Yellow tubes spread between scattered food sources, thickening along some connections and retreating from others. In experiments published in 2010, food placed around a central site represented settlements near Tokyo. The slime mould Physarum polycephalum formed networks that compared favourably with the railway system on measures including transport efficiency and resilience. Its routes emerged through local processes. The network's shape was inseparable from the activity sustaining it: this was an organism feeding and adjusting its connections, with no engineer assigning each route. 1
An economy also has to keep making the conditions of its own continuation. Calling it a living structure gives that requirement a particular force. Growth alone tells us little about whether the system can continue. A business can expand while exhausting a resource, just as a growing network can lose access to what feeds it. We have to distinguish the mechanisms that amplify economic activity from those that keep it viable. They can work together, yet they can also pull in different directions. 2
The physical starting point is a dissipative structure. Ilya Prigogine showed how organised behaviour can arise away from thermodynamic equilibrium, sustained by exchanges with the surroundings. Order need not be a frozen arrangement: it can depend on continuing flows and irreversible processes. This helps us understand why the persistence of an organised system has an operating requirement. It gives no automatic assurance that any particular arrangement is beneficial or durable. 3
Nicholas Georgescu-Roegen's economic argument makes the complementary point. Production is a physical transformation, drawing on resources whose usefulness cannot be described by conservation of energy alone. Equal energy quantities can have different capacities to perform work. His criticism of an economy imagined as a self-contained circulation of money directs attention towards its material conditions. A factory's accounts may close while its physical exchanges continue through the environment. 4 There is no conflict between local organisation and the second law. Entropy cannot decrease in an isolated system. An open system can maintain local organisation through exchanges that leave sufficient entropy elsewhere. Solar energy helps sustain the organisation of the biosphere, while energy ultimately leaves Earth as radiation. This does not undo the entropy produced overall, and it does not guarantee the recovery of every resource we damage. It establishes the physical setting in which maintenance has to operate.
Dissipation by itself is still insufficient for. Stuart Bartlett and Michael Wong propose a broader functional definition, which they call Lyfe, combining dissipation with autocatalysis, homeostasis and learning. Their proposal concerns the possible forms of living systems. We can borrow its questions to examine economic organisation: how activity sustains itself, how it amplifies, how it stays within viable conditions and how experience changes its behaviour. The borrowing is a way to investigate functions, with no implication that society has become one biological individual. 5
Autocatalysis is visible whenever an activity helps produce more of the conditions for its expansion. Profit finances new equipment, which can generate more profit. Experience makes a workshop better at obtaining the next order. A useful network attracts participants who make it more useful. Such processes explain much of an economy's energy and invention. They also explain why a profitable method can grow beyond the capacity of the resource feeding it, especially when that resource is cheap because someone else bears its deterioration.
Homeostasis provides a response to disturbance. In economics, taxes and resource rules can constrain an expanding activity before it destroys its own supporting conditions. Demand supplies some feedback, since unsold output makes production less attractive. It is a poor signal for losses excluded from the sale. Buyer and seller may both gain from a transaction while leaving a damaged water supply downstream. Their agreement has settled the exchange; the wider resource remains impaired.
My proposed corrective signal is the change in natural and social capital. An activity that consumes productive potential should recognise the work required to maintain an equivalent capacity. A method reducing that work would gain an advantage. Increasing output by leaving a larger maintenance burden behind would become less attractive. The aim is to give economic amplification a reference outside its own receipts, so the conditions supporting production enter the decisions that expand it.
That requirement is a deliberate rule for a sustainable economy. Physics does not command people to preserve a particular productive inheritance. It tells us that physical transformations have constraints and consequences. We decide that future users should retain capacities on which our own opportunities depended. Expressing that decision through an account gives it a place in ordinary calculation, where the balance between a private gain and a shared loss can otherwise disappear.
Learning changes the available methods. A workshop can discover how to assemble an object so that worn parts are replaceable. The revised design may take longer initially while saving materials throughout use. Which design prospers depends partly on what the accounts reward. If replacement burdens are absent from the calculation, experience may perfect an easier sale at the expense of harder recovery. Recognising maintenance gives invention a reason to pursue improvements that remain useful across that longer interval. A firm must also retain what it learns. Maintenance records linking a failure to a replacement help the next designer avoid repeating it. If such knowledge disappears when a supplier closes, later enterprises can be forced to pay for the same discovery again. Keeping it accessible is part of maintaining a productive inheritance.
The four functions must also operate at compatible speeds. Investment can expand rapidly while damage becomes apparent slowly. Knowledge of a replacement process may arrive after a resource has disappeared. Early information matters because it preserves room for adjustment. A credible obligation gives firms time to compare methods before equipment and habits harden around the old one. Sudden scarcity may force a response, but it offers fewer choices to the people whose livelihoods depend on the system.
People can also question the purpose of the arrangement. They possess rights and conflicting interests; they cannot be treated as disposable components in somebody else's model of social health. Our institutions must let them understand and contest the obligations imposed. The living-structure approach is useful when it helps us explain a failure of regulation or learning, and devise a response answerable to the people affected. Maintaining shared capabilities includes maintaining that capacity for cooperation and disagreement.
Physarum keeps reshaping the tubes that connect it to food, and a viable economy must likewise keep its growing connections answerable to the resources and learning that sustain them.
References
1. Tero, A., Takagi, S., Saigusa, T., Ito, K., Bebber, D. P., Fricker, M. D., Yumiki, K., Kobayashi, R., & Nakagaki, T. (2010). Rules for biologically inspired adaptive network design. Science, 327(5964), 439–442. https://doi.org/10.1126/science.1177894
2. Lascols, B. (2026). Capital and entropy: Theory of negentropic value (4th ed.). Accounting for externalities.
3. Prigogine, I. (1977, December 8). Time, structure and fluctuations [Nobel lecture]. Nobel Foundation. https://www.nobelprize.org/uploads/2018/06/prigogine-lecture.pdf
4. Georgescu-Roegen, N. (1971). The entropy law and the economic process. Harvard University Press. https://doi.org/10.4159/harvard.9780674281653
5. Bartlett, S., & Wong, M. L. (2020). Defining lyfe in the universe: From three privileged functions to four pillars. Life, 10(4), Article 42. https://doi.org/10.3390/life10040042
