A dissipative system takes in energy, does work, and gives back disorder. Cells, storms, teams, change programs, neighbourhoods, markets, planets. They look unrelated, until you see the structure they share.
This is what we have found.
The term dissipative comes from Ilya Prigogine. Working at the Free University of Brussels through the 1960s, he found that systems far from thermal equilibrium (systems taking in energy and giving back disorder) could spontaneously organise into stable structures that hold their shape only while the flow continues. He called them dissipative structures. A candle flame. A cell. A storm. A standing wave in a river. He won the Nobel Prize in Chemistry in 1977 for this work.
Prigogine's point was that order does not always come from equilibrium. Order can also come from continuous through-flow, when the flow is large enough to push the system into self-organisation. After him, the idea was carried into biology by Stuart Kauffman (self-organisation in living systems) and Humberto Maturana and Francisco Varela (autopoiesis, the self-producing cell), and synthesised across domains by Erich Jantsch in The Self-Organizing Universe. We work in that tradition.
Prigogine showed that dissipative structures exist and named what they have in common. What we have found, working on top of his foundation, is that they share more than a property. They share a shape.
When we looked carefully at dissipative systems in very different substrates (cells under selection, immune cycles responding to a tumour, organisations going through change, populations under environmental pressure), we found the same shape underneath each one. A cycle. Four regimes. Same order, every time.
The substrate changes the instrument we use to read it. The shape stays.
The unit we work with is a closed cycle that produces the ground for the next cycle. We call it a Change Unit: one full turn through four regimes that ends by depositing what the next turn needs to begin.
Each Change Unit turns through its four regimes once, and the output of the last regime is the substrate the next Change Unit needs to begin. Conservation of one cycle becomes the Potentiality of the next. The closure is the seed. The fractal is not infinite repetition of the same. It is stacked inheritance. Each turn deposits something for the next turn to stand on.
This is the structural building block of dissipative systems the way the golden ratio is the building block of growth. Once you see it, you see it everywhere: cells, organisations, immune responses, ecosystems, markets. Unlike the golden ratio, the Change Unit is not a ratio. It is a unit.
One Change Unit. Sixteen positions in a ring, four per regime. Potentiality blue (top-left), Construction gold (top-right), Encounter red (bottom-right), Conservation russet (bottom-left). The next layer of structure inside each regime type.
The four regimes are four different kinds of dynamical system. Each runs its own equations, has its own observables, and fails in its own characteristic way. They are not symmetric, they are structurally specialised. The names below are working names, derived from what each regime actually does. The animation on each tile shows the regime's character.
The order is forced. The output of the capacitor is the input the factory needs. The factory's shipped product is what the reactor consumes. The reactor's yield is what the homeostat conserves. The homeostat's drift is what loads the next capacitor. The system cannot skip a regime. Any path that tries to shortcut produces a stall, and the stall has a structural address.
Each of the four regimes contains its own four-regime cycle. The cycle at the cellular scale runs inside a cycle at the tissue scale, which runs inside a cycle at the organ scale, which runs inside a cycle at the organism scale. Zoom into any regime and you find the same four regimes again, smaller. Zoom further. Same shape. As far down as you care to look.
A golden spiral with four full turns. Each turn contracts by a factor of 1/φ ≈ 0.618. Each quarter-turn is one regime. The same shape at every scale, continuing inward toward the centre indefinitely.
This is what lets the framework reach across substrates. A tumour is one cycle. So is the immune response trying to contain it. So is the trial designed to test the therapy. The same shape runs at each scale, and the same questions can be asked at each scale.
The golden ratio φ appears throughout the framework not because it was chosen, but because it is forced. For a structure to look the same at every scale (for the cycle inside a regime to be the same shape as the regime inside its parent), the contraction ratio between nesting levels has to satisfy one equation:
This is the fixed-point equation for self-similar contraction. It has exactly one positive solution, and that solution is 1/φ. No other contraction ratio produces a structure that repeats at every scale. φ is not chosen; it is what closure looks like when written down.
The same number, raised to small powers, sets the cycle's natural rate constants in the closure-derived structural form: 1/φ for the fast transitions, 1/φ² for the slow ones, 1/φ³ for the minimum drain per cycle. These are the rates a cycle would run at if nothing else were happening. Real cycles, in real substrates, deviate, sometimes substantially. The structural form is the reference; the deviations are where the work lives.
There is also a single-number measurement worth running on any cycling system. Fit a damped oscillator to its behaviour, measure its damping rate γ and its period T. The dimensionless product γT, in the closure-derived form, is:
where L is the number of nested levels the cycle integrates across. For one complete four-regime cycle (L = 4), this is approximately 1.924. For an L5 system with an Observer function, approximately 2.41. Values above 2.41 indicate the reading is integrating across a meta-scale boundary. This is the kind of falsifiable surface we look for. Where the measurement does not land near the predicted value, the deviation tells us something specific about how the system departs from the structural form.
Cycles are not free. Every traversal loses mass. Prigogine's point (these systems exist only in conjunction with their environment) has a sharp form in our framework. We can name the rate.
A closed dissipative cycle loses 1/φ³ ≈ 23.6% of its accumulated mass on every traversal. Without continuous external input, the system relaxes toward zero. Open systems sustain themselves by replacing what the cycle drains.
A few other numbers that fall out of the same structure:
Structural form and empirical form. The numbers above are the closure-derived structural form: what a dissipative cycle would look like under closure alone, unconstrained by its substrate. Real cycles, in real substrates, deviate. The structural form is the reference; the deviation is where the substrate-specific physics shows up, and where the framework earns its keep diagnostically. The shape that survives is four regimes, traversed in order, each one outlasting the one before it. The exact residence ratios depend on what the cycle is made of.
The framework is useful for one reason. When a dissipative system stalls (a tumour that won't resolve, a trial that won't read, a team that won't renew, a change program that won't land), the stall has a structural location. It happens at one of the four regimes, or at one of the singularities between them.
The same surface symptom can come from stalls at structurally different places. Treating them as the same problem is a common reason interventions fail. The framework lets us locate where the stall actually sits, and from that location, the field of useful interventions narrows.
For more on how a specific engagement is shaped around this, see the method page.
Our open notebook is on the articles page. Predictions, structural reads, papers, and essays, including where things have not yet worked.
Get in touch.