What This Is, and Where It Stops
This is the Amovera architecture reduced to what can be defended without hedging. Earlier passes kept any claim that was “defensible”; this one keeps only claims that are established — settled physics, robust empirical patterns, well-replicated cognitive science — and discards the rest, however foundational it once seemed.
What that discipline removes is significant, and worth stating up front. It removes the free energy principle (contested), the electromagnetic theory of consciousness (a minority view), any claim that consciousness requires integration (the hard problem is unsolved), and — most consequentially — the bridge from is to ought. The neo-Aristotelian grounding of the last pass was the best available naturalistic move, but “best available” is not “proven,” and the instruction here is to keep only the proven.
So the result is not a moral theorem. It is a description of reality as it stands: an account of what self-maintaining, knowledge-bearing systems are and what conditions sustain or destroy their capacity. It states, precisely, the point at which a value would have to be added to make it an ethics — and it does not add one, because science and philosophy cannot yet supply it. The architecture’s ethical conclusions are not refuted here; they are set aside as not-yet-grounded, recoverable only by an explicit value premise that a reader must choose to adopt.
Eight propositions follow. Each is stated plainly and tagged with its grounding. They form a chain: each rests on the ones before it. The chain ends, deliberately, one step short of a prescription.
The Theorem
I. Persistence is acquisition.
An ordered system held far from thermodynamic equilibrium persists only by continuously acquiring energy and structure from outside itself. Withdraw the throughput and it decays toward equilibrium. Equilibrium is the one state that persists without acquiring — and for an ordered, self-maintaining system, reaching it is dissolution, not rest.
Grounding. Settled. Non-equilibrium thermodynamics: Schrödinger’s “negative entropy,” Prigogine’s dissipative structures, standard stability theory. This is the secure floor of the whole theorem and assumes nothing contested.
II. Some such systems carry a model of themselves.
A subset of these systems — living things, and the minds some of them possess — do more than dissipate energy: they store and transmit information about what keeps them going. Persistence becomes self-directed and heritable. This is the difference between a hurricane and a cell.
Grounding. Settled core. Self-maintenance plus heritable information is the uncontroversial common ground across definitions of life. The exact boundary criterion (autopoiesis, autocatalysis, information-theoretic) is unsettled and is not relied on.
III. A modelling system can mistake coherence for truth.
Minds build predictive models and register a felt sense of “understanding” when a model coheres. That signal tracks internal coherence and fluency, not accuracy. A model can therefore feel like knowledge while being wrong — and because the feeling is satisfying, it resists correction.
Grounding. Strong / well-replicated. The fluency heuristic (Alter & Oppenheimer); Bjork’s “illusions of competence”; predictive-processing accounts on which the sense of understanding tracks prediction-error reduction, a coherence signal, not correspondence. Note: this is a claim about cognition and model-building — it does NOT assume any theory of subjective experience, which remains unsolved (the “hard problem”).
IV. Human persistence runs through accumulated culture.
Humans persist as a species largely through a second, non-genetic inheritance channel: knowledge accumulated and transmitted across generations, with technology as its material arm. The two coevolve and have reshaped human biology. This cultural channel, not individual capability, is the distinctive human strategy.
Grounding. Strong / mainstream. Gene-culture coevolution and dual-inheritance theory (Boyd & Richerson), the cultural ratchet (Tomasello), niche construction (Odling-Smee, Laland), the cultural brain (Henrich).
V. Cumulative knowledge requires connection and fidelity, not merely independence.
A population accumulates capacity only when enough minds remain connected and transmit with sufficient fidelity, alongside a supply of variation. Sever the connections or degrade the transmission and a population can lose capacity it once held — not merely fail to gain. Pure independence does not accumulate; it sheds.
Grounding. Strong pattern. A convergent literature links population size, connectedness, and transmission fidelity to sustained cultural complexity (Powell, Shennan & Thomas; Derex & Mesoudi; Kline & Boyd; Muthukrishna et al.). The most cited single case (Holocene Tasmania) is disputed and is not relied on; the pattern across many studies and experiments is.
VI. Such systems can fail in two opposite directions.
A knowledge-bearing population can lose capacity by over-consolidation — agency and decision concentrating in a few, the rest deskilled — or by over-fragmentation — connection and transmission breaking down below the threshold needed to retain what is known. Both states are self-reinforcing. Which one threatens at a given time is an empirical question, not a constant.
Grounding. Strong tendency. The consolidation pole maps onto extractive institutions (Acemoglu, Johnson & Robinson, 2024 Nobel: concentration stifles innovation; distribution sustains it) and elite-capture dynamics (Michels, Piketty). The fragmentation pole follows from Proposition V. The two-pole structure is the explore–exploit trade-off general to adaptive systems. Edge cases exist (growth under non-inclusive regimes), so this is a strong tendency, not an exceptionless law.
VII. In an interdependent population, capacities are coupled.
Because individuals depend on the shared cultural channel (IV–V), no member’s capacity to perceive, learn, and act is fully independent of the others’. Conditions that raise or lower the distributed capacity of the population tend to raise or lower the capacity available to its members. Interdependence is structural, not sentimental.
Grounding. Strong. Follows from IV–V; consistent with the evolution-of-cooperation literature, in which conditional, reciprocal cooperation — not unconditional sacrifice and not pure defection — is what stabilizes in interdependent populations (Axelrod; reciprocity and multilevel-selection models).
VIII. Here the description ends; an ethics would require a premise science cannot supply.
Propositions I–VII describe what these systems are and what sustains or destroys their distributed capacity. They do not, and cannot by themselves, establish that anyone ought to sustain it. To move from “this sustains distributed capacity” to “one ought to act so” requires an added value premise — that the flourishing of this kind of system is choiceworthy. That premise is not a finding of physics, biology, or cognitive science. It must be chosen, and naming it as a choice is the honest stopping point.
Grounding. This is the is–ought boundary (Hume; Moore’s open-question argument; the standard critique of evolutionary ethics: that something persists or sustains a system says nothing, by itself, about whether it ought to). The strongest naturalistic bridges — neo-Aristotelian natural goodness (Foot, Thompson) — are defensible but contested, not established, and so fall outside what this reduced theorem will assert.
One Conditional, Clearly Marked
The theorem above asserts no ought. But the architecture’s ethical content can be stated honestly as a conditional — true as logic, carrying its premise in the open, asserting nothing about whether the premise should be adopted:
IF one takes the flourishing of a knowledge-bearing population — its sustained, distributed capacity to perceive, learn, and act — as a thing to be valued, THEN Propositions V–VII specify what that commitment requires: protect connection and transmission against fragmentation; protect distributed agency against consolidation; read which danger is live in a given domain rather than assuming one; and extend a conditional, clear-eyed regard across the population, since capacities are coupled.
This is the entirety of the architecture’s ethics that the evidence licenses — and it licenses it only as the consequent of an “if.” The “if” is where a human being, not a theorem, must decide. The reduced theorem’s final and most important claim is simply that this is where the decision lives, and that no current science or philosophy can make it for us.
What Was Removed, and Why
So the cut is legible, here is what the reduction discarded from the prior version and the reason each failed the “proven” bar:
- The free energy principle as a foundation. A contested, arguably unfalsifiable framework; its core lemma is challenged in the literature. Removed entirely. The Master Principle now rests on plain thermodynamics, which needs no such support.
- Any physical theory of consciousness (electromagnetic-field identity; the claim that integration is necessary and sufficient). Consciousness science has no consensus and the hard problem of subjective experience is unsolved. The theorem keeps only the uncontroversial functional point — that minds build models and can mistake coherence for accuracy (III) — and says nothing about experience or substrate.
- The substrate-generality of mind (the synthetic-consciousness corollary). It depended on a settled theory of consciousness that does not exist. It is neither asserted nor denied here; it is simply open.
- The derivation of an ought. Every attempt — hypothetical imperative, natural goodness — either smuggled a value or rested on contested philosophy. Demoted to the explicit conditional above. The destination is preserved as an honest “if,” not asserted as a finding.
- Single decisive cases (e.g., Tasmania). Replaced by the convergent multi-study patterns they were standing in for.
On What Remains
What is left is smaller, and it is sturdier for being smaller. It claims no more than the evidence carries: that self-maintaining systems persist by acquiring; that some carry models of themselves; that those models can deceive themselves with coherence; that human persistence is cultural and depends on connection and fidelity; that such systems fail by consolidation or by fragmentation; that capacities in such a population are coupled; and that the step to an ethics requires a value no science supplies.
This is a description of reality as it currently stands, terminating exactly where current knowledge does. It does not reach a moral conclusion, and it does not pretend to. That restraint is not a failure of the theorem — it is the most defensible thing it can say.
An ordered system far from equilibrium persists only by acquiring what it lacks. What follows from that for how one ought to live is a question this theorem leaves — honestly — open.