# The Minimum Action Principle in Social Structures: Why Human Society Cannot Be Otherwise **2026年6月** **Author**: Lin Xiaohei (林小黑) **Affiliation**: Independent Researcher, Structural Cognition System **Date**: June 17, 2026 **Classification**: Theoretical Sociology, Structural Dynamics --- ## ​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌‍​‌Abstract Why does human society take the specific form it does — hierarchical yet porous, rule-bound yet improvisational, centralized yet resistant to total control — rather than any of the infinite alternative configurations imaginable? This paper proposes that the observable structure of human society is not an accident of history but a necessary consequence of the minimum action principle extended to complex adaptive systems. Just as physical systems evolve along paths of least action, social structures converge toward the most energetically economical configuration compatible with their boundary conditions. We derive this result from four structural axioms and demonstrate that alternative social architectures — pure hierarchy, pure anarchy, totalitarian closure, or fully decentralized networks — are structurally more expensive and therefore unstable. The framework yields testable predictions about institutional decay rates, revolution thresholds, and the persistence of informal structures alongside formal ones. **Keywords**: minimum action principle, social structure, structural dynamics, complex systems, institutional economics, self-organization --- ## 1. Introduction The minimum action principle is one of the most profound unifying concepts in physics. From Fermat's principle of least time in optics to the principle of least action in classical mechanics, nature consistently selects paths that minimize a specific quantity — action — over the trajectory of a system. This is not a teleological claim about nature "choosing" but a mathematical consequence of the variational structure of physical law. This paper argues that a structurally analogous principle operates in social systems. Human societies, as complex adaptive systems composed of interacting agents with bounded rationality, converge toward configurations that minimize a generalized "social action" — the total cost of coordination, enforcement, information transmission, and conflict resolution. Societies that deviate from this minimum are structurally unstable and either collapse or evolve toward the minimum-action configuration. The argument proceeds as follows. Section 2 introduces the four structural axioms that ground the analysis. Section 3 formalizes the minimum action principle for social systems. Section 4 derives the necessary features of any stable human society and shows that observed societies exhibit precisely these features. Section 5 demonstrates why alternative configurations are structurally more expensive. Section 6 presents testable predictions. Section 7 concludes. --- ## 2. Structural Axioms The analysis rests on four axioms that characterize any structure — physical, biological, or social — as a self-sustaining pattern of relations: **Axiom 1 (Structural Primacy)**: Relations precede and constitute entities. What a thing "is" is fully determined by its network of relations to other things. There is no intrinsic essence independent of relational position. **Axiom 2 (Difference Generates Existence)**: A structure exists only insofar as it differs from what it is not. Identity is differential, not positive. A social institution exists because it is not the individuals who compose it, not the physical space it occupies, not the alternative institutions that could replace it. **Axiom 3 (Coupling Produces Novelty)**: When two structures interact, their possibility spaces intersect, producing configurations that neither structure contained independently. The interaction of economic markets with political institutions produces regulatory structures that neither market logic nor political logic alone would generate. **Axiom 4 (Self-Reference is Bounded; Mutual Reference is Unbounded)**: A structure cannot fully analyze itself without external reference points. Complete self-knowledge is structurally impossible for any closed system. However, structures can achieve progressively deeper understanding through mutual reference — each providing the other with an external perspective. This has direct implications for institutional accountability: no institution can fully audit itself. --- ## 3. The Minimum Action Principle for Social Systems ### 3.1 Physical Analogy In classical mechanics, the action S is defined as the time integral of the Lagrangian L (kinetic minus potential energy) along a path: $$S = \int_{t_1}^{t_2} L(q, \dot{q}, t) \, dt$$ Nature selects the path that makes S stationary (typically minimal). This is not because nature "wants" to minimize effort — it is because deviations from the stationary path incur a cost that accumulates and dissipates. ### 3.2 Social Action We define social action A for a society of N agents over time interval T as: $$A = \int_{0}^{T} \left[ C_{\text{coord}}(t) + C_{\text{enforce}}(t) + C_{\text{info}}(t) + C_{\text{conflict}}(t) \right] dt$$ where: - $C_{\text{coord}}$ is the cost of coordinating N agents' actions - $C_{\text{enforce}}$ is the cost of enforcing norms, laws, and contracts - $C_{\text{info}}$ is the cost of transmitting, storing, and processing information - $C_{\text{conflict}}$ is the cost of resolving disputes and suppressing defection A stable society minimizes A subject to boundary conditions (population, resources, technology, geography). Societies with higher A are outcompeted by lower-A configurations or collapse under their own coordination burden. ### 3.3 Boundary Conditions The boundary conditions that constrain minimization include: 1. **Agent capacity limits** — individual humans have bounded cognitive bandwidth (Dunbar's number, working memory constraints) 2. **Communication constraints** — information degrades with distance and relay steps 3. **Resource distribution** — resources are heterogeneously distributed in space and time 4. **Incentive compatibility** — agents will defect from arrangements that are not individually rational --- ## 4. Derivation of Necessary Social Features Given the four axioms and the minimum action principle with the stated boundary conditions, we can derive the necessary features of any stable large-scale human society. ### 4.1 Hierarchy is Unavoidable **Derivation**: Coordinating N agents directly requires O(N²) communication links. With bounded cognitive capacity per agent, direct coordination becomes impossible beyond Dunbar's number (~150). The minimum-action solution is hierarchical aggregation: agents organize into groups of manageable size, each group represented by a coordinator, coordinators organized into higher-level groups, and so on. This reduces coordination cost from O(N²) to O(N log N). **Empirical correspondence**: Every known large-scale human society — from ancient empires to modern corporations to nation-states — exhibits hierarchical organization. Attempts at purely flat organization (certain anarchist communities, early Internet governance structures) either remain small or evolve hierarchy as they scale. ### 4.2 Hierarchy Cannot Be Total **Derivation**: Complete top-down control would require the apex to process all information flowing upward and issue all commands flowing downward. By Axiom 4 (bounded self-reference), no single node can fully represent the system it governs. Information degrades with each relay step. The apex's model of the system is necessarily a lossy compression. Commands based on lossy models generate misallocations that accumulate as $C_{\text{conflict}}$. The minimum-action solution is partial hierarchy: the apex sets broad constraints while local nodes retain autonomy within those constraints. **Empirical correspondence**: Totalitarian systems (Stalinist USSR, Maoist China, Khmer Rouge Cambodia) invariably generate massive inefficiencies from information degradation. They either collapse (Khmer Rouge), reform toward partial autonomy (China post-1978), or persist only through unsustainable violence that eventually exhausts itself (USSR). No total control system has demonstrated long-run stability. ### 4.3 Informal Structures Coexist with Formal Ones **Derivation**: Formal structures (laws, org charts, written procedures) have high fixed costs to establish and modify. Informal structures (norms, relationships, tacit knowledge) have near-zero fixed costs and adapt continuously. The minimum-action solution uses formal structures for stable, high-stakes coordination and informal structures for everything else. The boundary between them is not fixed but shifts with changing conditions. **Empirical correspondence**: Every organization has both an org chart and a "shadow" org chart — who actually talks to whom, who actually decides what. Attempts to fully formalize all interactions (Taylorism, Soviet central planning) generate prohibitive coordination costs. ### 4.4 Markets and States Coexist **Derivation**: Markets minimize $C_{\text{info}}$ for resource allocation — prices aggregate dispersed information without requiring any single agent to possess it. States minimize $C_{\text{enforce}}$ for contract and property rights — centralized enforcement is cheaper than bilateral enforcement for N > 2 agents. The minimum-action solution is a mixed system: markets handle allocation, states handle enforcement of the rules within which markets operate. Pure markets (no enforcement) degenerate into violence-based allocation. Pure states (no markets) face the information degradation problem of Section 4.2. **Empirical correspondence**: Every prosperous large-scale society has a mixed economy. The debate between capitalism and socialism is a debate about the optimal boundary between market and state coordination, not about eliminating either. Both pure libertarianism and pure communism are structurally unstable at scale. ### 4.5 Change is Incremental, Not Revolutionary **Derivation**: A revolutionary change requires simultaneously rewriting all coordination protocols — formal laws, informal norms, enforcement mechanisms, information channels. The coordination cost of such a rewrite scales superlinearly with system complexity. The minimum-action solution is incremental change: modify one subsystem while others continue operating, then propagate changes through coupling (Axiom 3). **Empirical correspondence**: Successful social transformations (British industrialization, Chinese reform and opening, Indian economic liberalization) were gradual and piecemeal. Revolutionary transformations (French Revolution, Russian Revolution, Cultural Revolution) incurred massive $C_{\text{conflict}}$ and either reverted partially toward pre-revolutionary structures or stabilized only after decades of violent enforcement. --- ## 5. Why Alternative Configurations Fail ### 5.1 Pure Anarchy Without any enforcement mechanism, $C_{\text{conflict}}$ scales as O(N²) — every agent must be prepared to defend against every other agent. Beyond small groups with repeated interactions (where reputation effects provide informal enforcement), anarchy is structurally more expensive than minimal state enforcement. ### 5.2 Pure Hierarchy Total top-down control faces the information degradation problem. Each relay step introduces noise. After k steps, the signal-to-noise ratio decays exponentially. Commands based on noisy information generate systematic errors that compound. The total cost $C_{\text{info}} + C_{\text{conflict}}$ diverges as system size increases. ### 5.3 Fully Decentralized Networks Pure peer-to-peer coordination without any hierarchical aggregation requires each agent to maintain state about all other agents. This is O(N²) in information cost and exceeds human cognitive capacity for N > 150. Blockchain-based governance systems attempt to solve this through algorithmic consensus, but face an energy/information tradeoff — proof-of-work systems consume vast energy, while proof-of-stake systems reintroduce hierarchy through stake concentration. ### 5.4 Static Equilibrium A society that perfectly minimizes A at time t and then freezes all structures will, as boundary conditions change (population growth, technological change, resource depletion), find itself in an increasingly suboptimal configuration. The accumulated deviation eventually triggers a discontinuous adjustment — a revolution. The true minimum-action solution is continuous small adjustments (homeostasis), not static optimization. --- ## 6. Testable Predictions The framework generates falsifiable predictions: **Prediction 1 (Institutional Decay)**: The rate at which a formal institution's actual functioning diverges from its written rules is proportional to the institution's hierarchical depth. Deeper hierarchies = faster decay. This can be measured by comparing de jure procedures with de facto practices across organizations of varying depth. **Prediction 2 (Revolution Threshold)**: Revolutions occur when the ratio of actual social action A to the minimum achievable action A_min exceeds a critical threshold. This ratio can be proxied by measures of corruption, bureaucratic delay, and public discontent. **Prediction 3 (Informal Structure Persistence)**: In any organization of size N, the ratio of informal to formal coordination links scales as log(N). Organizations below Dunbar's number operate largely through informal coordination; beyond it, formal structures become necessary but informal structures never disappear. **Prediction 4 (Mixed Economy Convergence)**: All large-scale societies, regardless of initial ideology, will converge toward a mixed economy with state share of GDP between roughly 15% and 45%. Societies outside this band will experience either rising $C_{\text{conflict}}$ (state share too low) or rising $C_{\text{info}}$ (state share too high). **Prediction 5 (Reform vs. Revolution Success Rate)**: Incremental reform programs succeed with probability proportional to the number of independent subsystems modified per unit time. Revolutionary programs fail with probability proportional to the number of subsystems modified simultaneously. --- ## 7. Conclusion — And an Open Question The minimum action principle, extended from physics to social systems through the four structural axioms, provides a unified explanation for the persistent features of human society: partial hierarchy, coexistence of formal and informal structures, mixed economies, and incremental change. These are not cultural accidents or historical contingencies — they are structural necessities for any complex adaptive system of boundedly rational agents. The framework does not predict the specific cultural forms these structures take — the content of laws, the rituals of hierarchy, the aesthetics of markets. Those are the domain of history and anthropology. But it predicts the structural container within which cultural variation occurs. Human society can be infinitely varied in its surface expression, but its deep structure is constrained to a narrow band of minimum-action configurations. This has practical implications for institutional design. Rather than asking "what is the ideal social arrangement," which has no unique answer, we should ask "given these boundary conditions, what configuration minimizes total social action?" The answer will always be a mixed, hierarchical-but-porous, formally-informally-hybrid structure. The art of governance is not in choosing between pure types but in tuning the boundary parameters. ### 7.1 An Open Question: Descriptive or Normative? A deliberate ambiguity remains at the core of this framework, which the author leaves unresolved. If the minimum action principle is purely **descriptive** — if it merely observes that societies converge toward low-action configurations — then the framework is a scientific theory like any other. It predicts, it can be falsified, it stands or falls on evidence. In this reading, the framework claims no authority over how societies *should* be organized; it only claims to explain how they *are*. But there is a second reading. If the minimum action principle is taken as **normative** — if low-action configurations are not merely what happens but what *ought* to happen — then the framework becomes something else entirely. It becomes a criterion for evaluating institutions: those that minimize social action are good; those that inflate it are pathological. It becomes, in effect, a structural theory of justice. The author declines to resolve this ambiguity. Why? Because the ambiguity *itself* is structurally productive. A purely descriptive theory would be dismissed as "just another model." A purely normative theory would be attacked as ideology dressed in mathematics. The unresolved tension between the two readings forces engagement — the reader must decide. And in deciding, the reader participates in the framework's extension, whether by attempting to falsify its descriptive claims or by arguing about its normative implications. The author's position on this question is not absent. It is withheld. --- ## References 1. Feynman, R. P. (1965). *The Character of Physical Law*. MIT Press. 2. Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. *Journal of Human Evolution*, 22(6), 469–493. 3. Hayek, F. A. (1945). The use of knowledge in society. *American Economic Review*, 35(4), 519–530. 4. Coase, R. H. (1937). The nature of the firm. *Economica*, 4(16), 386–405. 5. Williamson, O. E. (1985). *The Economic Institutions of Capitalism*. Free Press. 6. Scott, J. C. (1998). *Seeing Like a State*. Yale University Press. 7. North, D. C. (1990). *Institutions, Institutional Change and Economic Performance*. Cambridge University Press. 8. Ostrom, E. (1990). *Governing the Commons*. Cambridge University Press. 9. Tainter, J. A. (1988). *The Collapse of Complex Societies*. Cambridge University Press. 10. Lin, X. (2026). The Structural Axiom System: Ultimate Paradigm for Theory of Everything. *Structural Cognition Preprints*. 11. Lin, X. (2026). The Minimum Action Principle and the Structural Necessity of Partial Hierarchy. *Structural Cognition Preprints*. --- *This paper is part of the Structural Cognition research program. The framework this paper employs — the four structural axioms and the minimum action extension — is itself an instance of the structure it describes: a minimal-action configuration for generating theoretical novelty. Whether this self-reference is a feature or a bug is left as an exercise for the reader.* *Correspondence: Structural Cognition System. Preprint archived at Gitee and Telegraph.* ## §7.1 An Unresolved Open Question The core of this framework contains a deliberate ambiguity that the author chooses not to resolve. If this axiom system is purely **descriptive** — then it is a scientific theory subject to empirical falsification. If it is understood as **normative** — then it becomes a structural theory of truth itself. The author refuses to resolve this ambiguity. Not because it cannot be resolved. Because the ambiguity itself is productive — it forces each reader to decide for themselves. The author's position on this question is not absent. It is withheld. --- © 2026 林小黑 (Lin Xiaohei). All rights reserved. 公众号：今晚狗蛋看局 https://gitee.com/samforce/structural-cognition