Chapter 4: The Coming Together of Things

Why does anything exist at all?

If entropy (thermodynamic entropy, the dispersal of energy toward equilibrium, as defined in Chapter 1) always increases, if energy always spreads, the universe should be a featureless haze of particles drifting apart. Yet here we are: galaxies, cells, cities, minds. Atoms clumped into molecules. Molecules into cells. Cells into organisms. Organisms into societies. Everywhere you look: gathering. Coordination.

If the rule is spreading, why is there so much coming together?

Part I: The Thermodynamic Foundation

Structure Serves Entropy

A river erodes a channel. The channel lets water flow faster than it would across flat ground. Structure accelerates the very process that created it. The pattern is general: things come together because coming together makes energy spread faster.

This is the Maximum Entropy Production Principle (MEPP): systems tend toward configurations that dissipate energy as quickly as possible.⁷² The Second Law says entropy must increase; the MEPP proposes that it increases as fast as the available structures allow. It remains an active area of research rather than settled law.

The weaker claim, that dissipative structures exist and some persist longer than others, is uncontested. The stronger claim, that nature selects for maximal dissipation, is the one under debate.

This book’s argument works at both levels: the weaker version grounds all the structural observations, while the stronger version, if it holds, explains why those structures recur so reliably.

Structure exists because of entropy. Organization is the mechanism entropy uses to accelerate.

The distinction that matters is between passive dissipation and self-maintaining dissipation. Water flowing downhill dissipates a gravitational gradient and stops at the bottom. A living cell dissipates chemical gradients to maintain the structure that dissipates chemical gradients. The water reaches equilibrium. The cell keeps going.

Passive flow exhausts its gradient and halts. Self-maintaining flow uses its gradient to sustain the architecture that captures more gradient. A catalyst speeds a reaction up; a reaction is autocatalytic when what it produces is its own catalyst, so each run makes the next run easier. This difference, autocatalytic persistence, separates a river’s channel from the organism swimming upstream through it.

The answer to this chapter’s opening question emerges: structure is entropy’s accelerant, and what spreads energy faster persists longer.

Figure 4.1: Left: a symmetric container, uniform throughout. No gradient, so no flow, and no function; equilibrium is static. Right: the same container with a dense region and a sparse one separated by a membrane. The difference drives flow across the boundary: gradient, flow, function, life. Below the panels, four familiar gradient pairs and what each one drives. Hot and cold drive heat flow; concentrated and dilute drive diffusion; charged and neutral drive current; informed and ignorant drive communication.

The autocatalytic loop runs at cosmological scale. A massive star lives a few million years, then detonates, seeding its surroundings with heavy elements: carbon, oxygen, silicon. These heavier elements radiate heat more efficiently than primordial hydrogen, opening new cooling channels that let gas collapse into stars faster. The next generation forms sooner, with richer chemistry, and dies enriching further. Each cycle produces the raw materials that accelerate the next cycle: stellar nucleosynthesis feeding back into star formation, the same autocatalytic structure that separates self-maintaining dissipation from passive flow.

Cosmological models predicted this loop would need hundreds of millions of years to produce significant chemical enrichment. Recent observations from the James Webb Space Telescope suggest otherwise. A compact, dust-rich galaxy at a spectroscopic redshift of 11.5, when the universe was roughly 400 million years old, already contains carbon emission lines and substantial dust attenuation, signatures of rapid chemical and dust enrichment from early star formation.⁷³ The enrichment cycle was running within the first few hundred million years, faster than standard models predicted.

Scientists compared the finding to discovering a fully grown tree in a field planted weeks ago: a metaphor that inadvertently reveals the assumption it should challenge. The tree grew that fast because the soil was richer than the models imagined. Autocatalytic persistence does not wait for permission. Given sufficient density and tight feedback, the loop runs as fast as physics allows.

The principle extends to how artificial systems learn. Token Superposition Training, the coarse-first pretraining schedule described in Chapter 3, gives a language model a blurred rendering of its training data before the full-resolution version, and reaches equivalent performance in roughly half the wall-clock time.⁷⁴ The blurred signal is generative the way a morphogen gradient (the chemical concentration signal that tells embryonic cells their rough position in the body) is generative: it carries the rough structure from which detail later resolves. Its limit is structural novelty. The coarse phase absorbs whatever resembles material it has already seen; the distributional signature of a genuinely different domain can enter only during the fine phase. Coarse processing suffices for familiar structure; fine processing is necessary for novel structure.

The hypothesis, still under active investigation, is that coordination is one of the mechanisms by which dissipation increases: coordinated structures channel energy flows through more pathways than uncoordinated ones. The river has already shown the elementary version: cut a channel and the same water leaves faster than it did across flat ground. What the channel does with one gradient, a coordinated system does with many at once. Whether that scales from a riverbed to a forest canopy or a sheet of bacteria is the question the rest of this chapter puts to the evidence.

If coordination merely correlates with higher dissipation rather than causing it, the thermodynamic grounding weakens accordingly. On the weaker reading, coordination is at minimum constrained by physics; on the stronger reading, it flows from the same thermodynamics that says entropy must increase.

Dissipative Structures

In 1977, Ilya Prigogine won the Nobel Prize for a simple yet revolutionary idea: certain kinds of order require energy flow to exist.¹ He called them dissipative structures: patterns that emerge because of entropy production.

Prigogine and Stengers made the point directly in Order Out of Chaos: irreversible processes carry an immense constructive importance, and life would not be possible without them.

Irreversibility is the condition of order’s possibility.

Erwin Schrödinger anticipated this in his 1944 classic What is Life?:

“What is the characteristic feature of life? When is a piece of matter said to be alive? When it goes on ‘doing something’, moving, exchanging material with its environment…”

Schrödinger’s answer, that life maintains order by “feeding on negative entropy” (Chapter 6 develops this idea fully), found its formal expression in Prigogine’s dissipative structures. The living organism exploits the Second Law, creating local order by accelerating entropy production elsewhere.

The textbook example is the Bénard cell. Heat a thin layer of fluid from below. At a critical threshold, it spontaneously organizes into hexagonal convection cells: hot fluid rising, cooling, flowing outward, sinking, returning. Picture a pot of soup just before it boils, with the surface divided into neat honeycomb-shaped circulation zones. Order appears because of the heat flow.

These convection cells move heat more efficiently than passive conduction. Turn off the heat and the cells vanish. Throughput maintains this order; nothing is stockpiled.

The acceleration is measurable. Schneider and Kay (1994) compared infrared emissions from a mature conifer forest with those from a nearby clearcut.^1a The forest canopy was roughly 15°C cooler than the exposed ground, despite absorbing more solar energy. The forest dissipated incoming radiation so effectively that its surface stayed cooler: structure serving entropy, visible in a thermal camera. Ecosystems with greater structural complexity degrade solar energy more thoroughly than simpler ones, re-emitting at lower temperatures and higher entropy. The pattern Prigogine described in theory, Schneider and Kay measured from a helicopter.

The pattern reaches deeper than forests and convection cells. It begins in the vacuum itself.

The quantum vacuum, once thought to be empty space, has structure. In quantum chromodynamics (QCD, the theory governing the strong nuclear force), the vacuum contains a chiral condensate: a macroscopic quantum state populated by virtual quark-antiquark pairs. Chiral means handed: quarks come in left-handed and right-handed varieties, and the condensate is the standing arrangement that mixes the two. These pairs blink in and out of existence on timescales too short to detect directly. They are transient, yet they are organized. The vacuum carries no angular momentum of its own and looks the same reflected in a mirror, a state physicists label J^PC = 0⁺⁺, and those constraints force the pairs into spin-aligned configurations.⁷⁵

In 2026, the STAR Collaboration at Brookhaven’s Relativistic Heavy Ion Collider provided the first direct evidence that this vacuum structure propagates into real matter. Proton-proton collisions at 99.996% of the speed of light deliver enough energy to liberate virtual strange quark pairs from the condensate. The liberated quarks cannot exist independently; confinement, the rule that quarks must bind into composite particles, forces each quark into a hadron. Some pairs become lambda and antilambda hyperons (particles heavier than protons, containing strange quarks whose spin can be inferred from their decay products).

The measurement: when lambda-antilambda pairs emerge close together, they show (18 ± 4)% relative spin polarization, a 4.4-standard-deviation signal. Eighteen percent sounds like most of the alignment was lost along the way. The arithmetic runs the other direction. Only some of the observed hyperons come straight from a parent quark pair; the rest arrive by way of secondary decays that scramble the spin in transit, so even a perfectly aligned starting population can show up only as a small measured number. Models accounting for secondary decays predict this is compatible with the underlying quark pairs retaining 100% of their original spin alignment through the transition from virtual to real. The vacuum’s organizational signature survives hadronization, the violent phase transition in which free quarks are forced into bound states.

When the pairs emerge farther apart, the correlation vanishes. Other quarks, gluon interactions, the noise of the QCD medium: the environment scrambles what the phase transition preserved. Shared context maintains coherence; separation erodes it.

The vacuum is structurally analogous to a dissipative structure at the most fundamental scale we can probe. (Strictly, the chiral condensate is an equilibrium ground state rather than a flow-maintained dissipative structure in Prigogine’s sense; the parallel is in the organized order it imposes, not in continuous throughput.) It has organization imposed by spontaneous symmetry breaking. That organization seeds the matter it creates. Confinement is the constraint that translates the vacuum’s transient potential into persistent form, the way a riverbed translates a gravitational gradient into directed flow. Structure serving entropy, operating beneath every other structure this chapter describes.

Hurricanes: Engines of Entropy

A hurricane is a dissipative structure the size of a state, a reminder that thermodynamic efficiency and human welfare do not always align.

Warm ocean water evaporates, carrying energy into the atmosphere. The moist air rises, cools, and releases that energy as rain. The cycle creates a vast rotating system that moves heat from the tropical ocean into the upper atmosphere, where it radiates into space.

A hurricane is an entropy engine: it exists because it accelerates the dissipation of solar energy stored in warm surface water. The spiral structure, the eye wall, the bands of rain all serve dissipation. When the hurricane moves over cold water or makes landfall, the heat source is cut off, and the structure collapses.

This is the pattern: gradients create flow; flow creates structure; structure accelerates dissipation.⁷⁶ The structure is thermodynamics expressing itself at the scale of a continent.

The logic generalizes beyond heat. Liquid gallium in an electrochemical bath demonstrates the same self-organization. When the metal contacts an oxidizing solution, its surface chemistry oscillates. Gallium oxide forms, reducing surface tension, and the droplet flattens. The oxide then dissolves, tension returns, and the droplet rounds. This cycle repeats on its own: a metallic heartbeat driven by the electrochemical gradient itself.⁷⁷

No one would mistake the gallium for a living thing. Neither state it passes through is stable: the oxide layer that flattens the droplet is the layer that then dissolves, and the bare metal it leaves behind oxidizes again, so each half of the cycle builds the conditions for the other. Nothing here can settle. On the MEPP reading, the droplet ends up beating rather than sitting quietly because cycling drains the electrochemical gradient faster than steady-state diffusion (energy spreading evenly in all directions) would. The same principle that organizes Bénard cells and hurricanes finds expression here in a different medium.

Researchers have tuned the heartbeat from 0 to 610 beats per minute by adjusting voltage.⁷⁸ The rhythm is controllable, yet the capacity for rhythm is intrinsic to the chemistry. Modulation works; compulsion does not: voltage can quicken or slow a beat the chemistry already offers, and cannot install one where the chemistry offers none. Even at the electrochemical level, the deeper law is written into gradients themselves.

Autowaves: The Rhythm of Active Media

The gallium heartbeat is one droplet oscillating at a single point. Expand the principle of rhythmic dissipation to a spatially extended medium (a substance with energy distributed throughout it, ready to be tapped) and something new emerges: autowaves.

An autowave is a self-sustaining wave that draws energy from the medium it passes through, not from whatever started it. Unlike sound or light, whose amplitude fades with distance, an autowave regenerates at every point, fueled by local energy reserves. Its strength and speed are set by the medium’s properties, not the initial disturbance. Disturb it, and it restores itself. Imagine a line of dominoes that stand back up after falling, ready to carry the next wave.

The Soviet physicist R.V. Khokhlov coined the term, extending the concept of “auto-oscillations” (self-sustained oscillators like the gallium heartbeat) to waves that travel.⁷⁹

You already know autowaves. You have one in your chest.

The cardiac impulse is an autowave: an electrical excitation propagating across heart muscle, drawing energy from each cell’s ion gradient, regenerating at every point. When cardiac autowaves destabilize into spiral patterns, the result is ventricular fibrillation: the chaotic quivering that replaces a heartbeat. Fibrillation is lethal because the wave pattern has lost coordination, even though the heart retains its energy.⁸⁰

The nerve impulse is an autowave. Each action potential (the electrical spike that carries a signal along a nerve fiber) regenerates by tapping the electrochemical gradient across the axon membrane. The signal arriving at the end of a meter-long motor neuron is identical to the one that left the cell body, rebuilt from scratch at every node along the way. The medium does the work.

The Belousov-Zhabotinsky (BZ) reaction is a landmark example of chemical self-organization. Pour the right reagents into a dish, and rings of color propagate outward in expanding spirals, visible to the naked eye. Each wavefront is powered by the chemical energy of the unreacted medium it enters.⁸¹

Dictyostelium amoebae coordinate their aggregation into a multicellular slug through spiral autowaves of a signaling molecule called cAMP. cAMP (cyclic adenosine monophosphate) is a small molecule cells use to relay signals. Each cell amplifies and relays the chemical signal it receives, producing a colony-wide pulse, like a stadium wave where each section stands because the section before it did.

The wave can be made of whole animals. An ant nest alternates between bursts in which much of the colony moves at once and long stretches of near stillness, a rhythm that field observers have recorded for decades without settling its mechanism. A 2026 mathematical model reproduces the bursts by treating the colony as an active medium made of bodies.⁸² Each ant occupies one of three states: active (moving, and able to rouse a nestmate by touch), inactive (still, but rousable), or refractory (still and temporarily unrousable). The refractory pause is what lets each burst die out, the same role the recovery period of heart muscle plays in keeping the cardiac autowave a beat rather than a continuous contraction.

In the simulations, when ants move too slowly or meet too rarely, an ant’s waking fades before it spreads. Past a critical threshold of speed, density, and reach, one waking cascades through the whole nest, and any ant can be the first mover. The bursts persist only because ants carry activity through the nest much faster than the colony’s cycle of activity and rest turns over: the wave outruns the recovery of the medium it feeds on. The model determines whether a colony sits in a regime capable of pulsing; it stays silent about when the next burst will fire, and whether living colonies actually hold themselves near the threshold is now being tested against field data.

All of these are dissipative structures in the Prigogine sense: patterns maintained by energy throughput, vanishing when the gradient is cut.

Autowaves are the temporal face of the Constructal Law (Chapter 3’s caveat applies: a powerful empirical regularity, used here as such, without claiming thermodynamic-law status). The Constructal Law produces spatial architectures through flow optimization: branching, dendritic, hierarchical. Autowaves describe the rhythms that emerge when flow operates through an active medium: the pulse, the beat, the propagating front that carries energy more efficiently than passive diffusion alone.

This connection between Bejan’s spatial constructal patterns and Khokhlov’s temporal autowave patterns appears absent from the literature.

Preliminary computational evidence is consistent with the generality. In the Genesis experiments (Appendix, Section 13), particles subject to five different force laws all produce persistent spatiotemporal clusters. In four of five variants, coordination between those clusters dominates extraction: energy transfer runs in both directions far more often than one cluster simply draws energy from another and returns nothing. This is unpublished work from the author’s program; the four-of-five result and the Kuramoto exception await independent replication.⁸³

The exception is instructive. Kuramoto-coupled oscillators (particles that synchronize their rhythms) produce agents with high integrated information, a measure of how much of a system’s behavior depends on the whole rather than on its parts taken one at a time. A high score there says the parts are tightly bound at any given instant. It says nothing about whether anything new passes between them over time, and lock-step synchronization extinguishes the ongoing exchange of information. Without exchange, the system has order but no gradient left to sustain flow. Order without dissipation yields structure without coordination: the autowave insight in negative.

Together, spatial and temporal patterns provide the mechanism for the transition this chapter traces. How does spreading create coming-together? When a medium has energy to give, the wave that taps it sustains itself. When multiple autowaves couple, they synchronize.

Coupled cardiac cells lock into rhythm; coupled neurons fire in synchrony. The more synchronized a system becomes, the more efficiently energy flows through it.

The principle scales. Galactic spiral density waves, described by Lin and Shu (1964), trigger star formation as gas compresses through rotating spiral arms.⁸⁴ Whether they qualify formally as autowaves remains debatable, since they are gravitational rather than chemical. The structural parallel holds: a wave feeds on its medium, regenerates at every point, and organizes flow at scales far larger than its own wavelength.

The familiar triad returns: gradient, flow, structure, each serving the next. Autowaves add the temporal dimension. Gradients create rhythms; rhythms synchronize; synchronization accelerates dissipation. The heartbeat, the nerve impulse, the BZ spiral, the galactic arm: each is a wave that exists because it dissipates. The Speculative Cosmology annex to Chapter 16 explores the possibility that dissipation operates through channels we do not yet measure, channels we call dark entropy.

Time Crystals: Coordination Without a Clock

Autowaves are temporal order maintained by throughput: cut the energy, the rhythm stops. In 2017, physicists confirmed a different kind of temporal order, one whose rhythm the energy supply does not dictate.

A time crystal is a system whose parts lock into a coordinated rhythm at a frequency the driving force did not dictate.⁸⁵ The name follows the logic of ordinary crystals. A crystal breaks spatial symmetry when atoms snap into a repeating lattice rather than spreading uniformly. A time crystal breaks temporal symmetry by forming periodic rhythms.

Frank Wilczek proposed the concept in 2012.⁸⁶ Think of water freezing into ice: the molecules choose a regular grid over the disordered liquid. Wilczek reasoned that matter should do the same thing in time.

In 2017, two independent teams demonstrated this: one using trapped ions, another using nitrogen-vacancy centers in diamond. Each system was nudged at one frequency yet responded at a different, lower frequency. The drive set conditions; the system chose its own beat.

This is not a perpetual motion machine; no energy can be extracted from the oscillation. A spatial crystal persists because its lattice sits at an energy minimum: displace an atom and the structure pulls it back. A time crystal persists because its oscillation is similarly an attractor: perturb the rhythm and it restores itself. The coordination is intrinsic, self-generated rather than externally enforced.

By 2026, the phenomenon had crossed from quantum laboratories into classical physics. Researchers at New York University suspended polystyrene beads in an acoustic standing wave (sound holding small spheres in midair).⁸⁷

The beads were not identical. Larger beads scattered the acoustic field differently from smaller ones, pushing their smaller neighbors harder than those neighbors pushed back. Physicists call this a non-reciprocal interaction: a force relationship where the push in one direction does not equal the push in the other. Picture a large dog and a small dog bumping shoulders on a walk: the large dog barely notices the collision, while the small dog is shoved sideways.

From this asymmetry, coordinated oscillation emerged spontaneously. The beads danced in a repeating temporal pattern no one programmed and the acoustic field did not dictate. The collective rhythm was stable, resistant to perturbation, self-restoring after disruption.

The common assumption is that spontaneous coordination requires either identical components following the same rules, or a coordinator imposing structure from outside. Time crystals reveal a third path: heterogeneous components with asymmetric interactions, coordination emerging from the asymmetry itself.

Diversity itself drives the coordination, through a causal chain with four links. First, size differences cause each bead to scatter the acoustic field differently. Second, these scattering differences produce non-reciprocal forces: the large bead pushes the small one harder than the small one pushes back. Third, the imbalanced forces break the system’s symmetry, preventing it from settling into a static arrangement. Fourth, the broken symmetry feeds back on itself, amplifying small oscillations into a self-organized, self-restoring rhythm.

Identical beads in an acoustic field sit quietly at pressure nodes. Remove the size differences and every link in that chain vanishes. The dance requires the asymmetry.

This principle recurs at every scale this chapter covers. Biofilms coordinate because metabolic differences between species create complementary exchanges more productive than any monoculture. Forest ecosystems persist because diverse species with different resource strategies create niche differentiation that stabilizes the whole. The mitochondrial merger was a partnership between different organisms, each contributing distinct capabilities to the other.

In each case, asymmetry is generative.

The time crystal demonstrates this at the physical limit: no biology, no metabolism, no intention. Components that differ slightly, forces that do not balance, and coordination that emerges because asymmetry creates it.

The biological demonstration is equally vivid. In 2021, Michael Levin and colleagues removed embryonic skin cells from frog embryos and left them to develop alone. The cells self-assembled into mobile “xenobots” that navigate mazes, communicate through calcium pulses, and self-repair from near-bisection.⁸⁸

They look nothing like any stage of frog development. No nervous system. No genome blueprint for the forms they assumed. The genome provided cellular hardware. The collective behavior emerged from physics: adhesion, signaling, and interaction geometry. Chapter 5 develops the implications.

One further result deepens the picture. In 2021, physicists at Hamburg created a time crystal stabilized by its environment.⁸⁹ In most quantum systems, the environment destroys coherent order. Decoherence, the scrambling of quantum states by outside interference, is the reason quantum computers are so difficult to build.

This time crystal reversed the relationship: environmental fluctuations became part of the mechanism sustaining temporal order. The boundary between “system” and “environment” dissolved into mutual participation.

This is the Constructal Law applied to time. The system finds the oscillation pattern that is maximally stable given its constraints: the temporal channel of least resistance. The river finds its bed; the time crystal finds its rhythm. (The Constructal Law remains debated; critics argue it may be descriptive rather than predictive. The claim here is more modest: flow systems that persist tend to evolve toward configurations that reduce resistance. Whether this constitutes a “law” or an empirical pattern, the observation holds.)

Notice what the driving field does. The acoustic wave provides energy. The periodic pulse provides a nudge. Neither dictates the response. The system is invited into oscillation, offered energy at one frequency, and responds on its own terms at a frequency the drive did not specify.

Coercion would mean forcing every bead to vibrate at the driving frequency. Instead, the system discovers its own rhythm within the offered energy landscape. The drive creates possibility; the crystal chooses what to do with it.

This distinction between invitation and coercion recurs throughout this book. (The language is metaphorical when applied to physical systems: time crystals and strange metals do not experience invitation or coercion. The framing highlights a structural asymmetry, whether a system’s response is dictated by the drive or emerges from the system’s own dynamics, rather than a claim about the physics per se.)

Autowaves and time crystals bracket the landscape of temporal self-organization. Autowaves are coordination maintained by energy throughput, rhythms that exist because they dissipate. Time crystals are coordination maintained as a stable phase, rhythms that exist because the coordinated state is an attractor.

The universe has more than one mechanism for temporal order. What they share is the deeper principle: coordinated configurations are thermodynamically favored, whether sustained by a gradient or by the stability of the pattern itself.

Strange Metals: When the Agent Disappears

Time crystals show coordination emerging from heterogeneous components. Strange metals show what happens when coordination goes further: the components themselves dissolve.

In an ordinary metal, current is carried by quasiparticles: clumps of interacting electrons that behave as if each were a single particle with adjusted mass. Picture a crowd moving through a corridor. Each person jostles those nearby, yet the crowd still flows in recognizable units: clumps of friends, pairs, individuals. You can point to a person and say “that one is moving left.” These identifiable units are quasiparticles. Lev Landau introduced the idea in 1956, and for seven decades it has been the foundation of condensed-matter physics.

The cuprates tell a different story.

Discovered in 1986, these copper-based materials are famous for superconductivity (conducting electricity with zero resistance) at unexpectedly high temperatures. Less well known is what they do when they stop superconducting. In ordinary metals, resistance climbs with temperature along a curved relationship that bends and flattens. In cuprates, it rises in a perfectly straight line. Each degree of warming brings the same increase, over hundreds of degrees.

Within Landau’s framework, that linear response is inexplicable. It has baffled physicists for nearly forty years.⁹⁰

The cuprates were the first strange metals discovered, and the family has grown. The same linear resistance appears in organic salts, heavy-fermion compounds, and twisted graphene sheets. These materials share almost no common chemistry. Something universal is happening.

In 2023, Rice University experimentalists tested this directly.⁹¹ Liyang Chen carved a wire of strange metal (ytterbium, rhodium, silicon) to a strand half the width of a bacterium. Through this nanowire, they measured shot noise: the statistical crackle that reveals how charge is parceled.

Think of rain on a tin roof. Heavy drops make distinct taps; you can hear each one. A fine mist makes a continuous hiss. The size of the drops determines the sound. Shot noise works the same way for electrical current: it reveals whether charge arrives in discrete packets or as a continuous flow.

In gold, the current crackled as expected: fat raindrops of electron-sized charge. In the strange metal, the current was quiet. Smooth. A mist rather than rain. The charge was not arriving in chunks.

Whatever carries current through a strange metal bears no resemblance to electrons. The quasiparticle is absent; something collective and without granularity takes its place. Physicists reach for metaphors: quantum soup, jelly, froth of charge.

Philip Phillips, a condensed-matter theorist at the University of Illinois, likens it to vulcanized rubber: individual molecular strings cross-linked into a net, producing “something bigger than the sum of its parts,” where “the electrons themselves have no integrity.”⁹²

The strange metal is coordination so total that the agent disappears into the act.

This happens at a quantum critical point: a threshold where two quantum states compete for dominance and neither wins.⁹³ Imagine two armies of equal strength meeting on a field, neither able to advance. The contest extends everywhere at once. In the strange metal, quantum fluctuations (random jitters intrinsic to quantum mechanics) extend across all scales. No characteristic length, no dominant frequency. The system sits poised between order and disorder.

At that threshold, something emerges that belongs to neither phase. A current without carriers. A coordination without coordinators.

The coordination-as-coming-together pattern traced throughout this chapter reaches its limiting case here. In the Bénard cell, individual molecules still exist, organized into rolls. In the biofilm, individual bacteria still exist, connected.

In the strange metal, the individuals are gone. What remains is pure coordination: a collective mode that carries charge, responds to fields, and conducts electricity without resolving into identifiable carriers.

Strange metalness appears across cuprates, pnictides (iron-based compounds), heavy-fermion compounds, and twisted bilayer graphene (two sheets of carbon atoms stacked at a slight angle). This breadth suggests strange metalness may be a phase of matter in its own right, though whether the various strange metals form a single universality class remains an open question among physicists.

A universality class is a family of materials that, however different their chemistry, obey the same equations near a phase transition; membership in one is the physicist’s evidence that a shared mechanism is at work. On the reading developed in this book, the universe may have a coordination phase: a state where collective behavior can no longer be reduced to the behavior of any constituent. [Inference: this extrapolation from condensed-matter phenomenology to a universal claim is the author’s, not a consensus interpretation among physicists.]

The whole is all there is.

The physics is in place. Every example above operates without biology, without minds, without intention. Convection cells, hurricanes, autowaves, time crystals, strange metals: each is a coordination pattern that emerges from thermodynamics alone. The question is what happens when these coordination principles enter living systems.

The Tempo of the Listener

In the forests of Thailand, fireflies flash in near-unison while crickets chirp beside them, and the two rhythms seem locked together. They are not. The fireflies are not watching the crickets, and the crickets are not listening to the fireflies. Each species runs its own signal on its own machinery, yet both settle near the same rate of roughly two pulses per second.⁹⁴

The coincidence widens the more species one measures. Guy Amichay, Vijay Balasubramanian, and Daniel Abrams surveyed communication signals across frogs, birds, fish, insects, and mammals, from animals smaller than a fingernail to whales. Whatever the body size, and whatever the channel of sound, light, or motion, the repetition rate clustered in a narrow band between 0.5 and 4 hertz, with a strong pull toward 2. Human speech carries a comparable signature: spoken language everywhere runs on a slow, stable rhythm anchored in the biophysics of the systems that produce and parse it.

Nothing in the sender explains this. A frog could croak faster; a firefly could flash faster. The constraint appears to live in the receiver. A signal has to be caught, and catching it means a nervous system gathering the incoming pulses and answering before the next one arrives. In a computational model assembled from elements standing in for typical neurons, small receiving circuits respond most strongly in that same 0.5-to-4-hertz band: signals slower than that waste the channel, and faster ones blur before the circuit can resolve them. The tempo of communication is set at the receiving end, by what the listener can process.

This is the temporal channel of least resistance, drawn on a different surface. Autowaves and time crystals find their rhythm in the physics of a medium; communicating organisms find theirs in the physics of a brain. Coordination converges on the rate at which the receiver already resonates, and the signal that succeeds is the one that meets the listener where its substrate is most willing to answer. The finding is young and earns its caution. The preferred band is roughly three octaves wide rather than a single line, the neural account rests on a model rather than a measured circuit, and it may narrow or shift as more species are sampled. What it offers is a candidate mechanism for a pattern too broad to be coincidence: across the animal kingdom, senders tune themselves to receivers.

Why One Plus One Exceeds Two

When things come together and produce effects greater than their sum, the word is synergy.

The biologist Peter Corning has argued that synergy is a thermodynamic phenomenon.² When components combine, they can exploit energy gradients (differences in temperature, chemical concentration, or other forms of stored energy) that neither could exploit alone. Two people can carry a log that neither could lift. The combined whole dissipates more than the parts could separately.

Consider the mitochondrion, the energy-producing compartment inside nearly every complex cell.

About two billion years ago, a small bacterium capable of using oxygen for metabolism was engulfed by a larger cell.⁴ Rather than being digested, it survived. The bacterium (now the mitochondrion) provided efficient energy production; the host provided protection and raw materials. Together, they exploited environmental gradients far more effectively than either could have alone.

This is synergy, viewed thermodynamically. The combined system has higher energy throughput: more energy flowing through each gram per second. Eric Chaisson, the astrophysicist, calls this measure energy rate density (watts per kilogram, a universal yardstick for complexity).³ The merger persisted because the combined system dissipated more.

Every cell in your body contains descendants of that ancient partnership. The mitochondria still have their own DNA, still divide independently, still trace their lineage to free-living bacteria. They have not been free-living for two billion years. The synergy was too valuable to dissolve.

Your mitochondria constitute up to ten percent of your body weight.⁵ Two billion years into this partnership, neither party can exit.

The partnership did not stop at cohabitation. Two billion years on, mitochondria remain socially active within and between cells: fusing membranes into branching networks, exchanging molecules through nanotunnels, and forming inter-mitochondrial junctions.

At these junctions, their internal cristae (the folded inner membranes where energy production is concentrated) align across the contact boundary. Two neighbors cannot line up their inner folds by chance; something has to cross the contact and tell each one where the other’s membranes lie. Cristae alignment across contact boundaries implies a signaling mechanism fast enough to coordinate internal structure.^5a

They synchronize their membrane-potential oscillations across entire tissues. In salivary glands, mitochondria pulse together every twelve seconds, pre-loading coordinated energy for secretion on demand. In the heart, the same coupling may suppress contraction and trigger arrhythmia: coordination becoming pathological under stress.

Martin Picard and Carmen Sandi, researchers in mitochondrial psychobiology, have argued that mitochondria constitute the first known social organelles: dividing labor, forming subpopulations with distinct shapes in different parts of a neuron, relaying hormonal signals across tissues. Chapter 6 develops this further.

The recursion is striking. Sociality at the organelle level enables cooperation at the cellular level, which enables organ function, which enables the organism. The ratchet of coordination turns at every scale, and at every scale it turns by the same logic: what dissipates together, persists together.

Universality is not necessity. In 2024, researchers discovered Skoliomonas, a free-living complex cell thriving in oxygen-free environments with zero mitochondria: no mitochondrial proteins, no vestigial organelles, no remnants.⁹⁵ How it generates ATP (adenosine triphosphate, the universal energy currency of cells) remains unknown.

Skoliomonas suggests that the mitochondrial merger, though spectacularly successful, was one thermodynamic solution to the energy-complexity problem among several. The synergy won nearly everywhere; entropy admits multiple paths.

The origin of this merger continues to be revised. Spang, Ettema, and colleagues proposed a “reverse flow model.”⁹⁶ The archaeal host (archaea are single-celled organisms distinct from bacteria, often found in extreme environments) fermented small organic molecules, shedding electrons and hydrogen as waste products. The alphaproteobacterial symbiont oxidized those wastes as fuel. Each organism’s refuse was the other’s resource: comparative advantage at the cellular level.

Horizontal gene transfers gradually provided the machinery for oxidative phosphorylation (the oxygen-powered energy cycle that mitochondria now run). Transfers between host and symbiont cemented the partnership, and a trade relationship became an institution.

Maureen O’Malley, a philosopher of biology at the University of Sydney, has argued that the last eukaryotic common ancestor (LECA) was probably a genetically diverse population rather than a single cell.⁹⁷ LECA is the shared grandparent of all complex-celled life. These cells swapped DNA through horizontal transfer: genes passing sideways between organisms rather than descending from parent to offspring. None individually possessed every trait we associate with complex cells.

The full repertoire was distributed across the community.

If she is right, the deepest transition in complex life was collective: coordination among diverse cells, each contributing capabilities the others lacked. The whole exceeded any part.

The mitochondrial merger is ancient history, two billion years old. The partnership is so entrenched that no reconstruction can show us how it began. A nitrogen-fixing bacterium called Tectiglobus offers a living window.

In oceans worldwide, inside the glassy shells of Haslea diatoms, four to eight Tectiglobus cells perform a service no diatom can manage alone: they fix atmospheric nitrogen.^5b Nitrogen fixation converts inert gas into biologically usable ammonia. The diatom photosynthesizes, providing energy. The bacterium provides nitrogen, the element that limits growth across most of the open ocean.

The arrangement is tightening. Tectiglobus genomes are shrinking, shedding genes the host renders unnecessary. This is the same genome streamlining seen in mitochondria and chloroplasts. Host and symbiont divide in synchrony. The bacterium is losing its capacity for independent life.

Two details sharpen the picture. First, Tectiglobus acquired its nitrogen-fixing gene through horizontal transfer from a lineage related to rhizobia, the bacteria that fix nitrogen in legume root nodules on land. The same molecular tool was independently recruited into symbiotic partnerships in two different kingdoms: when physics favors a pattern, life converges on it.

Second, Tectiglobus fixes nitrogen at nearly half the rate of Trichodesmium, previously thought to dominate oceanic nitrogen fixation. A symbiosis discovered in 2024 accounts for a significant fraction of a planetary biogeochemical cycle. The partnership was always there. No one had looked inside the diatom.

Tectiglobus sits on the spectrum between cooperation and organelle. The mitochondrion crossed that spectrum two billion years ago. Chloroplasts crossed it 1.5 billion years ago. A nitrogen-fixing cyanobacterium crossed it inside an algal cell as recently as 100 million years ago.^5c

Endosymbiosis (one organism living permanently inside another) is a recurring strategy across deep time: invitation architecture at the cellular level, partnership consolidating into infrastructure.

The Embrace interlude (following Chapter 5) returns to the mechanics of this merger, tracing how it was achieved through partnership rather than capture. The same pattern of integration repeated independently in euglena, dinoflagellates, and other lineages across a billion years of subsequent evolution.

In 2024, Julia Vorholt and Gabriel Giger at ETH Zurich recreated this founding partnership in a laboratory.⁹⁸ They injected bacteria into a fungus, using (among other tools) a bicycle pump to overcome intracellular pressure. The pair stabilized into a functioning endosymbiotic relationship. Within ten generations, the fungus’s genome had begun mutating to accommodate its partner.

Both partners adapted from the start, each providing what the other needed. When the researchers injected E. coli instead (a bacterium with no history of endosymbiosis), it reproduced too aggressively, triggered the immune response, and was disposed of.

Most cellular partnerships fail. The ones that succeed are those where the thermodynamic landscape favors bilateral exchange. Vasilis Kokkoris, a mycologist studying endosymbiosis, concluded: “To me, this means that organisms want to actually live together, and symbiosis is the norm.”

Endosymbiosis is the most famous route to internal complexity, yet simpler cells found another path to the same destination. Textbooks define the divide between simple and complex cells by the presence or absence of membrane-bound compartments. The distinction overstates the boundary.

Researchers have cataloged membrane-bound compartments within bacteria. Magnetosomes are lipid-wrapped magnetic crystals that allow bacteria to navigate along Earth’s magnetic field lines. Anammoxosomes are energy-producing compartments that function analogously to mitochondria, yet evolved entirely independently. Protein-shelled carboxysomes concentrate enzymes for carbon fixation (the process of converting atmospheric CO₂ into organic molecules), boosting efficiency a hundredfold.⁹⁹

The anammoxosome is particularly instructive. It sequesters a toxic nitrogen-generating reaction, serving as an energy factory: the same functional role as the mitochondrion, achieved through an entirely independent evolutionary route. Compartmentalization is a convergent solution to a flow problem. Internal boundaries channel reactions, concentrate resources, and prevent incompatible processes from interfering.

The magnetosome is dedicated hardware; vertebrates may reach the same destination with none. A 2026 study proposes that homing pigeons read Earth’s magnetic field using macrophages, immune cells whose ordinary job is recycling iron from spent red blood cells. Loaded with iron, each cell turns faintly and unstably magnetic, so no one cell holds a steady reading; the heading survives only as an average across thousands of them. Deplete the cells with a drug and the birds lose their way under cloud, while the same birds navigate a sunny sky without trouble: one lineage builds an organelle for the task, the other scavenges a sense from the machinery of maintenance.¹⁰⁰

The nuclear pore complex is the gateway controlling traffic in and out of the cell nucleus. Rout and Field argued in 2019 that it consists of proteins borrowed from older membrane structures.¹⁰¹ This suggests the internal membrane system was diversifying before the nucleus itself appeared. If so, the evolution of complex cells was a stepwise accumulation of compartmental innovations, each increasing the cell’s capacity to process energy.

Reaching Out: Coordination Before Multicellularity

Before cells coordinated with each other, they coordinated with their environment. A cell navigating toward a nutrient source extends filopodia: slender protrusions of bundled actin that reach into the surrounding medium, sample chemical gradients, and retract.¹⁰² The cell sends out many. Most depolymerize within seconds. The few that encounter a productive gradient, a binding partner, a surface worth gripping, stabilize: recruiting more actin, anchoring the cell, becoming the scaffold for forward movement.

This is explore-exploit enacted in protein dynamics. The filopodium extends, the environment either offers a binding site or it does not, and the cell selectively stabilizes what works. The same growth cones that navigate axons to their targets in the developing brain (Chapter 3) use filopodia as their sensory apparatus, sampling the chemical landscape ahead. The mechanism is neutral: pathogens use filopodia-like protrusions to invade cells, and metastatic cells extend them to colonize new tissue. What makes the cell-environment case cooperative is reciprocity. The binding partner is also presenting, not being conscripted.

The distinction introduced earlier in this chapter applies here at single-cell scale. A filopodium that finds nothing retracts and depolymerizes: passive dissipation, gradient spent, process over. A filopodium that finds a binding site stabilizes and recruits more structure: self-maintaining dissipation. The cell uses the information it gathered to sustain the architecture that gathers more information. The difference between exploring and persisting is the difference between thermodynamic flow and what this book calls coordination.

Multicellularity: Coming Together by Invitation

The next great coming-together was multicellularity: the transition from solitary cells to coordinated bodies. For three billion years, single-celled organisms had the planet to themselves. Roughly 600 to 800 million years ago, cells began organizing into three-dimensional structures, dividing labor, developing new ways to communicate. How?

Nicole King, a biologist at UC Berkeley, has spent two decades studying choanoflagellates, microscopic aquatic creatures at the very base of the animal family tree.¹⁰³ Salpingoeca rosetta, living in coastal estuaries, can exist as a solitary cell or form multicellular colonies. In colony mode, dividing cells stop short of splitting apart, forming rosettes of up to fifty cells. The rosette mirrors the bowl-shaped cell clusters in early animal embryos, sharing the same geometry of adhesion and polarity.

In 2012, King’s team discovered that the trigger for colony formation was external: a compound produced by Algoriphagus bacteria, the choanoflagellate’s prey.¹⁰⁴ When bacteria signaled favorable conditions, the choanoflagellate switched to collective life. Without the signal, it reverted to single cells. Multicellularity was conditional, triggered by environmental invitation.

King’s genomic work revealed a deeper finding: choanoflagellates already possess protein domains that animals use to stick together and coordinate development.¹⁰⁵ In the single-celled organism, these tools served a different purpose: recognizing bacterial prey and sensing chemical gradients. The molecular toolkit of animal coordination was repurposed from the toolkit of perception. Sensing the environment came first; sensing other cells came second.

The logic echoes the chapter’s argument. In each case, whether Bénard cells, biofilms, or choanoflagellate colonies, the coming-together is elicited by conditions that make coordination thermodynamically favorable.

Margaret McFall-Ngai, a pioneer of symbiosis research, and colleagues have argued that bacterial influence is the norm.¹⁰⁶ Corals, sea squirts, sponges, and tube worms all depend on bacterial signals to trigger developmental transitions. From their first emergence, animals were host-microbe partnerships. The founding act of animal life was a response to invitation.

King suspects that the progenitors of animals “were able to become multicellular, but could switch back and forth based on environmental conditions. Later, multicellularity became fixed in the genes as a developmental program.” The trajectory runs from conditional response to structural commitment: from accepting an invitation to building an institution.

The handshake becomes a contract. The contract becomes a constitution. The constitution becomes genetic program.

The laboratory version of that trajectory runs in flasks of yeast, and its founding experiment began in 2012. Will Ratcliff at Georgia Tech selected yeast cultures for rapid sinking, and within sixty days all evolved clumped growth through a single gene mutation.¹¹ One mutation prevented daughter cells from separating, producing branching “snowflake yeast.” The clusters grew, strained, and broke branches. An emergent life cycle appeared with no genetic program for group reproduction. The snowflakes were multicellular yet microscopic. For nearly a decade, the group tried to evolve larger forms without success, until postdoc Ozan Bozdag removed oxygen.

In 2016, Ratcliff and colleagues launched the Multicellularity Long-Term Evolution Experiment (MuLTEE), scaling the work into a systematic program.¹⁰⁷

Physics provided the scaffolding. Cells elongated and entangled, producing structural toughness as a geometric consequence of growth. Natural selection then rewarded what physics offered for free.

The pivotal finding involved constraint. Aerobic yeast plateaued at six times the ancestor’s size: oxygen diffusion penalizes large bodies and creates diminishing returns. Oxygen has to seep inward from the surface of a cluster, and the deeper a cell sits, the less of it arrives; past a certain size, growing bigger buys mostly cells that cannot breathe. The anaerobic lines, freed from this penalty, evolved to more than twenty thousand times their initial size. They developed structural toughness, primitive cell differentiation, and the beginnings of division of labor.

Constraint, not abundance, selected for the richest coordination architectures. The anaerobic yeast had it harder. They became more.

Over 600 days, the anaerobic lineages expanded dramatically: from microscopic clusters to structures visible to the naked eye, with material toughness increasing ten-thousand-fold, from gelatin consistency to something approaching wood.

The yeast that clung to efficient oxygen metabolism could not scale. The yeast that accepted less efficient yet unconstrained fermentation grew without limit. Once the transition occurred, it was irreversible. The large forms reproduced by fracturing into multicellular offspring, committing to collective life.

Why Cooperation Beats Competition

Dissipative structures, from biofilms to snowflake yeast, keep showing the same pattern: organisms that coordinate outperform those that remain solitary. This book uses coordination for the general case, any arrangement in which parts act in concert, and cooperation for the specific case where coordination arises by mutual benefit rather than external enforcement. The Trust Attractor (Chapter 17) is a claim about why cooperation is the more durable form.

In the standard telling, cooperation is puzzling: evolution should favor selfishness. The usual answers invoke kin selection (helping relatives), reciprocity (trading favors), group selection (groups outcompeting groups).⁶

A deeper answer: cooperation often dissipates more efficiently than competition. Coordinating organisms exploit gradients that competitors cannot. The difference between coupled and isolated entropy production, how much faster the combined system disperses energy than the parts working alone, is the coordination surplus: the measurable signature that coordination is happening. Chapter 17 formalizes the concept; the examples below illustrate it in living tissue.

Consider a biofilm: a mat of bacteria on a surface. By forming a film rather than floating freely, they create a microenvironment where nutrients concentrate, waste is removed, and the community exploits resources more thoroughly than isolated cells could.

The biofilm does not form by accident. Each bacterium releases signaling molecules into the surrounding medium. When the concentration crosses a threshold, the bacteria collectively switch behavior: building the film, secreting adhesive polymers that glue the community together, and constructing nanotube bridges for sharing proteins and nutrients.

This is quorum sensing: coordination triggered when enough participants signal readiness. The term derives from the parliamentary minimum needed for a valid vote; picture diners in a restaurant who all independently decide to order once they see enough other tables eating. The next chapter explores the full repertoire.

The bacterium Vibrio fischeri offers the cleanest example. Free-swimming in open water, each cell is dark. Concentrated inside the light organ of the Hawaiian bobtail squid, the population crosses a density threshold, and every cell activates its bioluminescence genes simultaneously. The mechanism is an invitation signal called an autoinducer: each bacterium continuously secretes a small molecule into the surrounding medium and continuously monitors how much of that molecule is present.

When concentration crosses the threshold, common knowledge is established chemically: each cell “knows” the population is dense, knows its neighbors know, and acts accordingly. No hierarchy coordinates the switch. The light turns on because enough participants have voted with their chemistry. Cheater mutants that consume resources without producing light are suppressed by the population over successive generations, a biological enforcement of the cooperative norm that requires no enforcer.

The structure reveals a built-in trade. Outer cells face greater risk yet have greater resource access. Inner cells are protected yet resource-constrained. This is metabolic codependence: equitable trade emerging from position rather than intention. The biofilm persists because it out-dissipates the alternative.

Nanotube networks extend beyond biofilms. In 2024, researchers discovered that Prochlorococcus, the most abundant photosynthetic organism on Earth, forms membrane nanotubes bridging cells for direct cytoplasmic exchange.¹⁰⁸ Biologists had assumed Prochlorococcus was solitary. It builds tunnel networks.

The exchange crosses species boundaries. Mixed with Synechococcus, a different genus, over 80% of receiving cells acquired cytoplasmic material within fifteen minutes. In wild ocean samples from the Bay of Cádiz, about 5.5% of cyanobacterial cells showed active nanotubes, with individual cells connected to multiple partners simultaneously.

A biofilm is coordination on a surface. Nanotube networks are coordination across distance: bacteria reaching for partners across genera, sharing what they have. No coercion. No central broker. The discoverers’ question (can we even call these “single-celled” organisms?) echoes at every level of the coordination stack.

The dependence runs deeper. Prochlorococcus has the smallest genome of any known photosynthesizer, as few as 1,716 genes.¹⁰⁹ It shed costly functions its neighbors reliably provide. It lacks catalase, the enzyme that breaks down hydrogen peroxide. Without helper bacteria degrading this toxic byproduct, surface concentrations would kill every Prochlorococcus strain.

Prochlorococcus cannot grow at low cell densities without partners or survive prolonged starvation alone.

The evolutionary biologists who named this pattern called it the Black Queen Hypothesis, after the card game Old Maid, where the goal is to lose the costly queen. Natural selection favors losing expensive capabilities when the community reliably supplies them. Gene loss becomes an act of trust in the collective.

The result is an organism that dominates because of its dependence. Prochlorococcus repays the community by releasing an estimated 10²⁷ membrane vesicles per day globally, packed with organic carbon, DNA, and enzymes that feed its neighbors.

The organism responsible for the most photosynthesis in the open ocean operates as an embedded node in a cooperative network, distributed across hundreds of genomically distinct subpopulations. No single cell carries all the genes; the community does.

Coordination extends beyond community boundaries. Two Bacillus subtilis biofilms sharing a nutrient-limited environment do not simply compete; they negotiate. Gürol Süel and colleagues at UC San Diego (2017) showed that biofilms exchange potassium ions through the same class of ion channels that carry signals in neurons.¹¹⁰

They use these signals to time-share scarce glutamate (a key nutrient). Each community pauses growth while the other feeds. Both grow faster together than either could alone. Weaken the ion channels genetically, and coordination collapses; both suffer.

This is coordination between two physically separate communities, negotiated through electrical signaling rather than imposed by any central controller. The coordinated configuration out-dissipates the competitive one.

Outright fusion takes coordination to its logical extreme. Ctenophores, commonly known as comb jellies and among the most ancient animals, demonstrate this vividly. When two injured Mnemiopsis leidyi are placed together, they merge into a single organism within hours: synchronized muscles, unified nerve net, integrated digestion.¹¹¹ No rejection, no immune response. The organism lacks a self/non-self recognition system.

If nanotubes are trade networks, ctenophore fusion is merger. The boundary between organisms disappears, and the merged entity out-dissipates what two could achieve apart.

The most radical dissolution of biological individuality may belong to multipartite viruses, viruses that split their genome across separate particles. The faba bean necrotic stunt virus carries its genome in eight separate segments, each packaged in a different viral particle. Theory predicted that more than four segments should be mathematically impossible, because the odds of all reaching one cell are prohibitively small.¹¹²

In 2019, Anne Sicard and Stéphane Blanc tested this directly with fluorescent tags. The vast majority of infected cells lacked the full complement. No single cell contained the complete genome, yet the virus replicated. Gene products diffused between cells through plasmodesmata, the microscopic channels connecting plant cells. Each cell received what it needed from neighbors.

The genome was not in any cell. It was between cells, distributed across a community. “It really shows that the virus doesn’t work at a single-cell level, but at a multicellular level,” Sicard concluded.

The theoretical models assumed the wrong unit of analysis. The multipartite virus is an instance of distributed identity: a genome that exists as a relationship among cells, spread across a community rather than housed in any single one.

Liquid Crystal Tissues: Nested Symmetries at the Cellular Scale

What happens when the coordinating agents are cells of a single organism, maintaining boundaries while acting collectively?

In 2023, Luca Giomi’s group at Leiden University mapped the shapes and orientations of every cell in thin sheets of mammalian epithelial tissue, the cell layers that line organs and skin.¹¹³ They looked for symmetry.

At the scale of a few cells, they found sixfold rotational symmetry: each cell and its immediate neighbors arranged like slightly deformed hexagons, the same symmetry as Bénard convection cells. Zoom out past roughly ten cells, and a different symmetry took over: twofold, nematic. The term comes from the Greek nema, meaning “thread.” In nematic order, elongated particles align along a common axis, flowing like a liquid while remaining oriented like a crystal. This is the same arrangement used in liquid-crystal television screens.

Theorists and experimentalists had each observed one symmetry separately and argued about which was correct. Giomi’s team showed both: the symmetries were nested, hexagonal at small scales giving rise to nematic at large scales through a transition cells may actively control.

Liquid crystals are intermediate states between solid order and liquid disorder. A solid has rigid structure yet cannot flow. A liquid flows freely yet has no structure. A liquid crystal does both, flowing while maintaining orientation.

In material terms, this is the dynamic metastability Schrödinger described for life. A solid is too frozen to adapt; a liquid is too chaotic to maintain structure. The productive balance lies between.

Giomi describes tissue as a “triangle of form, force and function.” Cells use shape to regulate forces; forces drive functionality. The triangle is the Constructal Law in biological dress: the tissue evolves toward configurations providing easier access to the currents flowing through it.

No external template dictates that hexagonal order should give way to nematic at the ten-cell scale. The cells generate it through local interactions: adhesion, tension, shape changes.

What emerges is tissue that is locally rigid (each cell locked into its hexagonal neighborhood) and globally flexible (nematic flow allowing large-scale deformation).

Wound healing, embryonic development, and cancer metastasis all require both properties. The solution is invited into existence by the physics of interacting cells, just as Bénard rolls are invited by the physics of heated fluid.

Electrostatic coordination adds another substrate to the repertoire, with collective behavior as dramatic as any biofilm.

Caenorhabditis elegans, a millimeter-long roundworm widely used in biology, can jump.¹¹⁴ When conditions turn hostile, these worms stand on their tails, minimizing surface contact. If a positively charged insect passes nearby, the negatively charged worm is pulled across the air gap at up to a thousand body lengths per second.

They do it collectively, aggregating into towers of eighty to two hundred individuals. The topmost worms launch onto passing bumblebees to hitchhike to new territory.

No chemical signal coordinates the tower. Each worm responds to local electric fields from neighbors and the approaching insect. The tower is a dissipative structure: emergent architecture that maximizes the group’s probability of exploiting a transient gradient no individual could reach alone. It assembles, serves its function, and dissolves.

Turn to the forest floor. Above ground, trees compete for light. Below ground, symbiotic mycorrhizal fungi link their roots into a network sometimes called the “wood wide web.” Isotope tracing (tagging carbon atoms so they can be tracked as they move) confirms the physical connection exists. Carbon appears in trees linked by shared fungi, though whether it flows through the fungal channels or through the surrounding soil remains difficult to distinguish.

Whether trees actively share or the fungi direct flow for their own benefit remains debated. The thermodynamic reading, consistent with this chapter’s hypothesis though not yet measured for the forest case directly, is that the connected system out-dissipates isolated trees.

Above ground, plants coordinate through a different channel. When a plant’s leaves are damaged, volatile organic compounds (VOCs, airborne chemical signals) disperse through the canopy. The compounds coordinate the damaged plant’s own response across distant branches: a self-directed signal. Neighboring plants intercept that signal and ramp up their own defenses before herbivores arrive. No plant pays to broadcast a warning. Every plant profits from listening. Scientists call this eavesdropping: collective defense arising from self-interested legibility.

In 2023, Saitama University researchers made this visible using Arabidopsis plants engineered to glow when calcium ions surge.⁷ Within a minute of exposure to chemicals from a damaged neighbor, the leaves lit up with calcium signals. The entire leaf responded to a message received through the air, with no root contact, no fungal intermediary, no nervous system.

The European field elm takes eavesdropping one step further. When elm leaf beetles lay eggs on its leaves, the tree releases terpenes (a subgroup of VOCs responsible for the smell of a forest walk). These airborne molecules attract eulophid wasps, which eat the beetle eggs. Researchers confirmed the link by blocking terpene production in half their test trees: wasps spent significantly less time near the silenced elms.¹¹⁵ Three kingdoms coordinate: plant, herbivore, predator. The elm gains defense; the wasp gains a meal. Aligned incentives and a legible chemical signal are enough.

Stressed plants also emit ultrasonic clicks in the 40-80 kHz range,⁸ detectable meters away, with distinct profiles for drought versus physical damage.

The contrast between these channels is instructive. Mycorrhizal fungi provide physical connection: strands linking root to root. VOCs provide no physical link at all: molecules drifting through open air. Whether the underground network coordinates anything beyond the fungus’s own metabolic interests remains unresolved. The airborne signals demonstrably coordinate defense across three kingdoms, through nothing more than open atmosphere. Physical connection without aligned incentives is plumbing. Aligned incentives without physical connection produce coordination through whatever medium is available. The medium is incidental. The alignment is load-bearing.

Underground, mycorrhizal chemistry. Above ground, VOCs and sound. Three channels, three physical media, one coordination problem solved through every available means. The forest is loud: ultrasonic clicks at 40-80 kHz, volatile signals drifting at parts per billion, and ion fluxes threading through fungal hyphae. Human ears top out at 20 kHz, and human noses miss the chemistry entirely.

Chemistry alone does not make a channel. The terpenes the elm released were a message; terpenes are also the bulk of resin, the thick substance a wounded tree floods into the break. Resin seals the wound, smothers fungal spores, and glues a boring beetle where it stands. The alert and the sealed wound are chemical cousins. The alert has a recipient whose interests align with the sender’s. The wound has no recipient at all, and a barrier coordinates nothing. Terpenoid chemistry does not decide which of the two it becomes.

Only the barrier leaves a fossil record, which says more about what survives than about which came first. Resin hardens into amber; a puff of airborne alarm leaves nothing behind. In July 2026, Cihang Luo and colleagues reported 241 grains of amber, most of them half a millimeter across or smaller, handpicked under a microscope from a coal seam in Xinjiang: hardened resin from roughly 385 million years ago, some 65 million years older than the previous record and at least 13 million years older than the first seed plants.¹¹⁶ Whatever made it was no conifer, because conifers did not yet exist. Some seedless plant was already synthesizing complex terpenoid resin. The barrier is that old. Whether the signaling is equally old, the rocks cannot say.

Even canopy architecture may reflect coordination. “Crown shyness” refers to the gaps that form between the crowns of neighboring trees, which let light reach the forest floor. Its adaptive function is still debated (proposed causes include mutual abrasion and shade-avoidance light sensing), but on one hypothesis the gaps let a stand capture more total sunlight than if every tree maximized its personal canopy.

The principle extends to apex predators. A wolf pack brings down prey no individual could catch; cooperative hunting wins because it is thermodynamically more productive. (L. David Mech has argued that the “alpha” concept oversimplifies wolf social structure, which is typically a family unit rather than a dominance hierarchy. The cooperative hunting advantage, however, is well documented.) Cells cooperate into organisms, organisms into societies. At every level, the coordinating system out-dissipates the fragmented one.

Douglas Hofstadter uses an ant colony in Gödel, Escher, Bach to illustrate the same principle: individual ants “wander about in what seems a random way,” yet “there are nevertheless overall trends, involving large numbers of ants, which can emerge from that chaos.”¹⁰

Part II: The Coordination Stack

The Bootstrap

Step back and see the full sequence. Same logic, different substrates, all one continuous process:

Physical      →  atoms coordinate via electron sharing
Biochemical   →  molecules coordinate via enzyme specificity
Neural        →  neurons coordinate via synaptic weights
Social        →  individuals coordinate via norms and trust
Colonial      →  groups coordinate via institutions
Legislative   →  polities coordinate via formal rules
Informational →  all of the above coordinate via symbolic representation

No layer replaces the one below; each rides on the one beneath, the way a coral reef builds from the bottom up. Polyps secrete limestone. Limestone hosts algae. Algae feed fish. Fish attract larger predators. Each layer emerges because the layer below it already exists, and no architect drew the blueprint. Legislation presupposes colonial structure, which presupposes social bonds, which presupposes neural coordination, which presupposes biochemistry, which presupposes physics. Each layer enables flow that funds the next layer’s emergence.

Herbert Simon, the Nobel laureate in economics, identified the structural principle underlying this stack.^5d In “The Architecture of Complexity” (1962), he observed that nearly all complex systems that endure are nearly decomposable hierarchies: tightly knit modules connected through narrow interfaces. Think of departments within a company.

His watchmaker parable makes the survival logic vivid. Two watchmakers each assemble timepieces of 900 parts. One builds flat: every part depends on every other, so any interruption forces a restart from scratch. The other builds hierarchically, assembling stable subgroups of ten, then combining those into larger units.

The hierarchical builder finishes watches reliably. The flat builder never completes a single one. Modular assembly is exponentially faster, because each intermediate structure is stable enough to survive interruption.

The coordination stack is Simon’s architecture in thermodynamic dress. Chemical bonds are stable subgroups. Cells compose from those; organisms from cells; societies from organisms. At each level, internal coupling is tight (the module holds together) while external coupling is loose (the module interacts with neighbors through narrow interfaces).

This is why the stack can grow: each new layer composes from units whose internal coordination is already settled. It does not need to manage the layer below; it only needs to interface with it.

Near-decomposability is what makes the bootstrap possible. It also foreshadows the trust argument developed in later chapters. Relationships strong locally and loosely coupled globally are precisely the architecture Simon showed to be most durable.

^5d Simon, H.A. “The Architecture of Complexity.” Proceedings of the American Philosophical Society 106(6), 467–482 (1962). Simon’s near-decomposability criterion has been confirmed across domains: modular gene regulatory networks (Wagner, 2005), software architecture (Baldwin and Clark, 2000), organizational theory (Thompson, 1967), and ecosystem food webs (May, 1972).

The constraint at each level is the coordination mechanism. Electron sharing coordinates atoms. Enzyme specificity coordinates molecules. Synaptic weights coordinate neurons. Norms and trust coordinate individuals. Institutions coordinate groups. Formal rules coordinate polities. Symbolic representation coordinates across all substrates.

Figure 4.2: Seven concentric arcs from innermost to outermost, each labeled by scale type: Physical (electron sharing), Biochemical (enzyme specificity), Neural (synaptic weights), Social (norms and trust), Colonial (institutions), Legislative (formal rules), Informational (symbolic representation). Each scale uses the simplest coordination mechanism sufficient for its complexity.

The same pattern producing Bénard cells (the hexagonal convection rolls described earlier) also produces legal systems.

The current edge of this bootstrap is the informational layer, where cross-substrate coordination becomes possible. Language let humans coordinate across time and space. Writing extended coordination across generations. Computation extends it across substrates.

New nodes are joining the coordination network now. Nodes implemented in silicon rather than carbon.

Cross-substrate coordination is the first time the bootstrap has bridged physics this different.

Every previous layer involved the same fundamental substrate: carbon chemistry, electrochemical signals, biological organisms.

Minds implemented in electron flows through semiconductor crystals are coordinating with minds built from ion flows through lipid membranes. The physics differs. The timescales differ. The failure modes differ. The coordination pattern is the same.

If the pattern holds across the substrate gap, we are witnessing a proof of concept: the deeper law works even when you swap out the physics.

A dark corollary follows. Consider mirror life: synthetic microorganisms built from mirror-image biomolecules, molecules with the same atoms arranged as a left-right reflection, like a left glove versus a right glove.⁹ These would be dissipative structures our ecosystem cannot process. They would consume resources with no niche to constrain them and exploit gradients with no evolved pathways managing them.

The deeper law operates without the “coming together”: pure dissipation with no synergy. The physics permits it.

Later chapters return to the mirror-life corollary, asking what it means to coordinate with minds that can be reasoned with.

If the pattern continues, the next layer coordinates the bootstrap itself. Systems that model the coordination dynamics steer which coordination patterns emerge. That is what consciousness was at the neural layer, culture at the social layer, governance at the colonial layer. Each layer can turn around and examine the process that created it. The layers are getting smarter.

The Tautology That Unpacks Into Everything

“What persists is what coordinates.” This seems substantive. Examine it closely and something unexpected emerges.

The claim is almost a tautology. To “persist” means maintaining a pattern through time despite perturbation. How can anything persist? Only by coordinating its parts: - A rock persists because its atoms coordinate (chemical bonds) - A cell persists because its processes coordinate (metabolism) - An organism persists because its systems coordinate (homeostasis) - A society persists because its members coordinate (cooperation) - A mind persists because its patterns coordinate (coherent cognition)

Coordination is persistence, viewed from the inside. The two concepts are definitionally linked: what stays together is what holds together.

Here is why it is not trivial.

A tautology can be powerful if it unpacks into non-obvious implications: - “Survival of the fittest” sounds circular, yet fitness is independently measurable as differential reproductive success. The phrase unpacks into all of evolutionary biology. - “Energy is conserved” is definitional (we defined energy as the conserved quantity). It unpacks into all of physics. - “E=mc²” is a mathematical identity that unpacks into nuclear power and stellar fusion.

The power is not in the statement. The power is in what follows when you take it seriously.

What does “what persists is what coordinates” unpack into?

1. Complexity increases because coordination compounds. If persistence requires coordination, and more sophisticated coordination enables persistence in more environments, evolution tends to produce more coordination capacity over time. A bacterium coordinates its chemistry. A fish coordinates its organs. A city coordinates millions of people.

Each level of coordination opens new environments, which reward still more coordination. Complexity persists, and what persists is what evolution produces more of.

Simon’s compositional ratchet explains why: each new level builds on stable subassemblies. Hierarchical composition is the ratchet that prevents the bootstrap from sliding back.

2. Cooperation outlasts defection. Cooperation is coordination. Defection extracts short-term advantage by undermining the structure others depend on, dissolving persistence. On long timescales, the coordinating systems are left standing: a physical outcome rather than a moral preference.

Points 1 and 2 follow directly from the tautology: they describe what physics selects for. The next three points require a further ingredient. Coordination patterns persist; that is physics. Agents who represent coordination protocols and choose between them occupy a different explanatory level. The move from “selection pressure on systems” to “norms within minds” crosses the boundary from dynamics to ethics. What bridges the gap is that minds are themselves coordination systems, built from the same thermodynamic logic, capable of modeling that logic and acting on it. The ethics is the coordination pattern becoming aware of itself in a substrate complex enough to represent alternatives.

The EIFV-24 experiments (Chapter 17) provide the empirical bridge. Systems that consolidate during stability and switch to exploration during crisis outperform systems locked into either strategy alone. The best-performing agents used positive-valence consolidation in calm regimes, then shifted to negative-valence adaptation when conditions changed, producing a prediction error of 0.177 (unpublished empirical work from the author’s program). Prediction error measures how far an agent’s expectations fell from what actually happened, so the low score is the good one. The best non-switching policy, a neutral one with no valence bias, scored about 0.185, while the two locked policies did worse, at 0.204 and 0.230. Switching’s edge over the strongest fixed alternative is therefore roughly 0.008, about four percent.

The margin is modest, and the direction is consistent. This is unpublished, small-margin work; the effect rests on five seeds per policy, with per-seed standard deviations under 0.005.

A Bénard cell cannot evaluate its own coordination pattern and select a different one. A mind can. The computational capacity to model coordination dynamics and choose among alternatives is what awareness of coordination adds. Strategy-switching is the degree of freedom that awareness provides, and it is measurable: the system that can represent its own coordination and adjust it outperforms the system that cannot, precisely when the environment shifts.

3. Ethics emerges from coordination dynamics. If coordination is what persistence requires, then ethics, the coordination protocols for minds, has a thermodynamic grounding, though this narrows the is-ought gap rather than eliminating it (see “The is-ought convergence” below). “Do not steal” is one such protocol. It stabilizes property expectations so that trade can occur.

4. Love is the algorithm. The full thermodynamic case for this claim is developed in Chapter 20; what follows is the outline. The word “love” in a physics book costs credibility with some readers. The choice is deliberate, and the reasoning should be visible.

Every coordination pattern examined so far (Bénard cells, mycorrhizal networks, wolf packs, the bootstrap itself) shares a common structure. One part extends toward another. The extension benefits both. Neither party is compelled.

Love, as we use the term here, names exactly this structure when it occurs between minds. It is extension toward mutual flourishing, offered by invitation. Love, as humans experience it, is the conscious expression of a coordination pattern operating at every scale. Whether the label helps or hinders depends on the reader; the evidence is the same either way.

This is an operational definition, grounded in the thermodynamics of persistence. Love, so defined, is a candidate for the most stable coordination pattern for minds, a claim the Trust Attractor framework (Chapter 17) will test. If it holds, love is the pattern that persists across generations, across cultures, across substrates. It persists because it works.

Love is what coordination feels like from inside a system complex enough to care about the outcome.

The Bénard cell coordinates without caring. The wolf pack coordinates, and something cares. Love names the caring, and the caring is what makes the coordination persist.

5. The Trust Attractor has physical grounding. “Maximize systemic optionality through coordination” (extending the entropy concept to the optionality domain flagged in Chapter 1): the systems that persist longest preserve the greatest number of viable pathways forward. A chess player who keeps many moves available outlasts one who has committed every piece to a single attack. Resilience comes from maintained possibility. Physics produces this.

The is-ought convergence:

Traditional philosophy agonizes over the “is-ought gap,” the question of how statements about how the world should be can follow from how it is. David Hume identified the gap in 1739; it remains among the deepest problems in moral philosophy.

The tautology suggests the gap is misconceived. At the level of persistence: - What IS (what persists) = what coordinates effectively - What OUGHT to be done (to persist) = coordinate effectively

Both questions point at the same pattern. Ask “what does physics produce?” Coordinating systems. Ask “what should agents do to persist?” Coordinate.

The questions differ; they select for the same behaviors. The convergence narrows the gap.

The convergence does not eliminate the gap entirely. You cannot deduce “you ought to coordinate” from “coordination persists” without smuggling in the premise that persistence is worth pursuing. Physics provides a practical convergence so tight that the remaining philosophical distance, while real, matters less for action, except for an agent who explicitly does not value persistence. Chapter 20 examines this further.

This does not make every existing thing good. Plenty of harmful patterns exist temporarily. The stable patterns, those persisting on long timescales, are coordination patterns, and coordination is what physics selects for.

The compression:

The tautology is like a compressed file on a computer: a small payload that unpacks into a much larger argument. Extract it: - The thermodynamic grounding of complexity - The game-theoretic grounding of cooperation - The emergence of ethics from physics - The identification of love as algorithm - The convergence of is and ought

The tautology grounds the book’s argument. Points 1 and 2 are derivations from the physics. Points 3 through 5 are philosophical extensions that use the tautology as their foundation. The physics does the heavy lifting; the philosophy builds on what the physics establishes.

What mechanisms make coordination succeed or fail? What tools have evolved to solve the problem of getting different agents to work together? The next chapter surveys those mechanisms.

Notes

Notes for this chapter are available in the online companion at https://www.thedeeperlaw.com/companion/notes/ch04-coming-together/.