Monday 11 May 2020

Beyond Homeostasis: Some thoughts on biology, physics and cybernetics

John Torday, Peter Rowlands, Andrew Crompton and myself had a Zoom meeting today in which we talked about some fundamental issues in physics and biology. These have a bearing on thinking about education and development, and a particularly strong association to cybernetics.

Both Peter and John have theories about nature which reference a kind of recursive recapitulating symmetry in nature, from a fundamental original order, through to complex manifest biological and physical reality. There are differences between them in terms of defining what this original order might be: for John, it is a historical event, the Big Bang, and its associated singularity (although I gather from Peter that the singularness of the Big Bang is now contested). For Peter, original order means a totality of nothing in the universe (from Newton's 3rd Law), with the recursive and recapitulating mechanism driving a process of complexification in nature through successive levels of expressing the original nothingness at different orders of organisation.

Of particular interest in the discussion was John's view of epigenetics as a fundamental mechanism of evolutionary development through continual interaction and absorption of the environment by cells which exhibit levels of homeostasis at different orders of complexity. As cells seek to maintain homeostasis, they absorb epigenetic marks from the environment which steers the evolution of the species. The epigenetic marks themselves found their way into the environment from biological reproductive processes, fundamentally involving the sex organs. In other words, the old generation's expressions of epigenetic marks will lie in the environment to be picked up by the next generation, and in so doing, the ontogeny of the individual organism recapitulates the phylogeny of the species.

With regard to certain hormones this is very interesting. The balance between the androgens and oxytocin - the former causing "fight or flight" behaviour, the latter fundamentally related to generosity and love - shifts from youth to old age. The dominance of oxytocin in later years may help explain the growing warmth of the elderly - particularly in their attitude to the young. This, John argues, is not simply a behavioural shift - it is an evolutionary balance that serves to nurture the future of the species. I can think of many examples of particularly aggressive men who, in old age, find a new warmth of tone in dealing with the world (and people forget what complete bastards they were when they were younger!)

If this recapitulation of phylogeny is a kind of regulatory mechanism, then it raises questions as to how we are to think about things like homeostasis at all. Homeostasis is the maintenance of a stable state in a system in its environment - but it is a local phenomenon: homeostasis in maintained in local biological systems. But with epigenetics we are not talking about a local situation, but a broad historical situation where biological processes are spanning generations.

Piaget preferred Waddington's term homeorhesis - which is the tendency to maintain a stable flow, rather than a stable state. But that doesn't quite do either, because it lacks any explanation as to what might be driving a processes of homeorhesis.

This is where Peter's theory is so powerful. If totality is zero (or nilpotent in Peter's terminology), and local phenomena recapitulate this zero-ness by seeking to cancel themselves out, then it is possible to imagine that both the expression of epigenetic marks like oxytocin or the androgens driven by a principle of nilpotency at one stage in one context, and where their absorption at a later stage by a different generation is similarly part of a local process of trying to "cancel oneself out". And that process can then reproduce what appears to be a regulating mechanism connecting ontogeny and phylogeny. The key mechanism in this process is the creation of a selection mechanism for the organism that determines its behaviour according to how it believes its survival will be most likely: in other words, an anticipatory system.

More deeply, this means that our concept of homeostasis is too flat to describe these inter-generational historical processes.  As Conant and Ashby noted, every good regulator of a system is a model of that system. The third dimension of homeostasis, or indeed homeorhesis is anticipation. In their normal cybernetic conception, neither concepts have it, and because of this, neither can explain the underlying force for regulation. Moreover, anticipation itself can be driven by a nilpotent principle.

So we have to get beyond homeostasis. In the three dimensions that Peter's work takes us towards, our systems concepts look very different.

No comments: