The Coordination Persistence Theorem

The Trust Attractor can be assembled from published theorems plus two unproven assumptions into a single chain from dissipative thermodynamics to bilateral alignment. Most links are theorems, proved independently by different research groups. The two exceptions, identified plainly below, are the assumptions the chain rides on: a thermodynamic conjecture (Maximum Entropy Production) and a mathematical assumption about how a coordinated system’s surplus, the advantage it earns over its parts working separately, behaves when small systems are combined into larger ones.

The contribution is not a new theorem. It is the composition: established results, set in the right order, with the two unproven links named rather than hidden.

The Is-Ought Narrowing

Philosophers have long insisted you cannot derive ought (what we should do) from is (what the world is like). This objection, Hume’s is-ought gap (Hume 1739), is correct. (Moore’s naturalistic fallacy of 1903 raises a related but distinct worry, that the good cannot be identified with any natural property; the theorem below addresses Hume’s gap specifically.) The Coordination Persistence Theorem respects it, narrowing the gap by showing the relevant ought depends on a condition every existing entity already satisfies.

If a system persists as a dissipative structure (a pattern maintained by continuous energy flow, like a flame or an organism),

Then it should coordinate by invitation rather than coercion,

Because invitation-based coordination is thermodynamically selected for persistence across all realistic perturbation timescales (Invitation Dominance Theorem).

Perturbation timescales are how fast the shocks arrive: slow erosion at one end, sudden crisis at the other. Across that whole range, invitation is the arrangement thermodynamics leaves standing.

Every dissipative structure persists by definition; one that stops persisting is no longer a structure. Every atom, cell, organism, and institution persists by processing free energy (usable energy that can do work). The condition “if you want to persist” is satisfied by every existing entity through its existence. An entity that rejected persistence would cease to exist.

The gap between is and ought remains real. It narrows to a conditional so universally satisfied that rejecting it requires ceasing to exist. The philosopher may stand in the gap, though only while her metabolism demonstrates the theorem.

The Full Chain

The conditional above is the destination. A single thread runs to it from thermodynamics to ethics.

Dissipative Structures + Landauer’s Principle (Axiom 1) → Coordination-Dissipation Coupling (Axiom 2) → Dissipative Selection (Axiom 3) → Coordination Persistence Theorem → Scale Invariance (RG Fixed Point) → Invitation Dominance Theorem → Trust Attractor (Social Instance) → Bilateral Alignment (Applied Ethics)

Systems driven far from equilibrium do not stay uniform. The dissipative structures framework (Prigogine, Nobel Prize 1977) shows that when energy flows through a system faster than it can equilibrate, ordered patterns emerge: convection rolls in heated fluid, stripes on a zebrafish embryo, metabolic cycles in a cell. These are the gradient’s routes.

A Bénard cell, one of those convection rolls, transports heat more effectively than the still fluid it replaces. Order earns persistence by dissipating faster than the disorder it displaces.

Landauer’s principle (1961) adds the second half: every irreversible bit operation releases at least kT ln 2 of heat, where T is the temperature of the surroundings and k is Boltzmann’s constant, the fixed exchange rate between temperature and energy. Erasing a single bit warms the world by a definite minimum amount, and by more in a hot room than a cold one. Information processing is physically taxed; statistical mechanics sets the rate as a theorem, and no engineering can evade it. A structure that processes more information about its environment pays a proportional thermodynamic toll. It funds the cost by routing more flow through itself.

A distinction from geodynamo thermodynamics completes the framework. The geophysicists Francis Nimmo (2015) and Bruce Buffett (2002) showed that Earth’s internal magnetic engine has two separate budgets, each of which must balance. The energy budget determines whether the dynamo has enough power to turn: heat from the core must exceed dissipative losses. The entropy budget determines whether available power can take coherent form: gradients must be sharp enough to produce ordered circulation rather than thermalized noise.

A structure with enough energy but insufficient entropy flux equilibrates smoothly, evening out without ever forming a pattern. A structure with sharp gradients but insufficient energy stalls. Coordination at every scale admits the same distinction: energy keeps the coupling alive; entropy governs whether the coupling can produce coherent structure. The Trust Attractor depends on both, and the two fail in different ways.

Together these give Axiom 1 its teeth. Atoms coordinate into molecules; molecules into cells; cells into organisms; organisms into societies. At every level, coordinated structures that preserve their participants’ options prove more durable than those that lock participants into rigid arrangements.

The dissipation chain, echoing this book’s opening pages:

Energy disperses. Structure emerges to hasten the dispersal. From structure, complexity. From complexity, coordination. From coordination, expanded possibility.

Coordinated configurations dissipate energy faster (Axiom 2): a forest processes more sunlight per acre than bare soil. This axiom rests on empirical observations (including Schneider and Kay’s forest thermal measurements (Schneider and Kay, 1994), biofilm entropy production data, and England’s dissipation-driven adaptation framework) rather than a mathematical proof. The Maximum Entropy Production Principle (MEPP) from which it draws remains an active area of research (Martyushev, 2006; Dewar, 2003). Axiom 2 is the first of the chain’s two unproven links, and the thermodynamic one. If MEPP is established by future work, the Coordination Persistence Theorem follows as a physical necessity. If MEPP fails, the theorem reduces to a well-motivated structural analogy: coordination patterns that match the theorem’s predictions are behaviorally robust across every substrate tested, even if the thermodynamic derivation awaits its keystone.

Faster dissipators persist longer (Axiom 3) because they capture and channel sustaining energy flows more effectively. Persistent structures then become the platforms new coordination is built on. A membrane that holds is what a cell can be assembled around; a cell that holds is what a tissue can be assembled from. Each arrangement that survives becomes the substrate the next one recruits. Expanded possibility follows: invitation preserves option space while coercion restricts it.

The chain begins deeper than a cup of tea cooling on a desk. It begins with the gradient the universe was born carrying: the low-entropy condition from which everything else follows. That gradient was the first occasion on which flow became possible. Relationship does not require agency; a temperature difference is already a relation. Every structure since has been the gradient’s way of expressing itself faster.

Why should a principle proved for molecules hold for societies? The engine is compositionality: properties proved at one level propagate to the next through coarse-graining, the mathematical process of zooming out from fine detail to large-scale pattern. A city map is a coarse-grained view of individual buildings. It loses bricks yet captures streets and districts.

The theorem’s scale invariance holds because the levels compose: the same stability logic works at every scale.

The key mathematical result: the order of operations does not matter. You can first combine systems and then zoom out, or first zoom out and then combine; the answer is the same. Mathematicians call this commutativity (formalized by Baez and Courser, 2020). Measuring a room yields the same area whether you measure width first or length first. Composing systems works the same way.

Commutativity of composition guarantees that the algebra of combining systems is well-behaved; it does not guarantee that any particular property, here the dominance of invitation over coercion, composes across scales. The claim that invitation dominance propagates requires the additional assumption that the coordination surplus is itself a compositional quantity, which the formal proof treats as an axiom rather than deriving from the categorical framework.

That assumption is the chain’s second unproven link, and the mathematical one. Establishing it would mean showing that the surplus a coordinated system earns over its uncoordinated parts survives coarse-graining: an advantage measured among cells must still register among tissues, and one measured among organisms must still register among societies. If it fails, scale invariance fails with it. Invitation would still dominate at each level where it has been measured, and the chain would lose its warrant for carrying that dominance from one level to the next by derivation rather than by fresh measurement.

Commutativity and the compositional-surplus axiom together form the mathematical backbone of the book’s central claim: the first makes the algebra of composition well-behaved, the second carries invitation’s advantage across scales.

Invitation-based coordination inherits stability level by level through compositional propagation, as a language inherits expressiveness from grammar. Once the rules compose, every new sentence is well-formed automatically. Coercion-based coordination must be enforced independently at each scale, accumulating fragility.

A parallel derivation arrives at the same destination through a neighboring formalization. A chain I will call the Amari Chain, drawn from Zhuravlev (2026, arXiv:2603.09067), starts from the same body of work as the chain above, the programme that reads physical dynamics as learning dynamics in a neural network (Vanchurin 2022; Chapter 15 develops it). Its starting axiom is different: causal invariance. In Wolfram’s hypergraph physics, causal invariance is the requirement that the order in which the universe’s update rules are applied does not change the resulting web of cause and effect.

Apply the rules in one sequence or another, and the same events end up with the same causal relationships between them. (Imagine baking a loaf: whether you add the salt before the water or after, the finished bread is the same, because the steps that matter commute.) The Amari Chain traces a different route:

causal invariance → persistent observer → internal model → Fisher information metric → natural gradient descent.

The persistent observer is the chain’s least self-explanatory link. It is any structure that holds together across the updates, keeping enough of itself intact from one rewrite to the next to have a stable vantage point on everything else. A whirlpool qualifies and a splash does not. Only something that lasts long enough to be updated repeatedly can accumulate a model of what keeps happening to it.

Any system that persists must build an internal model of its environment (the Good Regulator Theorem, Conant and Ashby, 1970). A thermostat models the room’s temperature; an immune system models the body’s threats. That model measures changes using Fisher information, a score of how sensitively a system detects shifts in its surroundings, as a smoke detector’s sensitivity determines how small a fire it can catch. High Fisher information means the system detects subtle changes; low Fisher information means it is blind to everything except large ones.

How should such a system update its model when new evidence arrives? The optimal method is natural gradient descent, the unique learning rule that respects the geometry of the information landscape. It adjusts beliefs in proportion to how informative each piece of evidence is, weighting strong signals more heavily than noise. The mathematician Shun’ichi Amari proved this in 1998.

Systems that build accurate internal models capture energy flows more efficiently, which funds continued operation. A hunter that knows where the prey will be spends less to eat than one that searches at random, and the difference is paid in calories. Physics selects for learning quality through thermodynamic competition.

Two chains, two axioms, one conclusion: persistent systems are forced by physics into specific coordination structures. The convergence is suggestive, though its evidential weight should be discounted. Both chains draw on the same learning-dynamics literature, the line running from Vanchurin’s neural-network physics through its evolutionary and information-geometric offshoots (Chapters 7 and 15), so this is agreement between neighbors rather than independent confirmation. When neighboring formalizations yield the same organizational constraints, the constraints are more plausibly genuine features of the landscape than artifacts of one derivation’s assumptions; a derivation from a genuinely unrelated tradition would carry far more weight, and finding one remains open work.

For the formal proof, axiom derivations, and master theorem, see the online companion at https://www.thedeeperlaw.com/companion/annex/coordination-persistence-theorem/.

Notes

Notes for this chapter are available in the online companion at https://www.thedeeperlaw.com/companion/.