Eyre Lines

I’ve been reading and blogging about “How I found God in Everyone and Everywhere”, a collection of essays edited by Andrew Davies and Philip Clayton, for which there is currently the “Cosmic Campfire” book group, a crossover between Homebrewed Christianity and the Liturgists, studying the book over the next few weeks. If you haven’t yet read my first post, you should probably read that first!

This is not part of the chapter-by chapter reactions, however. It’s the result of one of those things – coincidence, synchronicity, divine providence – I don’t know – whereby I will be talking about a subject in one place on the internet, and in an unconnected place, I see something which puts things in a new light. In this case, in a post in the Partially Examined Life facebook group, John Shannon posted a link to an article in “Wired” about the work of Karl Friston. The article is headlined as being about the “key to real AI”. In the comments following that, Wilson Alexander posted a link to a paper in “Physics of Life” about free energy and the Schroedinger Equation. (The first of those is very readable and I strongly suggest that you go and read it; the second is really very technical… but John Shannon posted a set of very brief and excellent outlines of many of the technical bits, which he’s kindly let me steal).

So, what do those have to do with discussions about panentheism? Well, after I posted some of my earlier reactions to some of the chapters, I got into discussions with a few people (notably Paul Gideon Dann and Bryce Haymond) in the book group about idealism and panspychism. Keith Ward is an idealist; Rupert Sheldrake is a panpsychist. Several of the authors are process theologians, and process certainly seems to me at least to tend in the direction of panpsychism.

I’m definitely not an idealist; I’d definitely a methodological naturalist (in that I expect to find naturalistic explanations for things, i.e. explanations which are rational and scientifically testable), and I tend to lean towards thinking that if there is one most fundamental “stuff” of the universe, it’s material (which would make me a materialist), and that phenomena such as mind and consciousness can probably eventually be reduced to being emergent properties or epiphenomena of matter. That is very much the direction of Karl Friston’s thinking, which leans heavily on the Free Energy Principle.

So what is that all about? Friston considers that the hallmark of life is that it involves prediction of what is going to happen, and that it acts to reduce “the gulf between your expectations and your sensory inputs”, i.e. to reduce free energy. The main other key concept referred to in the article is that of the Markov Blanket, which Christopher Frith describes (as quoted in the article) as “a cognitive version of a cell membrane, shielding states inside the blanket from states outside.” Friston’s thinking is then described as follows:- “Each of us has a Markov blanket that keeps us apart from what is not us. And within us are blankets separating organs, which contain blankets separating cells, which contain blankets separating their organelles. The blankets define how biological things exist over time and behave distinctly from one another.”

Again, “When the brain makes a prediction that isn’t immediately borne out by what the senses relay back, Friston believes, it can minimize free energy in one of two ways: It can revise its prediction—absorb the surprise, concede the error, update its model of the world—or it can act to make the prediction true. “ In other words, there must be a feedback loop (or, more likely, several) which operate to minimise the difference between prediction and actuality by acting or by revising the prediction (and as an aside, I think this will link with the work of Douglas Hofstadter).

Moving on to the second article, John Shannon’s precis version is here in red, parts of the original which he quotes are in brown, bits I’ve added are in black, to aid those who find the original article hard going:-
1. Survival by minimizing energy expenditures through good guessing.
2. Wiggle room for mistakes is mitigated by accuracy in forming a good idea of what’s “out there”
3. “All biological systems exhibit a specific form of self-organisation, which has been sculpted by natural selection to allow them to actively maintain their integrity…” – Survival bubbles, cells
4. “…All other self-organising systems, from snowflakes to solar systems, follow an inevitable and irreversible path to disorder.”  – No internal world-in-itself. No emulation of importance if you will. No “being-for-itself” as Sartre would put it. And no feedback loops.
5. Attracting states = “…extended phenotype of the organism—its morphology, physiology, behavioural patterns, cultural patterns, and designer environments…” – Creature habits, probably including rest!
6. “The implications of this are profound. It means that all biotic agents move, systematically, towards attracting states […] -This means living systems are effectively self-evidencing—they move to maximise the evidence of their existence…” – Retreat to habits of regularity and regulation, as if seeking comfort in its niche. I picture something snapping back into some kind of neutral ground in order to save energy  – perhaps including the consideration that thought consumes energy.
7. “…all biological systems maintain their integrity by actively reducing the disorder or dispersion (i.e., entropy) of their sensory and physiological states by minimising their variational free energy…” -Don’t work hard, work smart… though a thermostat might be thought to do that…
8. “Thus, an organism’s distal imperative of survival and maintaining functional states within physiological bounds (i.e., homeostasis and allostasis) translates into a proximal avoidance of surprise…” -Intelligence (I might argue that intelligence as we understand it might need another feedback loop, but possibly no more than that).
9. “…this propensity to minimise surprise is the result of natural selection (that itself can be seen as a free energy minimising process; see below)—self-organising systems that are able to avoid entropic, internal phase-transitions have been selected over those that could not…” – Natural selection breeds in this necessary perpetuating propensity
10. “…one needs to differentiate between the system and its environment—those states that constitute or are intrinsic to the system and those that are not. To do this, we have to introduce a third set of states that separates internal from external states. This is known as a Markov blanket. Markov blankets establish a conditional independence between internal and external states that renders the inside open to the outside, but only in a conditional sense…” – Membranes as world-makers, emulators… They also act as curbs on the more extreme conceptions of relationality found in Process Thought.
11. “The Markov blanket can be further divided into ‘sensory’ and ‘active’ states that are distinguished in the following way: internal states cannot influence sensory states, while external states cannot influence active states. With these conditional independencies in place, we now have a well-defined (statistical) separation between the internal and external states of any system. A Markov blanket can be thought of as the surface of a cell, the states of our sensory epithelia, or carefully chosen nodes of the World Wide Web surrounding a particular province.” -Division of labor and also boundaries between levels in emergence.
12. “…free energy is a function of probabilistic beliefs, encoded by internal states about external states (i.e., expectations about the probable causes of sensory input).” – The need of communication
13. “then, how do Markov blankets relate to the FEP (free energy principle)? The FEP tells us how the quantities that define Markov blankets change as the system moves towards its variational free energy minimum (following Hamilton’s principle of least action) … In other words, an organism does not just encode a model of the world, it *is* a model of the world—a physical transcription of causal regularities in its eco-niche that has been sculpted by reciprocal interactions between self-organisation and selection over time…” – Heidegger’s “being-in-the-world” comes to mind here… I also note that those regularities (patterns) can be in an extremely simplified form – indeed, science tends to the simplest explanation which is consistent with the facts, and even to the simplest explanation which is sufficiently consistent with the facts for the purposes of prediction. I could regard this as an energy minimising process, the energy in question being thinking time.

At that point, John’s interests diverge from those which I’m following in this post. The article goes on to put forward a concept of the Hierarchically Mechanistic Mind (figure 4 in the article) which, it seems to me, does not require the panpsychist’s concept of “mind all the way down” (as opposed to turtles), and definitely does not make any form of idealism look attractive – indeed, the whole article describes a system in which ideas are effectively no more than guesses.

Backtracking in the article to just before it introduces the Markov Blanket concept (End of section 2.1), I find this:- “Thus, an organism’s distal imperative of survival and maintaining functional states within physiological bounds (i.e., homeostasis and allostasis) translates into a proximal avoidance of surprise [26]. Although surprise itself cannot be evaluated, since free energy imposes an upper bound on surprise, biological systems can minimise surprise by minimising their variational free energy. From the point of view of a physicist, surprise corresponds to thermodynamic potential energy [41], such that minimising (the average) variational free energy entails the minimisation of thermodynamic entropy.” (I have left in the hyperlinks which are in the original).

This, I think, is where we get to Process. Process theologians have a tendency to talk of God as continually introducing novelty, new experience, new possibilities. Or, if you like, surprise. And surprise is now another way of talking of entropy, which is arguably the most powerful principle in the universe. So far, so good (we may have a formulation of “God is Entropy”), but the whole line of argument through these two articles is demonstrating that life is profoundly surprise-reducing, and anti-entropic.

So, are we (as life forms and so devoted to reducing surprise, and thus entropy) fundamentally set in opposition to the Entropic God?

I think back to Genesis 1, with the earth being without form and void (chaotic) and the spirit of God moving over the waters and establishing some kind of order, and wonder whether the God of Order or the God of Entropy is being considered here. On balance, and taking into consideration that the Ruach Elohim is both the spirit of God and the breath of life, I would incline towards God being the God of life, and of order, if forced to make a choice.

But maybe, just maybe, God is both? For we panentheists, “there is nothing that is not God” is a meaningful statement; for mystics, the coincidence of opposites is something we tend to just have to live with.