Logo notas.itmens

Michael Levin, Ingressing Minds

Ingressing Minds: Causal Patterns Beyond Genetics and Environment in Natural, Synthetic, and Hybrid Embodiments; File Here

A long read that for me the references were more helpful than what Levin himself wrote, since his writing seems heavily confused to me. He uses some fancy words without actually understanding them.

The point is nearly identical to what is discussed in the previous post “Can a Biologist Fix a Radio? ”

>The crucial point of the article is not that one should take the prescription that is in analogy to electrical engineering, which is to develop a formalized language that is predicated on the analogy between a living system and a electrical system, but that the fundamental operational logic may not have been grasped at all.

What's important here is the "fundamental operational logic". When we study circuits we rarely care about the physical, material properties of, say, copper cables. Circuits are physical entities, but how they work cannot be explained by studying their materials, but by studying idealized circuits, and it becomes mathematical. Mathematics is a general science of patterns and structures, it is mythologizing to say that Platonic mathematics "ingress" into physical space, there's nothing that fancy when we study circuits we study its structure and pattern rather than copper cables.

The actual problem, as I can see, is that we do not have a mathematics that can probe patterns and structures that are very complex. For now we only have empirical studies, done by people like Carl Jung, regarding these patterns. Furthermore, since these patterns are themselves alive and dynamically self-constructing and dynamically interacting with those who probe them, it is hard to imagine a way to study them, and what should we expect when we study them - what sort of knowledge can be obtained?

One thing Levin fails to notice that physical world is by itself highly agental and interactive, even the fundamental particles are not "what's there", they interact with the observers, and respond to them.

Some philosophical problems that have actual impact are not discussed, for example, is the history that led to the arrangement of matter that constitute an entity really important? It is known that microscopically it is impossible to clone any entity, but macroscopically this is still possible; what if an entity, as a specific arrangement of matter, is copied? Will it exhibit same, or similar behavior, to the original one? Certainly the two entities won't be the same, but they are close to each other. Also I need to read more about non-cloning theorem (quantum mechanics is a general theory of information - even physics is informational).

To Read#

  • “Ones represented by equations such as those describing gene-regulatory networks can even learn from experience:”
    • Biswas, S., W. Clawson, and M. Levin, Learning in Transcriptional Network Models: Computational Discovery of Pathway-Level Memory and Effective Interventions. Int J Mol Sci, 2022. 24(1). 100.
    • Biswas, S., et al., Gene Regulatory Networks Exhibit Several Kinds of Memory: Quantification of Memory in Biological and Random Transcriptional Networks. iScience, 2021. 24(3): p. 102131.
  • Popper, K.R. and J.C. Eccles, The Self & Its Brain: An Argument for Interactionism. 1984: Routledge.
  • More Bergson. Matter & Memory and Creative Evolution (again).
  • Thom, R., Mathematical models of morphogenesis. Ellis Horwood series in mathematics and its applications. 1983, Chichester, West Sussex, England New York: Ellis Horwood ; Halsted Press. 305. 130.
  • Pythagorean/Platonist stance in which some of the causal input into mind and life originates outside the physical world
    • Kastner, R.E., S.A. Kauffman, and M. Epperson, Taking Heisenberg’s Potentia Seriously. Adventures in Quantumland, 2017.
    • Tegmark, M., The Mathematical Universe. Foundations of Physics, 2008. 38(2): p. 101.
    • Tegmark, M., Is “the Theory of Everything” Merely the Ultimate Ensemble Theory? Annals of Physics, 1998. 270(1): p. 1-51.
    • Tegmark, M., Our Mathematical Universe. 2014.
    • Deutsch, D., Fabric of Reality. 1997
    • Ellis, G.F.R., Top-down causation and emergence: some comments on mechanisms. Interface focus, 2012. 2(1): p. 126-140.
    • Penrose, R., Shadows of the Mind. 1994.
    • Penrose, R., The emperor's new mind : concerning computers, minds, and the laws of physics. 1991, New York, N.Y.: Penguin Books. xiii, 466.
    • Penrose, R., Consciousness, the brain, and spacetime geometry: an addendum. Some new developments on the Orch OR model for consciousness. Annals of the New York Academy of Sciences, 2001. 929: p. 105-10. 

Scattered Remarks#

A general science of patterns. More Gary William Flake's The Computational Beauty of Nature. Complex entities, emergence, renormalization group…

Computer is certainly not just a bunch of circuits. The way the circuits are organized is not physical.

Xenobots. The collective behavior of chemical robots.

  • Li, F.e.a., Liquid metal droplet robot. Applied Materials Today, 2020. 19: p. 100597.
  • Points, L.J., et al., Artificial intelligence exploration of unstable protocells leads to predictable properties and discovery of collective behavior. Proc Natl Acad Sci U S A, 2018. 115(5): p. 885-890.
  • Cejkova, J., et al., Droplets As Liquid Robots. Artif Life, 2017. 23(4): p. 528-549.
  • Egbert, M., et al., Behaviour and the Origin of Organisms. Orig Life Evol Biosph, 2023. 53(1-2): p. 87-112. 

Also, what should we expect? Recall what is summarized succintly by Gary Williams Flake,

In a very strong sense, Chaitin 's result is also a statement about the ability of humanity to describe nature . Science as a whole has the goal of providing simple explanations for complex phenomena . Specifically , scientists strive to explain patterns by natural laws that are simpler than the observations . To do this , a scientist will often look for patterns and try to mold old explanations into new ones that fit the new data . However , this process of building explanations is another incarnation of the more general process of algorithmic compression . If an observation for a natural phenomenon is more complex than the algorithmic complexity of our identification procedure , then it is impossible to explain the data by anything more compact than the data themselves .

The actual problem, as I can see, is that we do not have a mathematics that can probe patterns and structures that are very complex. For now we only have empirical studies, done by people like Carl Jung, regarding these patterns. Furthermore, since these patterns are themselves alive and dynamically self-constructing and dynamically interacting with those who probe them, it is hard to imagine a way to study them, and what should we expect when we study them - what sort of knowledge can be obtained?

Mathematics and Biology#

This is not strange or surprising at all since whatever is structurally or pattern-related is mathematical. All these sayings centering around Platonic mathematical space seems bizarre to me, by the way. Platonic forms do not “inject” anything into physical events, it is simply that physical events are and can only organized in the specific forms described by mathematics, Levin's philosophy seems extremely confused to me. 

I propose that the objects on which we often fixate in physics, biology and AI - the embryos, machines, language models running on PCs or in robots, etc. are just pointers (or, per Hoffman, interfaces) to the deeper space of patterns.

What does the word-change indicate? What is a pointer? See Hoffman, D.D., M. Singh, and C. Prakash, The Interface Theory of Perception. Psychon Bull Rev, 2015. 22(6): p. 1480-506. and Hoffman, D.D., The Interface Theory of Perception, in The Interface Theory of 
Perception, Stevens, Editor. 2017. 

After all, the physical, material details of the properties of circuit materials such as copper cables, has little to do with how circuits work. The theory of electric circuits is that of idealized circuits. 

“Life makes extensive uses of the properties of prime numbers”: Cox, R.T. and C.E. Carlton, A commentary on prime numbers and life cycles of periodical cicadas. Am Nat, 1998. 152(1): p. 162-4. 

“Being amenable to liar paradoxes and other incompleteness results, gene regulatory networks are even subject to patterns from the field of logic”. Isalan, M., Gene networks and liar paradoxes. BioEssays, 2009. 31(10): p. 1110-5. 

Morphogenesis itself is a cognitive process#

  • Grossberg, S., Communication, Memory, and Development, in Progress in Theoretical Biology, R. Rosen and F. Snell, Editors. 1978.
  • Levin, M., Bioelectric networks: the cognitive glue enabling evolutionary scaling from physiology to mind. Anim Cogn, 2023. 

Is it possible that the relationship goes deeper, in that the core of what it means to be a mind, with inner perspective, embodied in the physical universe, is fundamentally linked to the kinds of autopoietic patterns a given construct can access? 

Form and agential behavior is a combination of ingressing meaningful information patterns and physical constraints in how it can manifest in the physical world determined by structural architecture, limitations of time and energy, etc. 

Some network structures have the simply goal-directedness and self-aseembly capabilities#

  • Serra, R., et al., On the dynamics of random Boolean networks subject to noise: attractors, ergodic sets and cell types. J Theor Biol, 2010. 265(2): p. 185-93.
  • Kauffman, S.A., The origins of order : self organization and selection in evolution. 1993, New York: Oxford University Press. xviii, 709. 

What does it mean that “Platonic patterns are themselves intelligent to a degree”?

Digression: Classical Liar Paradox#

In pp.14 Levin mentions Patrick Grim's perspective on the classical liar paradox, who claims that 

there is no paradox if we allow the truth value to change and consider the time-extended behavior; the paradox arises from our trying to freeze a fundamentally dynamic pattern down into an assumption that a proposition should have a static truth value. 

Hence what is proposed is that the statements in the classical liar paradox should include indexicals, basically Kripke's position. But this doesn't work, since there are other paradoxes that is self-referential but can totally be paradoxical even with the indexicals. For example,

  • Berry's paradox. “The smallest positive integer not definable in under sixty letters.” Maybe a way to evade this is that the integer defined by this sentence wasn't defined at all before this sentence appeared, but that seems like a way to say that self-reference is by itself impossible because by self-referencing the referencing entity is modifying itself.
  • “Suppose that my father asserts the mendacity of all one-legged men in town; suppose also that there is only one one-legged man in town who, unbeknown to us, has asserted the veracity of my father.” This doesn't in any way involves time, unless one of the assertion is made to only include what is not said.

What indexicals do here is nearly identical to what hierarchy does. For example Russell's paradox can be evaded by restricting sets to the cumulative hierarchy, and byproposing a hierarchy of semantically open languages; each language in the hierarchy has a truth predicate which can be applied legitimately to the sentences of the language below, and only to those of the language below.

i.e. “indefinitely extensible”.