remember saying that it was like a garden with 9 shishiodoshi (a Japanese garden water-distribution/temporal awareness system) runnign at the same time. Usually you have only one, right? (Laughs)
– Are there recurrences?
RS: ‘No there aren’t. Each morning when we boot up the Macs, 30 new random examples are generated. With 30 there are a lot of possibilities for recombination, and so each event is unique. The ending is decided though, and once the ending has been reached the program automatically shuffles the data so that a new pattern of 30 new groups is generated. If you’re playing one group of 30 there are times when a random cycle will be uninteresting so for this eventuality, there are secondary rules governing groups. or to say it more accurately, there are situations which it is better to create rules for. Situations wherein random play is interesting are left to run as is. If we say that within a given group there are 30 different sound files at play, they can be played at random, or in an enormous number of different sequence combinations. And of course there are nine different devices playing them. So in the end, there should be no time during the terms of the exhibition when the same sound will be playing.
– It sounds perfectly random, but in fact they’re not.
RS: The point is that they are not chaotic.
& just for the record, here is an example of a shishiodoshi (& for the train spotters, one appears prominantely in Kill Bill when Lucy Lieu loses both her final battle & her head)
which of course, someone just had to go & mash up:
I would quite like to hear an ambient dub track with a shishiodoshi though a space echo, but JPop gabba or whatever that is….hmmm… shame…