On Contingency

Experimental Studio

09 04 2018

-------------------------

Luc Döbereiner, David Pirrò, Hanns Holger Rutz, Daniele Pozzi



One of Almat's formats consists in selecting a text which tackles topic of common interest and discuss it together with the artist in residence. For this meeting, Luc suggested an article by Yuk Hui entitled "algorithmic catastrophe: the revenge of contingency", published on Parrhesia – a journal of critical philosophy, no. 23, pp. 122-143 (2015). 


This text revolves around the idea that contingencies are fundamental for the development of technology. Hui affirms that the twofold nature of technics (to both overcome and generate contignency) in its relation to contingency can be understood as pharmakon, both as remedy and poison.

"Simulation and Computer Experimentation in Music and Sound Art"

21 03 2019

-------------------------

Luc Döbereiner, David Pirrò


LD

The types of computational processes that we are interested in have an inherently contingent character. They are not random, but they cannot really be predicted without performing them. These processes can die out, they can fail to get started and so on. In terms of contingency, we see a connection between computation and the aesthetic experience. So there's also a kind of openness, a kind of contingency to the aesthetic experience itself that we are interested in, a kind of over-determination of the aesthetic experience. While we are working we of course constantly navigate perceptually that kind of space, we have certain thresholds that interest us and we are always perceiving this, and reacting to that. So there's a kind of mutual sensitisation, because we are also trying to make these systems sensitive. Sensitive to other aspects that we perceive: sensitive to space, sensitive to our bodies, sensitive to other parts of itself. And at the same time we are learning to listen and to perceive it. So there's a kind of entanglement of computational contingency and the openness of aesthetic experience as something over-determined. 

Meillassoux

"We will call 'contingent' any entity, ting, or event which I know could be, or could have benn, other than it is. I know that this vase have not existed, or could have existed otherwise. I know that the falling of the vase could have not happened."

Talking about some aspects of "Contingency and Synchronisation", the sound installation Luc and David presented at the Orpheus Institute (Ghent) during the seminar "Simulation and Computer Experimentation in Music and Sound Art" (21-22 March 2019) . 


---------------------------------


LD

"An algorithm is open to contingency which occupies another order of magnitude". That's also a crucial point for our installation I would say.

 

DP

Yes, in a sense we square the contingency. But also not.. it's another order of magnitude, but I don't think it's necessary higher.

 

LD

No. But I mean also, for example, in the installation that you had in the esc gallery, taking one sample from the microphone, there is a link to a completely different order of magnitude. Also sonically, different from what we hear etc. 

 

DP

That is true. And how is it for us, for the installation we did in Ghent? It's different than this example, which is extreme. There I almost really used tuche. I really used reality as a random number generator. But in our case (installation in Ghent) the contingency we are thinking it's not really completely random, and we are not thinking something which is really completely random. There's a very special contingent property which is already in it, in the process, but it's not random. 

 

POZ

What's this contingent property?

 

DP

I'm trying to formulate what for me algorithmic contingency is. With a sense of algorithms as I use them, as recursive machines, is that there are things in the output, things being aesthetic artefacts that cannot be formulated in any way.

 

POZ

So like, the outcome escapes the formulation.

 

DP

Yeah. Contingency is something which is not formulatable.

 

LD

Not kind of direct, but just as a product of the process.

 

DP

It is a product of something, of a behaviour. This probably applies only to algorithms and computation. But I'm not sure. I would still try to understand something as contingent when I cannot formulate it..

 

LD

If you cannot derive the rule to do it, basically.

 

DP

Exactly. If I cannot derive any rule, if I cannot describe it. If any kind of formulation would be the mathematical words or pictures or whatever, contingent is anything that escapes all of this. 

LD

What is also interesting is also this being there, the presence. Which is also mentioned somewhere briefly. That there is a kind of absence to contingency. Something that hints towards something which is not present. Like a potential, like a phase space which is not completely actualised. Somehow, if you say it could be different you are already hinting to something that is not present.

 

DP

Well, that is probably the driving force of development. If you recognise a contingency, you always have already the necessity to expand the algorithm, or your reason, in a way to diminish this contingency, or understand, or include or whatever. The recognising already points to a lack in your understanding, or your reason. And to fill that blank, you have to fulfil another round of your algorithmic thinking.

LD

And you mean that is also because it could be different? I mean, our installation definitely at all the time could also be different. Because, even if things could be different, like if you listen to a symphony, to a Brahms symphony or something. It makes so much sense. I mean, you can think "ok, this could have been a little bit different". It basically tries to convince you that it should be like this, right? And I think a lot of music suggests that, gestural electroacoustic music and so on. Of course, you somehow fell that there are some arbitrary decisions behind that could have been made different. But what you are talking about is also the kind of perception of this as an aesthetic artefact, as you say. So how could we deal on the contingency at that level, you know?

 

DP

Yeah, probably you are right. That there's an expect to it that tells you: I could have been different. But still you would recognise me as the same. In our installation, for example, locally you have the impression that things could have been completely different. But still, you would recognise it as an overall behaviour basically. And still that thing would escape any formulation. 

 

POZ

But still you recognise the model of synchronisation, in a way.

 

DP

Well, it depends how you do it. Of course, if you look at the solution as the arrival point, then everything else is uninteresting. But if you look at how things get there, all the states in between, if you look at one state and the next one, locally they are all different, in any case. But still they have something that links them transversally, so that you can recognise something that is contingent to that computation, which is not in how you formulate it mathematically, it's not in how you formulate it algorithmically, that links all the possibilities.

 

DP

If you use, for example, the algorithm we used with the opposite of synchronisation, it can have a very clear behaviour, which is not always the same, and you cannot reduce it to the simulation that you used. You cannot really see it, I cannot mathematically nor algorithmic. But still it has something really contingent in a sense: it's not in the formulation but it's something you perceive as being there. So, I think contingency also need some recognition.

 

LD

Presence, also.

 

DP

Presence, and also being able to recognise it. You don't see something contingent if you don't recognise it. If you cannot separate it from the rest. You need the rest, in order to recognise what is contingent or not. So we need the algorithm, in order to recognise what is not part of the algorithm, of the formulation of the algorithm.