ALife day 2

Day 2 of ALife, and more great science!

First was Ezequiel Di Paolo’s keynote on “Gilbert Simondon and the enactive conception of life and mind”. There are a lot of French philosophers whose work is relevant to complexity and systems and ALife who I have heard of but not read (Gilles Deleuze, Maurice Merleau-Ponty, Edgar Morin), but Gilbert Simondon was one I hadn’t heard of (and so presumably haven’t read, either). As I understand the presentation, Simondon was critiquing the duality of matter and form: form is intrinsically embodied in the rich physical material. Examples range from bricks, turbines, to life itself. The systems more and more exploit the rich embodied properties of their material embodiment. There is a process of individuation, or taking of form, but it is the process, not the end-point, that is key. There is a pre-individual state full of potentialities, which are exploited during the individuation process, and the fully individuated system has no more potentialities, and so is “dead”. Matter is more than mere stuff, it includes potentialities, transformations, operations, and changes. There was a lot more in the talk, all fascinating. I may have to start reading this particular French philosopher!

I didn’t get to go to any of the morning technical sessions, as I was in a committee meeting. So the next thing, after perusing the posters in the reception area, was the next keynote, Alexandra Penn’s, on “Artificial Life and Society: Philosophies and Tools for Experiencing, Interacting with and Managing Real World Complex Adaptive Systems”. Alex described how her group was using a participative approach to systems modelling, including diverse stakeholders. The models built are deliberately rough and ready, partly because there is no hard data, but partly to make it easier for the stakeholders to challenge them. Modelling allows for the discovery of relevant factors, and for diverse stakeholders to appreciate each other’s concerns. Analysis of the resulting models then allows the exploration of scenarios and identification of system levers. Since the system will respond to manipulations, there needs to be a continual modelling and monitoring process. The metaphor is system steering, rather than system control.

Then it was off to the snappily titled “Synthesising Concepts from Biology and Computer Science” (SCBCS) workshop. This was a bunch of short presentations about potentially suitable areas for writing review articles: diversity, fitness, open-ended evolution, self-modification, plasticity, modularity, recombination, co-evolution, Each of these areas is important in computer science (“natural” computing) and in biology. What could each discipline learn from the other? I find review articles that engage in deep synthesis some of the most valuable publications: they bring together small patches of research, and actually build the subject area. A good review is not a mere “annotated bibliography”: it is a constructive part of science itself. I am not qualified to contribute to many of these proposed articles, but I certainly want to read all of them! If only half the proposals were to be taken forward, it would be an extremely valuable contribution to the relevant disciplines.

So, another day, another ton of thoughts to process.