Evolution as a Universal Learning Process

Initial reflections after reading Brendan Graham Dempsey's "A Universal Learning Process"

“Meaning arises out of the most fundamental dynamics of the physical world: energy and information…[I]t is precisely via the thermodynamic forces driving cosmic evolution as a whole that meaning evolves into more complex biological and social registers such as we experience as conscious agents…[M]eaning is a particular kind of information—namely, information linking entities with their environments in ways that causally impact their viability and flourishing.”

-Brendan Graham Dempsey

I’ve just finished reading

Brendan Graham Dempsey

’s important metamodern contribution to the emerging genre of “Big History” cosmologies: “A Universal Learning Process.” After spending some time with his lucid weaving of cutting-edge complexity science, it has become somewhat clearer to me where we converge and diverge on core issues. The following is a somewhat scattered attempt to throw some ideas up into the air that I hope to pick up with Brendan again in a few days when we meet for a Zoom teleportation session.

First, the convergence. I stand in complete alignment with Brendan’s overall argument that (as he summarizes it in the final pages) “Learning is the archetypal act of transcendence” (p. 85). I would also characterize cosmic evolution as a learning process from the very beginning, with an indefinite number of creatively emergent phase transitions thus far in evidence, and more likely to arise into the future. In other words, we have every reason not to assume that the universe is finished creating itself, as if the final layer of complexity had been achieved once and for all. Cosmic evolution is a process of self-transcendence. Whether or not God existed in the beginning, the World has since created God. Brendan’s claim that an ongoing leaning process characterizes all levels of cosmic evolution brings a smile to my face. Might we christen this approach “panmatheism” (from Ancient Greek μάθησις (máthēsis, “learning”)?

It is somewhat ironic, then, that the whole shebang is supposedly governed by entropy.

When it comes to OG “Big History,” it is hard to out wit Alfred North Whitehead’s Philosophy of Organism. The year after Arthur Eddington gave his 1927 Gifford Lectures at University of Edinburg granting entropy “the supreme position among the laws of nature,” Whitehead ascended the same lectern to add more to the cosmic story. I defer to him on many matters not because I believe him to be infallible, but because as a mathematical logician, physicist, and philosopher he was uniquely positioned to understand and integrate the evolutionary, quantum, and relativistic revolutions of the late 19th and early 20th centuries into a truly post-mechanistic vision of the cosmos. In his view, evolution is a learning process by virtue of the fact that cosmogenesis continually generates novel actualities that transcend and include what came before (Process and Reality, p. 238). In other words, physical processes do not just blindly repeat themselves until they disperse into nothingness. They grow. Insight accumulates. Intelligence blooms. For Whitehead, the cumulative character of evolutionary time means matter-energy must have been mind imbued from the beginning.

Brendan has achieved a masterful, far reaching synthesis of cutting-edge systems science that I hope will be widely read. But my heart tells me we’re going to need more than systems science to awaken people from the opium of materialist nihilism. It seems to me (pending further discussion!) that Brendan’s metamodern cosmology diverges rather significantly from at least my own understanding and interpretation of a process-relational approach. Which is fine! But I think his physical explanations of meaning still leave many metaphysical questions unanswered.

It is clear that Brendan’s aim is to overcome reductionism by affirming emergent complexity and open-ended learning. But his approach (which culminates in a “thermodynamic theory of the sacred” [p. 84]) ultimately amounts to no more than a kind of extended emergentist physicalism. Perhaps he will protest this label. I would be glad to hear it!

I have also done my best to take the research of natural scientists as seriously as possible. But I do so keeping in mind Whitehead’s “fallacy of misplaced concreteness.” And I’m not sure all the science agrees with Brendan’s take (nor that the scientists he cites agree with each other, as I note below). It takes a great deal of careful attention to our living experience to avoid substituting abstract models for what we are actually aware of in perception. Nature is not made of the bloodless equations printed in physics textbooks or programed into supercomputer simulations. “Real facts are happening.”

Whitehead (Modes of Thought, p. 144):

What are those primary types of things in terms of which the process of the universe is to be understood? Suppose we agree that nature discloses to the scientific scrutiny merely activities and process. What does this mean? These activities fade into each other. They arise and then pass away. What is being enacted? What is effected? It cannot be that these are merely the formulae of the multiplication table—in the words of a great philosopher merely a bloodless dance of categories [Whitehead means F. H. Bradley]. Nature is full-blooded. Real facts are happening. Physical nature as studied in science is to be looked upon as a complex of the more stable interrelations between the real facts of the real universe.

Science can expand and elucidate our experience, but too often it ends up trying to explain it away by reference to some supposed causal substrate hidden behind the scenes (whether branching infinitely in a wave-function multiverse or curled up into seven extra dimensions of vibrating strings). Whitehead protests against the bifurcation of nature into primary measurable or mathematizable stuff and accidental psychic additions, reminding us that the only truly physical world we’ve ever known or could know is the one we can feel, touch, taste, smell, hear, and see. The math must mean something. And while learning is the archetypal act of transcendence, we cannot pretend to transcend our own embodied perspectives. We do not learn our way out of this world, but can only learn our way into deeper, more intimate, more correlate embodiment of the intelligibility of the Universe. Whitehead:

It is the accepted doctrine in physical science that a living body is to be interpreted according to what is known of other sections of the physical universe. This is a sound axiom; but it is double-edged. For it carries with it the converse deduction that other sections of the universe are to be interpreted in accordance with what we know of the human body (PR, p. 119).

In other words, we as experiential creatures have direct access to an aspect of reality that even the most advanced instruments would fail to measure. Physicists measure the facts. Measurable material facts are the consequences of energetic activities. Getting at these activities means metaphysically generalizing beyond the physics of space-time and matter-energy. Both are emergent from an “ether of events,” or what Whitehead later calls “actual occasions of experience.” No fact is final. New actualities are happening everywhere all the time, exercising not only their physical power to feel the past but some modicum of mental power to divine the future. Mind goes all the way down, diverting the flow of energy (not that unlike Terence Deacon’s “absential” constraints) from entropic dispersion. Even the thermodynamicist needs to account for the origins of constraint itself, since without it energy could do no work and the laws of conservation and entropy become meaningless.

Whitehead again (Process and Reality, p. 238):

Physical science is the science investigating spatio-temporal and quan­titative characteristics of simple physical feelings. The actual entities of the actual world are bound together in a nexus of these feelings. Also in the creative advance, the nexus proper to an antecedent actual world is not destroyed. It is reproduced and added to, by the new bonds of feeling with the novel actualities which transcend it and include it. But these bonds have always their vector character. Accordingly the ultimate physical entities for physical science are always vectors indicating transference. In the world there is nothing static. But there is reproduction; and hence the permanence which is the result of order, and the cause of it. And yet there is always change; for time is cumulative as well as reproductive, and the cumulation of the many is not their reproduction as many.

[The account of] simple physical feelings lays the foundation of the treat­ment of cosmology in the philosophy of organism. It contains the discus­sion of the ultimate elements from which a more complete philosophical discussion of the physical world—that is to say, of nature—must be derived.

Whitehead here makes reference to “physical feelings.” His technical term for “feeling” is prehension, and he acknowledges both physical and conceptual forms of it. In other words, there is always both a physical and a mental pole involved in every concrete process at whatever scale it occurs. In the pre-biological world, the transmission of feelings is almost entirely “conformal,” meaning there is a high degree of repetition. This allows physicists to develop mathematical models with increasingly impressive predictive accuracy when applied in the idealized conditions they are better and better at building technological means of maintaining. In effect, contemporary techno-science is in the business of manufacturing artificial environments wherein nature behaves as its models predict. The tremendous success of this mechanistic approach to knowledge-power (measured in terms of its ability to transform the entire planet in only a few generations) has blinded us to its terrible cost. Nature is not just a bloodless dance of numbers, Earth not merely a machine. Real facts are happening. Gaia is a living being. And if modern techno-science has not yet come close to killing her, it has surely already left a deep scar.

If we can study contemporary physics with new eyes, perhaps we can come to intuit in the cloud of plasma that likely characterized the cosmos at earlier phases of its evolution that there is already some tendency for nature to be in tune, some degree of allurement toward creative possibilities not yet actualized. There is already a longing to evolve. Otherwise nothing as complex as a hydrogen atom, much less a star or a galaxy, ever could have emerged (to say nothing of biological cells and conscious psyches!).

As a panexperientialist (to use Whiteheadian philosopher David Ray Griffin’s favored term), Whitehead affirms that what physicists measure and calculate as “energy” is, concretely speaking, a creative activity imbued with some degree of experiential quality or unconscious subjectivity, even if of minuscule intensity relative to animal sentience or human sense-perception and self-consciousness. This follows from his understanding of what it means to seek to develop a truly metaphysical account. Metaphysics is the attempt to develop a general scheme of ideas that applies to everything we can experience. Note at this point that by making experience the criterion he is not doing metaphysics in the way post-Kantian or “postmetaphysical” approaches complain about (ie, he is not relying on something that is beyond experience in order to explain experience). Whatever the metaphysical categories we construct end up being, they should apply everywhere to every entity, regardless of scale and complexity. Brendan rejects this approach to metaphysics, arguing that life, mind, and culture (levels he borrows from

Gregg Henriques

’ UTOK) “require different metaphysical concepts above and beyond efficient causation” (p. 20). This is a fundamental divergence from process-relational ontology: Whitehead argued that making sense of even just the physics of cosmic evolution already requires more than efficient causation.

Metaphysics is just the search for categories generic enough to apply to everything that is or could be, whether physical, biological, psychological, or spiritual. So while feeling grows gradually more intense as it evolves, eventually becoming sentient and self-conscious, in a process-relational ontology, there is no entity (or actual occasion) that is entirely devoid of feeling. Whitehead is a speculative philosopher, but he is also a radical empiricist in the Jamesian sense. We simply cannot know anything that is beyond experience, and while self-conscious human experience may be very different in degree, it cannot be different in kind from everything else which exists. At least not if we accept Whitehead’s wager and join him in seeking a comprehensive metaphysical scheme of categories that remains rationally coherent and empirically adequate.

…

I share Brendan’s fascination with thermodynamics. There is much that remains mysterious to me about the motion of heat. But I’ve come to be reasonably sure there’s more to it than what a thermometer detects.

Conservation and entropy laws developed out of a study of closed systems. They presuppose and do not themselves explain the origin of constraints, much less constraint closure. Using thermodynamics as fundamental explanation leaves us with at least two unanswered questions: 1) why and how did the universe begin with such a low entropy state?; 2) where did the first constraints come from?

Brendan defines entities as “stable transmissions of information through time,” adding that “accurate transmission has intrinsic significance to that entity” (p. 8). The concept correlate to what Brendan is describing in Whitehead’s scheme would not be his “actual entities,” but what he called “societies.” A society is a historical route or genetic lineage of actual occasions sharing definite characteristics. The tables and chairs, plants and animals, planets and stars that we see around us are all examples of societies. Our own living bodies are societies. But to account for the fact that we not only feel but are conscious of our bodies, and self-conscious of our minds, Whitehead is not satisfied with an ontology made up solely of enduring societies. There are also acts of experience occurring, which are the only means by which facts can come to matter.

Brendan writes (p. 33):

“By being able to store meaningful information in the genetic code and transmit this information through reproduction, an entirely new level of complexity has taken hold in the universe.”

Genetic templating is an important advance, but we risk reification of nucleic acids as some kind of magical molecule if we make them the sole carriers of information. It makes just as much (or as little) sense to say that the information required for making an organism is stored in the intracellular matrix and in the extracellular environment and historical sequence of environments through which it has evolved. I worry that leaning on 1960s evolutionary science from reductive materialists like Richard Dawkins only signals that an abstract and decontextualized idea of genetic information is being mistakenly hypostatized (see also p. 29 of my chapter on cosmological evolution for more on this issue).

Biological life arises when not only the past but the future matters to a being. But where and when does life as such originate? While with their capacity for imaginative anticipation, biological organisms exhibit a truly impressive creative advance over the far more repetitive physical and chemical domains, the cosmological evidence suggests that energy has already been engaged in evolutionary learning for billions of years before they emerged. Learning doesn’t start with “genes” or with species. Metaphysical coherence requires that we affirm that life was latent in energetic activity already. Whitehead (The Function of Reason, p. 21):

The root principles of life are, in some lowly form, exemplified in all types of physical existence.

What emerges in the course of cosmic evolution are more intense (in terms of inner experience and outward complexity) forms of life, not life as such. In some sense, the primordial organism is the Universe as a whole (an incomplete, growing whole): call it the world-soul or cosmic organism. This cosmic life recapitulates itself in the lives of individual organisms. In Whitehead’s terms, it is the societies or enduring bodies that are undergoing emergent evolution through a series of levels. Actual occasions are sheltered by the social order thus achieved, their experience amplified into ever-higher grades of aesthetic satisfaction (for more on the Teilhard/Whitehead connection, check out this chapter).

I believe this distinction between actualities and societies is crucial, and I must thank Brendan for putting forward such a clearly written attempt at a physicalist explanation of life and mind that relies solely on the latter, that is, on the stable information transfer that allows for enduring bodies. This has allowed me to see the difference between physical and metaphysical explanations far more clearly than I had before. What Whitehead adds with his category of the actual occasion is an account of how all that information becomes concretely felt and interpreted, that is, how it is conformally prehended and, in some cases, conceptually altered, as it is transferred from occasion to occasion.

…

While I need to study their ideas more closely, what Brendan has presented of Kolchinsky and Wolpert’s work on the emergence of semantic information does not to my mind explain but rather presupposes organismic conatus.

Eric Chaisson’s idea of energy rate density shows that evolutionary selection processes precede biology. I also draw on his work in my chapter (with astrobiologist Bruce Damer) on the cosmological context of the origin of life.

I am curious to learn more about Seth Lloyd’s approach to quantum gravity and cosmic computation, but for now I am hoping Brendan can clarify for me whether his claim is that quantum computers could provide better simulations of space-time (which seems likely, since classical computers can already simulate it), or that space-time itself really is a playing out of computational logic gates (even if quantum)? Does this mean he (or Brendan) affirms pancomputationalism? Computational approaches stand in sharp conflict with other theorists Brendan cites, including John Vervaeke and Evan Thompson, who explicitly reject the idea that information-processing or algorithmic approaches can account for relevance realization and living organization, much less cognition and consciousness. Thompson and other enactivists also argue forcefully that the concepts of autopoiesis and enactive cognition are incompatible with Karl Friston’s “Free Energy Principle” approach. I can understand where those arguments are coming from, but my frequent interlocutor Tim Jackson thinks otherwise.

Another divergence I believe I detect in Brendan’s cosmology from Whitehead, but also Vervaeke, Thompson, and Michael Levin is that they all affirm that life has always already been more than a genetic system. Mind or cognition begins at least with single cells, not just with nervous systems (Vervaeke argues as much in this recent article with Jaeger, Riedl, Djedovic, and Walsh, which I discuss with biologist Tim Jackson here).

…

Brendan writes (p. 39):

“Before, at the level of Life, learning occurred only at the population (phylogenetic) level, as natural selection ‘updated’ the species’ genetic code to better match environmental conditions.”

It doesn’t seem like Brendan’s UTOK-derived levels of matter, life, and mind adequately accounts for the fact that protists are already minded agents capable of memory and learning beyond what natural selection has imprinted on their genetic inheritance, and bacterial colonies were engaging in complex communication networks via bioelectricity long before nervous systems evolved in more complex animals. For these reasons and others, I’m more convinced by Thompson et al’s mind-life continuity thesis. The “gene pool” or species is not the right or at least the only level at which to capture what is characteristic of biological life. We not only need to affirm multilevel selection; we need to understand life as at least a planet-scale phenomenon, and also as a process with roots in a Cosmic Life.

Brendan writes that “natural selection hit on the bilateral symmetry of the animal body plan,” but there’s plenty of evidence that symmetry is no happy accident achieved by gradual mutations but is rooted far deeper in physics. Many systems biologists (eg, Brian Goodwin) would argue that bilaterality is a result of emergent “order for free” that does not need to wait for natural selection to stumble upon it. Rather, like the spherical shape of liposomes, bilaterality is an emergent function of mathematical possibility spaces and the formal constraints operative in morphogenesis.

Brendan quotes Ginsburg and Jablonka, who argue that “Learning was the basis on which [our conscious experience] was constructed” (p. 45). But I would reverse this: experience is the basis on which learning can occur. From a process-relational perspective, cognitive learning presupposes subjectivity. The universe evolves ever-richer meanings because it is experiential through and through.

Are animals really solipsists until language allows them to share thoughts (p. 62)? I don’t think this adequately accounts for the emotional resonance and common fields of feeling that are basic to the biological and even physical world. If anything, the illusion of solipsism is a function of being able to linguistically reflect to oneself about the cognitive states of others, concocting a “theory of mind” which overlays the basal emotional resonances or non-symbolic “mind meld” that otherwise dominates the experiential inter-being that is undeniably basic to biology, and arguably basal in physics, too. Even in human beings, our symbolic mouth squeaks and finger squiggles only function as mediums of communication because of a pre-verbal transcendental intersubjectivity and embodied sense of contextualized meaning.

…

Brendan affirms (and I agree!) that (p. 37):

“Meaning and value exist, not as mere social constructs or subjective inventions, but as real aspects of the world. They do not, we must stress, inhere in things themselves, as essences bestowed (such as traditional theological thinking presupposed); but neither are they purely subjective experiences overlaid onto the world (such as modern reductionistic thinking has tended to assume).”

No disagreement here. In short, meaning is relational, not essential. “Real relationship” is akin to Whitehead’s concept of “prehension,” which is the means by which feelings are transmitted at whatever scale energy can be measured and beyond (“beyond” since in addition to the objective data of physical science he also grants reality to subjective forms or emotions, subjective aims or purposes, and conceptual prehensions of possibilities as yet unactualized anywhere in the measurable universe).

This is why Whitehead’s God is an all-caring Relator and not an all-controlling Creator. The divine aim as the base of all things cannot predict the self-differentiating aims it inspires. God and the world recreate in one another. In a Teilhardian sense, evolution is not only a process of learning, it is incarnation, a process of divine embodiment. As Whitehead puts it (Religion in the Making, p. 155-156):

Every event on its finer side introduces God into the world. Through it his ideal vision is given a base in actual fact to which He provides the ideal consequent, as a factor saving the world from the self-destruction of evil. The power by which God sustains the world is the power of himself as the ideal. He adds himself to the actual ground from which every creative act takes its rise. The world lives by its incarnation of God in itself.

…

See also