Author: Psybertron

Subir Sarkar was interviewed by Sabine Hossenfelder last month, but I didn’t capture the link then:

Interesting content in the “Einstein was right when he said he was wrong” domain when it comes to the cosmological constant. Pointing to to some radically “new” ideas being needed to fix anomalies in physics. (That’s new as in old, but ignored.) But as I tweeted at the time, it is a fine interview anyway – proper respect between scientists with different disciplines of expertise and levels of experience.

Was prompted also to read this “Heart of Darkness” by Subir Sarkar on the same topic in a magazine called Inference. More spooky convergences, as “Active Inference” is this month’s topic in Cybernetics & Systems Thinking generally.

A quickie to capture this link:

Very impressed watching this recorded Web3 Foundation talk by Michael Zargham. He’s a name I came across from making contact with the “Active Inference Lab”. I already know Anatoly Levenchuck and Karl Friston on the AIL Advisory Board and discovered that Zargham is another board member.
(I’m intending to participate in the .edu domain of the AIL.)

The Age of Networks
and the
Rebirth of Cybernetics

Highlights:

• • • Very positive non-apology for focussing on many layers of abstraction above the bits & bytes. The essence of systems thinking is knowing what details to ignore in various levels of complex systems.
    • Very familiar recap of the history of Cybernetics starting from Plato (Kybernetes) via the Macy conferences. With “systems thinking” and network architectures front and centre of response to complexity.
    • Being comfortable with circular reasoning (Hofstadter for me). “Second Order” Cybernetics, positive as well as negative feedback loops. Future consequences are causal now. (There is active predictive inference involved – hence AI-Lab.)
    • Attention cost of participation (eg in social government). The more the “infrastructure” can handle invisible processes we don’t have to worry about, the better for us. Transparency is a distraction from what really matters. Noise means we always fall back to lowest common denominators. [See Mental Switching Costs]. What we need to trust is that the design of the decentralised system knows its own limitations.

(And, great to hear someone use that quote “All models are wrong, they simply have a valuable domain of intended use.” 3 decades (!) since I heard Julian Fowler use it.)

Anyway, a new “hero” (with no mention of John Doyle).
Connected on Twitter.

=====

Contrast with this pm’s talk with Iain McGilchrist’s elaboration of his “Sense of the Sacred”. Tremendous audience (and Iain) prejudice against “engineering” and machine language. These two domains just don’t get how close they really are. Same as deep thinking physicists being very close to the same sense of (something) sacred. In the Solms / Friston (bio-psycho) story, the turnaround of Damasio is telling, from the same prejudice against mechanistic algorithms to understanding the human subject involvement.

=====

 

The feeling of the brain being actively engaged with too many thoughts, to properly address any new issue, never mind any of the existing issues, is a common feeling – for me anyway.

Once you have several mental balls in the air that are connected to some strategy to get something delivered productively, it’s impossible to pick up a new one without letting drop at least one. [The other metaphor is the plate-spinning circus act.]

In correspondence today Richard Emerson coined the expression:

Mental Switching Costs

As well as reflecting the existing thought process above, that formulation instantly suggested its relationship to the (Friston) Free Energy Principle and all those systems-thinking consequences of Markov-blankets and active inference for living and sentient organisms. It’s all about efficient and effective use of resources, and when one of those costly resources is conscious attention itself, maximising which tasks can be left to the sub-conscious.

Isn’t it great when a plan comes together?

Regular commenter AJOwens (“Staggering Implications“) posted a very astute thought below my post on John C Doyle and Zombie Science.

Whether we see problems with “current” science as a bug or a virus, or simply the current state of ever-contingent, imperfect science, the switch to a new dominant view within science is of course exactly what Kuhn was talking about in his revolutions of scientific paradigms. And they’re always revolutions because – for whatever specific reasons – the existing paradigm naturally resists change. (I’d still say the current shift is special, somewhat meta, in that it’s about science not about any particular content of science. But he makes a good point.)

As an engineer / technologist I had always focussed on the techno-economic industrial paradigms (TEP’s after Freeman & Perez, previously Kondratiev Waves) enabled by advancing science, not the revolutions of or within science itself. Doubly meta here, because the current paradigm we’re struggling to get to terms with is the Electronic Information & Communications “wave” in human culture and economies more widely. This is quite distinct from the science and technology market-place that has enabled it, and quite distinct again from the revolutionary idea that information and communications may in fact be the very foundations of any kind of science.

Understood in [Kuhnian] terms, the “bug” in science is a very old one, and its roots are epistemological. All scientific research is conducted within a paradigm, but the paradigm influences what counts as “evidence.” Phenomena contrary to the reigning theory are at first not even noticed or recognized as important “facts.” If they become more persistent obstacles to current theory, they are explained away, dismissed as anomalies, or otherwise resisted. Eventually the reigning theory becomes so riddled with inconsistencies and beset with contrary observations that its very paradigm is overturned, and a new one is adopted which can accommodate the new evidence.

I believe we are in the middle of such a paradigm shift, and the work of people like McGilchrist and Solms and Doyle are part of it.

AJOwens, comment April 11th, 2022.

(And he goes on to suggest some other current sources.)

The point – we are in the middle of a Kuhnian paradigm shift – and being revolutionary, the process will have its downsides as well as its progress.

And this particular paradigm revolution is a complex, ubiquitous, many layered on multiple meta-axes. It is – or will be when it reaches a tipping point – going to be painful on a profound and grand scale. This is not just horse-drawn canal boats being replaced by steam railways. The e-Comms enabling is running full-steam ahead of the consequences in all aspects of humanity.

“The paradigm influences what counts as evidence.”

Indeed, as I’ve said before.
And resistance is futile.

Published this week, On Quality is a collection of writings by Robert Pirsig, prefaced and selected by his widow Wendy Pirsig, almost exactly five years after his death.

The Robert Pirsig Story

Apart from introducing us to Bob’s interest in the ubiquitous presence of Quality and to his two main writings, the books Zen and the Art of Motorcycle Maintenance (ZMM) and Lila, the preface also gives us “The Robert Pirsig Story”. Ironically, Wendy points out that most relevant parts of Bob’s early biography are to be found in the pages of ZMM, despite the “for rhetorical purposes” warning which led some readers, myself included, to research which aspects did indeed correspond with reality. I say ironically because for quite some time there was speculation, even a direct suggestion from Bob, that Wendy would one day write his biography. Here she gives us an eight page summary – including, despite its brevity, several newly public details – and lets us know that the selections in On Quality are themselves “loosely chronological”.

Previously published selections come not just from Lila and ZMM, and Bob’s paper Subjects, Objects, Data and Values but also from DiSanto & Steele’s Guidebook to ZMM and Dan Glover’s Lila’s Child. New selections come from Bob’s letters (to unattributed correspondents) and from his notes of the very few talks (*) he gave on quality.

[(*) Post note: In fact the whole of the introductory chapter “The Right Way” is a selection from the transcription of a talk he gave just a month after first publication of ZMM – now available in full here.]

On Quality

And the focus really is on quality. Whilst naturally acknowledging that his Metaphysics of Quality is elaborated within Lila, the multi-level “patterns” that form the full ontology – the model of evolved existence in the world -are not mentioned. Dynamic Quality, originally simply “quality” in ZMM, is the fundamental – radical empirical – essence of what is experienced.

‘Quality is just experience. It is the essence of experience of what is sensed. That’s all.’

‘It is not an intellectual category or any kind of thing that is independent of experience itself.’

RMP, Letter October 2, 1993

On Quality focuses on the quality monism itself and on its first division into static and dynamic, contrasted with the more orthodox subject-object split:

‘That line, “Without Dynamic Quality the organism cannot grow. Without static quality it cannot last. Both are needed,” is emerging in retrospect as the most important one in Lila.’

RMP, Letter September 4, 1993.

That statement itself pre-figures what today would be seen as fundamental to “homeostatic” models of life and consciousness in both science and philosophy, where all empirical knowledge is at root “affect” a categorically good or bad felt property before any more specific kinds of thing can develop in biology or in intellect. On Quality includes several references to Bob’s archetypal “hot stove” example of categorically good vs bad immediate experience. Elsewhere he went further and used also the classic “thermostat” example of what would be instantly recognisable as homeostasis today.

Also included in On Quality are selections from Buddhist texts where Bob saw parallels with his original quality thinking and was inspired that quality must indeed be fundamental and ancient, independent of Western scientific progress.

Still Important Today

Existing Pirsig readers – and there are millions – will welcome this sympathetic selection of “the most important” basic thoughts on quality from their source. For those readers, the notes from the few talks he gave form the bulk of the newly published material [(*) above and (**) below]. For a new reader who may have resisted the urge to dive into two rhetorical best-selling and cult road-novels from 1974 and 1991, On Quality provides a gentle introduction to their core thoughts, and may tempt you to follow-up on what all the fuss was about and why they remain important today.

=====

[For more on Pirsig from me on Psybertron, start from this summary page and follow links to my Pirsig Page short-term and the Robert Pirsig Association page (longer-term)]

[Post Note (**) one such large extract was shared on Literary Hub, by the publisher Harper Collins / Mariner Books.]

=====

Amazingly, after so many convergent threads on systems architectures and their fragility or resilience to well-placed viruses or bugs – in my agenda that “system” being the whole of science-led rational orthodoxy – I have may times used the systems thinking approach that complex systems (like politics plus media plus social-media) need moderation in the speed of communications in key layers. (As opposed to yes/no censorship of content.) We need this kind of thinking in order not to degenerate to lowest common viral denominators through social-media. (Sadly the prevailing “virus” is that any kind of moderation is a constraint on the much fetishised idol of “freedom of speech”.)

The threat / promise that Elon Musk likes Twitter so much he might buy it, in order to support that fetish, has got a lot of people thinking, and sure enough systems architecture thinking has gained a little traction:

When talking about how to improve social media, don’t focus so much on content moderation; focus on the architecture of viral dynamics. [full conversation coming soon] https://t.co/sxLiyibMD4

— Jonathan Haidt (@JonHaidt) April 15, 2022

Some of us have been watching the exact same cycle play out since 90s-era IRC chat room days pic.twitter.com/xdGsBIdhaT

— Jason Pargin, author of John Dies at the End, etc (@JohnDiesattheEn) April 4, 2022

Naturally, I enthusiastically agreed with both.

=====

Freedom runs on rails.
There are rules of engagement.

An information-and-computation-based metaphysics is fundamental to several recent sources (as well as my own metaphysics) – even if the specialist scientific researchers often don’t concern themselves with metaphysical aspects of ontology. Most recently John C Doyle and Mark Solms for example – information and its processes are simply taken as more fundamental than any other aspect of physics. The self-organisation and emergence of multi-layer systems and their architectures, Markov-blankets, active-inference, systems-thinking and the like then explain how more recognisable physical (and living and conscious) things arise and behave.

Sadly there’s a “bug” in orthodox science (Doyle) that rejects these descriptions where objective chains of causality appear to be broken by abstract patterns of information and which naturally bring the subjective aspects of consciousness into consideration. Empathising with the subjective is a “Rubicon” orthodox science needs to cross (Solms).

So.

“Minimal physicalism as a scale-free substrate for cognition and consciousness.”

– Chris Fields, James F. Glazebrook and Michael Levin

Is a paper from an August 2021 special edition of the journal “Neuroscience of Consciousness” – citing several authors already relevant here – and referenced earlier by Anatoly Levenchuk as being relevant to systems thinking and an information-and-computation-based metaphysics. I’ll say, (MP = Minimal Physicalism) it concludes:

In direct contrast with strict Cartesianism, MP holds that we can better understand our own awareness by understanding the awareness of our more basal cousins. Our homeostatic/allostatic drives and the mechanisms that satisfy them are phylogenetically continuous with those of prokaryotic unicells …

… The tradeoffs that we implement, and adjust in real time, between perception, memory, and planning are tradeoffs that have been explored and adjusted in niche-specific ways by all organisms throughout evolutionary history. We can take advantage of these fundamental mechanistic similarities to design theoretical and experimental paradigms that reveal and assess scale-free properties of consciousness in both natural and engineered systems.

Note:

“better understand our own awareness by understanding the awareness of” more primitive organisms … “both natural and engineered systems”.

No real review, just some extracted highlights which I can link to previous work here:

MP is Minimal Physicalism – That is, there are no physical assumptions beyond quantum information theory. (Not sure why “quantum information” specifically – but certainly information theory/ies more fundamental than anything else in physics.)

All physical interaction IS information exchange. (Agreed)

There is no Hard Problem. That is HP is not a problem to be solved, rather a set of inhibitions to be overcome. (Absolutely! It’s only orthodox science’s denial of subjectivity that gets in the way of explanatory understanding – see Solms above.)

There is no Combination Problem of psychist / subjective elements. (Ditto. Never was!)

“Bow-tie” systems topology. (Interesting. Something I’ve used in real world systems engineering before, and which Doyle’s work  often uses. Maximum diversity in higher and lower layers, minimally diverse, exploitable bottlenecks, in middle layers. Everything comes in threes, even individual layers.)

Markov Blankets (MB) – both Pearl and Friston forms covered. Also Tononi. (but no Dennett or Doyle). Also Boltzmann (1995 !*) Hacker, Heilighen, Damasio and Csikszentmihalyi (flow!) (* 1844 – 1906)

Not just Homeostasis (steady state) but Allostasis predictive of future demands.

Many testable “predictions” (the point of this paper?) … including
– use of “Quantum Zeno Effect” (Henry Stapp is also a joint reference)
– interoception and “the self“. (More Solms).

Prediction 15: The “self” comprises three core monitoring functions, for free-energy availability, physiological status, and organismal integrity, and three core response functions, free-energy acquisition, physiological damage control, and defense against parasites and other invaders. These will be found in every organism. Indeed they are found even in E. coli, which has inducible metabolite acquisition and digestion systems (Jacob and Monod 1961), the generalized “heat shock” stress response system (Burdon 1986), and restriction enzymes that detect and destroy foreign, e.g. viral DNA (Horiuchi and Zinder 1972). All of these responses act to restore an overall homeostatic setpoint, i.e. an expected nonequilibrium state; hence they can all be viewed as acting to minimize environmental variational free energy or Bayesian expectation violation. (Friston 2010; 2013).

Templeton funded – Christian religious funding an interesting feature common to much research that questions fundamentals of science itself.

Further to the previous post, let’s try and elaborate some specifics of what John C Doyle has to say. What is clear, after the throwaway “scientists will hate this” remarks, is that this is the reason he is unpublished in more popular journals and publishing formats. Because he is pointing out “a problem with science” he is experiencing resistance to getting published.

[It’s interesting in Gazzaniga, where I first came across Doyle, the most interesting read was the more autobiographical “Tales from Both Sides” which I originally understood to be a reference to the two sides of our bicameral-mind / divided-brain, but which was in fact a reference to the politics between researchers with unpopular findings and the story of whose work got published with which content, and Sperry who eventually won the Nobel Prize. I’m not, never have been, a conspiracy theorist. The institutional defence mechanism is a bug in scientific thinking, not some nefarious active conspiracy of secret interests. Essentially the bug is ignorance of the multi-layered architecture of “systems thinking” which is artificially flattened into one-dimensional logical objective “rationale”.]

It is a seriously degenerate problem, which is winning, because it’s self-reinforcing and we humans are poorly evolved to resist it. The bug is like a virus exploiting rational human weakness. Multiple timescales are part of the problem too – from speed of light global comms, to the pace of biological evolution, and the enormous range of calendar-based individual and collective human activities in between.

This is a disaster bigger than Anthropogenic Climate Change, not least because Zombie science, Zombie law and Zombie politics compromise our chances of successfully addressing it.

In this October 2021 presentation, Doyle gives us his take:

Ironically, Dan Dennett was one of those who used real virus-driven behaviour to illustrate issues the “Four Horsemen” had with religion in the early 21st C religion vs science wars. The classic example being a parasitic fungus that behaves virally, its spores infecting the primitive nervous system of a particular species of wasp, so that it not only effectively kills the insect, but changes its behaviour to ensure it is eaten by large mammalian hosts when it does die – a massive resource to multiply the virus numbers and spread them through the host population. A neat viral trick. (See Cordyceps if that’s not already familiar). Doyle uses exactly this example, and more classic variations – like the virus infecting mice which reduces their fear of predators like cats. Same propagation trick.

Zombies were a popular meme in philosophy – a thought experiment – about whether organisms’ (like humans’) behaviour reflected internal knowledge of what they were doing and why. How would we know if they had any internal sense of self? These virally compromised insects and mice also became known as Zombies for their so obviously self-disinterested behaviours.

Doyle’s contribution is Zombie Science or “Zience”
(Rough paraphrase 15 mins from ~28:30 to ~43.00 in this presentation.)

Vaccines – in the biological and social sense – are an example of a “Diversity Enabled Sweet Spot” in the enormous stack of human systems. As we have seen with Covid, the medical science is only one small part of the stack from Policy setting and enforcement, the medical processes and procedures, virus mutations, to the levels of individual and social psychology and behaviours. Many layered and massively complex, massively distributed asynchronously around the globe.

But that’s just the warm-up. Here’s the big thing.

Things are going wrong. And things are going to get worse. And almost everything we are doing with IT/Comms networks – like “Digital Transformation” – are actively making our problems worse.

We really need to understand fundamentally what is happening, not just anecdotally, individual examples.

Viruses exploit the universality of operating system architectures. And viruses rule – they kill HALF of their hosts everyday (most of those hosts are bacteria and other single-celled creatures).

As well as Viruses, we also have more active predators in our systems – Malware. Social Media is itself the most important Malware.

The awful thing about our most recent viral experience across all these levels is that it reinforces existing inequalities (race, wealth etc.)

Language itself is hijackable – it’s an important part of our operating system – we have many issues around the globe where exactly that is happening. Zombie memes. Contagious misinformation – false, unhealthy and dangerous. Previously ‘solved’ human rights and freedoms problems are coming back as well as new ones.

And science is not immune. Zombie Science.
It’s own self-correcting processes are not protection against the problem. Science will in fact reject all these multi-layer / diversity arguments. We are losing this battle. Good science is NOT winning the war against “Zience”.

Legal systems too. Laws and enforcement.
Zombie Law too. Unintended consequences. Zombie corporations, Zombie capitalism … endless.

It’s the architecture, NOT the individual viruses or humans.

I feel I’m fighting in the same trenches as John Doyle.

(Also note the significance of “Diversity” in the “Systems Level Synthesis”.
“Vive la Difference” as I so often say. Our systems will always have layers to be practically functional – fast and accurate enough – a single layer system can never work. But such a system will always have a “diversity enabled sweet spot” and many layers will be virtualised relative to the explicit layer in which “we” operate. These are vulnerable to viral attack, and we need to ensure we don’t lose sight of what matters in each layer so we can protect & manage them, not allow them to become Zombies.)

We need systems thinking – about the right things in the right layers in the architecture – not about all the “objects” (individuals) in the system and their direct logical / causal relations in the explicit layer. We need to consider and protect against viral fragility in the virtualised layers.

As in the preamble note above, the “bug” in science – and the reason Zombie Science is not helping us solve this problem – is that it rejects independent causality in multiple layers – flattens everything into one layer of explicit objects.

Good science
is NOT winning the war
against “Zience”.

Following the science can be dangerous.

=====

 

I mentioned John C Doyle as a candidate for a new real-life (living) “hero” in my research quest here in 2019 and again here in 2021. I say “new” hero because my long term hero has been Dan Dennett. Of course since then, both Iain McGilchrist and Mark Solms have taken up a good deal of attention with their own heroic contributions, but I mentioned Doyle again the other day in an exchange with Anatoly Levenchuk and “Systems Thinking”.

(Doyle is actually a reference in Gazzaniga who – like Sperry – is an important source in the cognitive science space, used directly by McGilchrist and many others, but not Solms so far as I can see. It’s how I first came across Doyle.)

I sensed, and still sense, that his work is going to prove important to how architectural systems thinking is applied to everything from fundamental physics to global human issues as well as brains / minds and IT/Comms networks. Trouble is, he admits, he’s very “disorganised” – unstructured presentations (oral and slides)  given to technical audiences and files on public shared drives. He’s prolific, but it’s all papers written with collaborators and students, no book(s) beyond his original control systems specialist field, with no obvious indexing or structure to his topics. In a sense that’s probably justified by the content of his current subject matter which demonstrates “universal” trade-off features of all multi-level systems – almost all his graphic abstractions are versions of each other.

I had already shared this presentation:
John Doyle, “Universal Laws and Architectures in Complex Networks”
March 2018 @ OFC Conference

Anatoly shared this recent one (as well as many papers, and in fact Doyle drops many paper references into his presentations, acknowledging his student contributions):
John Doyle – “Universal Laws and Architectures and Their Fragilities”
October 2021 @ C3.ai Digital Transformation Institute
(And this folder of public papers highlighting Social Science Architecture and Systemic Fragility.)

Now there is a thread of overturning scientific orthodoxy running through all the above, counter to the received wisdom of logical objective rationality of causal chains where wholes are reduced to summation of the history of their parts. In doing so, ignoring (a) ergodicity, that not just the end states of individual parts, but their network of paths through possible histories affects the whole outcomes, and (b) strong emergence, that wholes have their own properties and causes not causally determined by their parts.

At the level of political and aesthetic endeavours, no-one would bat an eyelid. The problem is, the problem exists in would-be science too and rational thinking more generally.

Dennett – warns scientific types to avoid greedy reductionism, and to suspend disbelief and hold-off on definitively objective definitions as rational arguments themselves evolve over repeated cycles.

McGilchrist – having debunked cortical & hemispherical misunderstandings of how our brains and conscious minds evolved to work, pleads for recognition of the naturally sacred beyond the reach of our orthodox (objectively verifiable) scientific model of reality.

Solms – having debunked cortical and mid-brain misunderstandings of those same brains and conscious minds and having established the basis of consciousness in subjectively felt experience and their evolved existence as distinct causal entities through fundamental information computation processes, makes a plea for objective scientists to cross the Rubicon to take-in the view from the subjective side.

Doyle – whose work arises explicitly from IT/Comms&Control computing networks, demonstrates repeatedly, with all manner of real-word examples (from talking or riding a bike to mobile apps & social media), that there are universal abstraction features of multi-layer systems network architectures that mean the virtualised wholes do more – better, different – than any of their parts. “Scientists will hate this!” he repeats in throwaway remarks to his technical audiences, recognising that strongly emergent causal identity of virtual entities is contentious for objective science, STEM and Engineering. (Slightly infuriating, unlike the better informed brain scientists above Doyle uses “cortex” as shorthand to mean that part of the human system inside our heads. The cortical fallacy.)

There is a common “problem with the received wisdom of orthodox science” running through all of this, and a lot of “systems thinking” and “information processing” common ground in where the problems arise.

It’s a “bug” in the received wisdom of “science-led” human rationality. The one that’s been driving this Psybertron project for 22 years.

We’ve barely scratched the surface with Doyle, I’ve mentioned elsewhere that in his terms this problem really is a bug. Viruses are especially adapted to hijacking vulnerable layers in multi-layer-architected complex systems, without needing to carry the overheads of more complex organisms such as ourselves and our social organisations. Humans are particularly badly adapted to deal with viruses that work against human interests – especially memetic ones in society’s information and communication layers. Our social systems – including science – are much more fragile than our rationality admits. Unless we want to give-up on humans and declare viruses and the simpler single-celled organisms as “the winners by headcount” in the cosmic game of evolution we need to find memetic vaccines that work.

(With Anatoly’s help) I need to dig further into Doyle.

Received and started reading Sexual Personae by Camille Paglia. The book that made her famous as a radical feminist but who also “identifies” as transgender and has been a critic of post-modernism’s consequences. Clearly someone of intellectual subtlety – and balls – on a topic that exemplifies our (21stC) modern polarisation predicament at a time when we desperately need careful discourse to make progress.

(Hat tip to Lila @commonclione for the recommendation).

Only just started the read, and already loving the style, so I expect I will digest the whole. Here an early sample:

Western love is a displacement of cosmic realities. It is a defense mechanism rationalizing forces ungoverned and ungovernable.

Sex cannot be understood because nature cannot be understood. Science is a method of logical analysis of nature’s operations … But science is always playing catch-up … Science cannot avert a single thunderbolt. Western science is a product of the Appollonian mind: its hope is that by naming and classifying, by the cold light of intellect, archaic night can be pushed back and defeated.

Name and person are part of the west’s quest for form. The west insists on the discrete identity of objects. To name is to know; the know is to control. I will demonstrate that the west’s greatness arises from this delusional certitude.

Our delusional certitude. Spot on.

The traditional contrast to Appollonian is Dionysian, but she uses “chthonian” instead – from the bowels of the earth. Being post-modern whilst criticising post-modernism is the trick. I call myself PoPoMo (post-post-modernist). The naming and classifying problem is my #GoodFences vs #IdentityPolitics agenda. Lots to look forward to.