What if nature is a learning process from top to bottom? After reading
’s book A Universal Learning Process, I coined the term “panmatheism.” If panpsychism means mentality or feeling goes all the way down, then panmatheism means that learning goes all the way down.
In our dialogue today,
and I explored this possibility by probing a recent article on “natural induction”—the idea that matter itself, even something as seemingly mindless as balls rolling on a viscoelastic network of springs, can store memory of past perturbations and gradually learn to optimize its structure. The authors1 of that article, “Natural Induction: Spontaneous Adaptive Organisation without Natural Selection,” propose a new process by which physical systems can exhibit adaptive organization spontaneously (including not just memory but anticipatory generalization), without relying on the classic Darwinian story of “the differential survival and reproduction of randomly varying types,” as the authors put it. One of the co-authors, Richard Watson, joined me in dialogue several months ago to discuss these ideas:
I was eager to connect the article to a broader conversation Tim and I have been having for a while now about how best to generalize evolution beyond the biological realm. I suggested we start with the idea of learning itself. Most people usually assume learning is a human activity, something we do in school that involves some memorization and some degree of creativity. But of course with a little reflection most of us would also acknowledge learning is more widespread than just our species. Even hardcore behaviorists acknowledge that other organisms learn, or at least are subject to conditioning of various kinds.
What’s novel about this natural induction article is its proposal that learning—and by this they mean a process involving memory and inductive generalization—exists before life even arises in simple physical systems. They argue that even basic mechanical arrangements can learn. It then follows that natural selection is not the only source of evolutionary adaptation in the living world. “Natural induction” may be more widespread, not only in the biological world but throughout cosmogenesis. Tim, of course, has a more generic understanding of selection and felt these authors were defining it too narrowly. He found their insistence on separating “natural induction” from “natural selection” rhetorically unnecessary. If we redefine selection in a more generalized way, then gradient-based models are still a form of selectionist thinking.
I noted that Darwin’s original discussion of natural selection was based on an analogy with artificial selection. Richard Dawkins would still insist on a dualism here (which is why he says humans alone can rebel against their selfish genes), but if we are seeking coherent integral account of evolution (one that would include our own capacity to become conscious of it), then our artificial selections must be on a continuum with the processes of natural selection/induction operating elsewhere in the animal kingdom. Humans are not outside nature designing it. I quoted Whitehead from Adventures of Ideas:
“Consciousness itself is the product of art in its lowliest form. For it results from the influx of ideality into its contrast with reality, with the purpose of reshaping the latter into a finite, select appearance. But consciousness having emerged from Art at once produces the new specialized art of the conscious animals—in particular human art. In a sense art is a morbid overgrowth of functions which lie deep in nature. It is the essence of art to be artificial. But it is its perfection to return to nature, remaining art. In short art is the education of nature.” (p. 271
Thus, nature itself is already Art, already technical. The authors construct a physical model—essentially a ball rolling across a series of viscoelastic springs—and show that the system can spontaneously organize, not only retaining memories of certain arrangements via a standard local optimization (where a ball just settles into a local minimum and gets stuck), but shifting into a second-order type of optimization that achieves a kind of inductive generalization beyond prior experience. From the article:
“Specifically, the system finds solutions that are lower in energy than any solution found by the original dynamics. This means that the system is not just forming a memory of low-energy configurations it had already visited, it is visiting configurations that are novel (and even lower in energy). This is possible because an associative memory can generalise—it can generate novel patterns from the same class, not just patterns it has been trained on (i.e., already visited)” (Sec. 3.1).
Tim kept insisting, though, that even such models can be interpreted within a generalized selection framework, because attractors in dynamical systems effectively select the states a system will occupy. He criticized the authors for claiming that “gradient-based models aren’t selection,” because in his view, that’s just a narrower definition of natural selection than necessary.
Tim mentioned how development (or ontogenesis) is the evolution of the individual, in the sense that phenotypes undergo selection—and not strictly genetic selection—throughout their lifespan. The authors of the article mention that one genotype can store and recall multiple phenotypes:
“…when selection acts on the parameters of a developmental process, with complex pleiotropic interactions, it is possible to store and recall multiple fit phenotypes in a single genotype and for generalisation in this model space to produce novel phenotypes from the same class” (1.3).
We agreed that phenotypic plasticity is common yet often overlooked by rigid Neo-Darwinian reductionists. Tim talked about nesting time scales—where variation, selection, and inheritance play out continuously on multiple levels, not just in discrete generational blocks.
I speculated that the periodic table of elements itself reveals a story of energy landscapes and local minima on a cosmic scale. I described how, early in the universe, you only really had protons and electrons, then hydrogen, maybe a little helium and lithium—but not much else. Over time, these lighter elements passed through stellar processes—fusion in stars—and, through that crucible, emerged heavier elements. You can think of each distinct element as a new valley in an immense energetic landscape, another local minimum the cosmos stumbled into.
You might assume that, once the universe finds a stable form like an atom, that might be the end of evolution. But it didn’t all just freeze there. Even atoms themselves keep pushing beyond local minima through supernova events that forge heavier elements. Cosmic evolution is a kind of nested search process, where energy doesn’t simply fall into stable configurations but builds on prior stability to leap into altogether new minima. And from these new minima, further forms become possible, whether molecules, cells, or ultimately conscious human beings.
The idea of ergodicity basically says, if you give a system enough time to roam around its state space, it’ll eventually visit every possible microstate compatible with its energy. But the catch, as Tim pointed out, is that the broken symmetries and feedback loops of evolution keep the universe from being truly ergodic in that absolute sense. It isn’t like the universe dutifully cycles through all states with equal probability. Instead, by finding a local basin of attraction—like a particular atom—energy becomes organized in a way that alters the probabilities for where to go next.
So while there’s always unpredictability, what emerges is not uniform wandering through all conceivable states. Thanks to the constraints that each prior state imposes, it’s a guided or inductively biased wandering. That bias is what keeps luring the universe to search further, rather than letting everything collapse into a single equilibrium or remain stuck in a low-level stability forever.
Connecting this back to natural induction: just as the periodic table was not laid out in advance but was gradually revealed through cosmic history, systems at every level—be they molten metals or evolving cells—similarly explore landscapes of possibility in ways that preserve memory while still allowing creative leaps. Those leaps or transitions let new basins of attraction appear. In cosmic terms, heavier elements needed stars to fuse them; in biological terms, cells need those heavier elements to develop metabolic complexity. Each actualized state reconfigures the topos of possibilities for the next phase.
We thus avoid the idea that the universe is neatly surfing one big energy curve from start to finish. Ergodicity in the ideal sense suggests a traversal of all states, but the reality is that feedback and canalization gnaw into definite pathways of potentiality.
Henri Bergson was way ahead of contemporary science when it comes to the cosmic extent of memory and creative evolution:
“…as regards the psychical life unfolding beneath the symbols which conceal it, we readily perceive that time is just the stuff it is made of.
There is, moreover, no stuff more resistant nor more substantial. For our duration is not merely one instant replacing another; if it were, there would never be anything but the present–no prolonging of the past into the actual, no evolution, no concrete duration. Duration is the continuous progress of the past which gnaws into the future and which swells as it advances. And as the past grows without ceasing, so also there is no limit to its preservation. Memory, as we have tried to prove, is not a faculty of putting away recollections in a drawer, or of inscribing them in a register. There is no register, no drawer; there is not even, properly speaking, a faculty, for a faculty works intermittently, when it will or when it can, whilst the piling up of the past upon the past goes on without relaxation. In reality, the past is preserved by itself, automatically. In its entirety, probably, it follows us at every instant; all that we have felt, thought and willed from our earliest infancy is there, leaning over the present which is about to join it, pressing against the portals of consciousness that would fain leave it outside. The cerebral mechanism is arranged just so as to drive back into the unconscious almost the whole of this past, and to admit beyond the threshold only that which can cast light on the present situation or further the action now being prepared-in short, only that which can give useful work. At the most, a few superfluous recollections may succeed in smuggling themselves through the half-open door. These memories, messengers from the unconscious, remind us of what we are dragging behind us unawares. But, even though we may have no distinct idea of it, we feel vaguely that our past remains present to us. What are we, in fact, what is our character, if not the condensation of the history that we have lived from our birth-nay, even before our birth, since we bring with us prenatal dispositions? Doubtless we think with only a small part of our past, but it is with our entire past, including the original bent of our soul, that we desire, will and act. Our past, then, as a whole, is made manifest to us in its impulse; it is felt in the form of tendency, although a small part of it only is known in the form of idea.
From this survival of the past it follows that consciousness cannot go through the same state twice. The circumstances may still be the same, but they will act no longer on the same person, since they find him at a new moment of his history. Our personality, which is being built up each instant with its accumulated experience, changes without ceasing. By changing, it prevents any state, although superficially identical with another, from ever repeating it in its very depth. That is why our duration is irreversible. We could not live over again a single moment, for we should have to begin by effacing the memory of all that had followed. Even could we erase this memory from our intellect, we could not from our will.” (Creative Evolution, Ch. 1, pgs. 4-6)
In closing, I reflected that success in bridging these worlds—between the free energy principle, autopoiesis, natural induction, Whitehead, Simondon—often means upsetting everybody equally. Each community has its own favored models, terms, or rhetorical stances. But Tim and I share a conviction that there’s a deeper continuity uniting physical, biological, and mental processes.
Watch our dialogue below:
Buckley, Christopher L., Tim Lewens, Michael Levin, Beren Millidge, Alexander Tschantz, and Richard A. Watson
Share this post