Civilizational Architecture · Working Paper

AeviternaAeviterna

Adaptive Stability, Transitional Governance, and Civilizational Stewardship

Version
v2.15
History
Date
July 2026
Status
Active Draft
Domain
Political Philosophy
Framework
Post-Capitalist Transition
Structure
6 Parts, 3 Appendices
Citation ID
EQ-v2.17-2026-07
Stable: core argument settled
Draft: structure present, content evolving
Open: active debate, not ready to cite

How to read this. The argument runs through the body: the diagnosis (Part I), the foundations (Part II), and the architecture (Part III) carry the load, and the human stake and governance (Parts IV and V) build on them. The front matter and Part VI are the rigor scaffolding, the methodology, falsifiers, and critique that let the claims be tested rather than trusted; read them second unless you came to attack. The practical path lives outside this document, on the on-ramp and in the starter kit.

Table of Contents
FM–01 · Abstract

Abstract

Stable

Contemporary political frameworks presuppose one of two civilizational modes: static equilibrium, in which the goal of governance is to hold systems in balance; or revolutionary rupture, in which existing systems are replaced through force or collapse. Neither framework adequately accounts for a third mode, adaptive transition under conditions of active cognitive capture. This paper introduces Aeviterna as a post-capitalist civilizational framework designed for that third mode. Aeviterna proposes that healthy civilizations do not seek perfect balance. They oscillate within survivable bounds while preserving the capacity to detect instability, correct proportionally, and redistribute recovery. The framework rests on five structural commitments: adaptive stability, replacing growth-as-imperative with bounded self-correction; cognitive sovereignty, treating perception itself as a contested civilizational resource requiring protection; cognitive ecology, the governance requirement to maintain productive tension between coherence and generative ambiguity; the justice of light, establishing that a civilization's response to harm must be as legible and corrigible as its governance structures; and the knowledge commons, reframing shared understanding as primary infrastructure rather than commodity. Aeviterna is presented not as a final design but as a versioned, falsifiable, and iteratively refineable architecture. Its claims now travel with the apparatus to test and adopt them. A register of sixteen dated falsifiers (F–01 through F–16) states what would refute it, a balance sheet prices what the mechanisms cost to run, and an on-ramp lets a group enter in a single meeting. Live instruments read the vital signs, track real adopters with their decline coded as data, and grade the workshop's own dated forecasts.

FM–02 · Research Problem Statement

Research Problem Statement

Stable

The central problem Aeviterna addresses is that existing political and economic frameworks were designed for conditions that no longer hold. They assume bounded information environments, legible power, and the possibility of stable equilibrium. None of these assumptions survive contact with the present.

Capitalism assumes that market signals are sufficient to coordinate behavior and allocate resources. It does not account for the weaponization of information as a market instrument, the systematic capture of regulatory bodies, or the externalization of civilizational costs onto commons that markets cannot price. The result is not market failure in the technical sense. It is the successful optimization of a system against its own stated purpose.

Socialism, in its historical implementations, assumes that collective ownership resolves the coordination problem. It does not account for the capture of collective institutions by administrative hierarchies, the suppression of feedback mechanisms necessary for adaptive correction, or the cognitive homogenization that results from centralized epistemic control.

Both frameworks share a fatal structural assumption, that the system, once correctly designed, will tend toward a desirable stable state. Aeviterna rejects this assumption entirely. Living systems do not stabilize. They oscillate. The question is not how to achieve stability but how to keep oscillation within survivable bounds while the conditions of the world continue to change.

A second problem compounds the first. The emergence of AI-mediated information environments has created conditions under which perception itself, the raw material of political judgment, can be shaped at scale, adaptively, and without the knowledge of those being shaped. This is not merely a problem of misinformation. It is a structural problem of cognitive capture, the systematic redirection of attention, emotion, and belief in service of concentrated interests. At scale, this process constitutes cognitive terraforming, the large-scale reshaping of the perceptual and epistemic conditions under which populations think, judge, and act. No existing political framework was designed to address this. Aeviterna is.

FM–03 · Literature Positioning

Literature Positioning

Stable

Aeviterna draws on and departs from several intellectual traditions. Locating those departures is necessary for rigorous engagement.

From complexity theory and systems thinking (Meadows, Sterman, Kauffman), Aeviterna inherits the core insight that living systems require feedback loops to remain viable, and that the suppression of feedback through opacity, capture, or design is the primary mechanism of systemic collapse. The departure is that complexity theory tends toward description. Aeviterna is normative. It specifies which feedback structures a civilization ought to preserve and why.

From commons theory (Ostrom, Hess), Aeviterna inherits the empirical finding that shared resources can be governed sustainably without either privatization or centralized state control, provided that governance structures meet specific design conditions. The departure is that Ostrom's work focuses on physical and digital commons. Aeviterna extends the commons framework to epistemic resources, meaning knowledge, attention, and the infrastructure of perception.

From critical theory and discourse ethics (Habermas), Aeviterna inherits the claim that legitimate governance requires communicative rationality. Decisions affecting people must be grounded in reasoning those people can access and contest. The departure is that Habermas assumes a relatively undistorted public sphere is achievable through institutional design. Aeviterna treats distortion of the public sphere as the default condition under cognitive capture. It requires active structural defense rather than procedural correction.

From cybernetics (Wiener, Beer), Aeviterna inherits the concept of adaptive control, meaning the capacity of systems to adjust their behavior in response to feedback from the environment. Cybernetic governance models have historically tended toward technocratic centralization. Aeviterna insists that adaptive control must be distributed, auditable, and corrigible.

From the attention and surveillance economy (Simon, Wu, Zuboff), Aeviterna inherits the account of how attention became a scarce resource to be captured and how behavioral prediction became an industry. The departure is that this tradition largely diagnoses an economic harm to consumers. Aeviterna treats the same process as a civilizational harm to the epistemic commons, one that requires structural defense of cognition rather than market correction alone.

From degrowth and steady-state economics (Daly, Kallis, Hickel), Aeviterna inherits the principle of sufficiency and the empirical literature showing that growth has not been decoupled from ecological damage at sufficient scale. The departure is that degrowth largely prescribes a lower throughput level, whereas Aeviterna prescribes bounded oscillation rather than a fixed equilibrium, and supplies the governance architecture that degrowth tends to leave underspecified.

From democratic confederalism and social ecology (Bookchin, Öcalan), Aeviterna inherits confederated, scale-nested decision-making and the use of mandated, recallable delegates, which operationalize accountable and revocable power. The departure is that confederalism relocates sovereignty to the assembly, whereas Aeviterna dissolves sovereignty into domain-specific, decaying stewardship, and treats anti-capture as explicit design rather than a property assumed to emerge from face-to-face democracy, a property the Rojava record shows is not automatic.

From the capability approach (Sen, Nussbaum), Aeviterna inherits a non-utilitarian account of flourishing as the real freedom to achieve valued functionings, which grounds the non-instrumental activity rights of EQ–03b and links cognitive sovereignty to the capacities for imagination and practical reason. The departure is that the capability approach is methodologically individualist, whereas Aeviterna extends flourishing to collective, intergenerational, and ecological subjects, and answers the Sen and Nussbaum dispute over a fixed list with a commons-curated, revisable core.

Open questions
  • The traditions above are positioned. Residual debates: green-growth decoupling, one-party drift in confederalist practice, and the Sen and Nussbaum dispute over a fixed list of capabilities.
FM–04 · Methodology and Falsifiability

Methodology and Falsifiability

Stable

Aeviterna is a normative framework, not an empirical theory. It makes claims about what civilizational systems ought to do, grounded in claims about what they need to do to remain viable. This distinction matters for how the framework should be evaluated.

The ledger has two columns, and honesty starts with keeping them apart. The value axioms, the inviolable dignity floor and non-domination, are normative commitments and are not up for empirical refutation. The load-bearing causal claims, P–01 and P–02 below and the slogan that compresses them, are wagers about how systems behave, and they inherit empirical answerability. The framework does not get to bank the rhetorical force of a causal claim while claiming a normative exemption from its test.

The primary method is structural argument, identifying the conditions under which specific failure modes occur, and deriving design principles that address those conditions. Each section of this document makes a structural claim that can be contested on its own terms.

Aeviterna is falsifiable in the following senses:

F–01 If systems designed with Aeviternal feedback structures (rotational oversight, distributed memory, decay mechanisms) do not demonstrably outperform comparable systems without them on measures of resilience and recoverability, the framework's core claims about adaptive stability are refuted, not merely weakened. The adoption tracker is this falsifier's measurement instrument in embryo. It defines the variables, anchors, and recording discipline the test needs, and it is not yet the test.
F–02 If epistemic commons can be shown to function sustainably without the structural protections Aeviterna specifies, the framework's claims about cognitive capture are refuted.
F–03 If standing decay mechanisms (EQ–05) produce coordination failures worse than the stewardship capture they are designed to prevent, that section must be revised or replaced.
F–04 If domain-scoped decaying standing cannot be calibrated to outperform a permanent or convertible reputation scalar at preventing the concentration of cross-domain influence while holding coordination at least level, with no tunable band that avoids both ossification when decay is too slow and the loss of the expertise floor to unelected permanents when decay is too fast, the standing engine of EQ–05 is withdrawn rather than patched.
F–05 Candidate, in-silico only. In the harness's adoption model, prefigured capacity and momentum jointly gate the transition cascade, and removing either wastes the legitimacy window (EQ–04d). If field trajectories of adopting polities show no such gating, the transition engine is wrong.
F–06 Candidate, observational. If a polity or institution running the EQ–06 anti-capture layer traverses three consecutive stages of the capture cascade without any watching mechanism firing, the cascade's detection claim fails as specified. If documented capture campaigns show no stage-ordering tendency at all, the cascade model is withdrawn in favor of the static threat table it replaced.
F–07 Candidate, observational. If the commonwealth trust's mandate breadth or its discretionary share of spending grows across two consecutive scheduled reviews without a forced contest firing, the anti-regrowth claim of EQ–02e fails and the organ is wrong as designed. If no rent-funded provisioning trust anywhere holds its enumerated mandate across a change of governing coalition, the trust is a fund that holds while the floor it finances is bargained away, and the mechanism is withdrawn.
F–08 Candidate, observational. If reading seats bound to pre-registered, dated, leading-indicator forecasts that reality later scores show no better calibration and no less capture than unbound seats across a decade of the tracker, the anchor of EQ–06e buys nothing and its third halt is wrong.
F–09 Pre-registered, self-directed. By July 31, 2028 the workshop that writes this framework must show at least one adoption of a named mechanism outside the maintainer's control, at least one outside critic answered on the record, and a forecast register scoring no worse than chance. Failing all three, the program is degenerating by its own definition (EQ–10b), and new-organ production halts while the corpus is archived as transplantable parts.
F–10 Candidate, observational, three-scale. If running the applicable stack requires more than 2 percent of operating budget or 5 percent of member work-hours at cooperative scale, more than 50 dollars per member per year or 2 percent of member work-hours at federation scale, or more than 1 percent of the public budget at polity scale, the stack is too heavy at that scale and the adoptability claim of EQ–02f fails there. Tracker-coded adopters abandoning named organs over cost within two years fails the design regardless of the ceilings.
F–11 Candidate, observational, two legs. If communities running the full bootstrap ladder, patient acquisition finance plus the veto lock plus a recycling levy, complete attempted acquisitions at rates the tracker cannot distinguish from the ambient funnel base rate, roughly one in twelve to one in fifty by asset class, the ladder buys nothing and EQ–02e's bootstrap claim fails. If ladder-assembled corpora are subsequently lost at rates within an order of magnitude of conventional ownership rather than the stewardship record's advantage, the lock claim fails independently.
F–12 Candidate, observational, two legs. If tracker-coded adopters that meet the compliance-load budget still show governance participation decaying to the ambient voluntary-association floor, single-digit turnout within three years of adoption, the demand mechanisms of EQ–04d fail and the engine's demand term is wrong. If organs whose protected value is visible and personal retain no more participation than organs whose value is abstract, measured within the same adopter, the visibility rule fails independently.
F–13 Candidate, observational, two legs. If tracker-coded adopters that pre-write the borrow, governance encoded as enforceable bylaws or operating agreements with a seated arbitration path, resolve documented internal breaches and survive them at rates indistinguishable from handshake-governed peers, the ladder's fifth rung buys nothing and the bootstrap claim of EQ–06f fails. If, within adopters, organs protected by stakes, locks, or federated sanction show compliance and survival no better than organs held by reputation alone, the ladder above the first rung is decoration and the organ is withdrawn.
F–14 Candidate, two legs, in-silico and observational. In the harness's siege scenario, cells carrying the wall, living internal clearing, a mode-gated reserve, federation treaty lines, and a published days-of-cover, must hold their floor functions under clamps on external capital, clearing, and demand that collapse matched unwalled controls, or the wall is wrong as designed. In the field, if tracker-coded adopters carrying named wall elements survive documented siege episodes at rates indistinguishable from matched unwalled peers, or survive only by breaching the dignity floor or exporting losses to non-members, the economic wall of EQ–04f buys nothing.
F–15 Candidate, observational, demographic. If any member polity's old-age support ratio falls below about one and a half working-age adults per person over sixty-five while the framework still owes both its full carer entitlements and its full governance-participation budget, and the measured informal care hours plus mandated governance hours per working-age member exceed that member's measured discretionary time, the care-and-participation arithmetic of EQ–08f is unfundable in labor-hours and its assumption fails. A polity that sustains both from residents past the threshold, without importing carers or throttling governance, leaves the assumption standing.
F–16 Candidate, observational, two legs. Using the mandated provenance trail, if across a decade a single durable network drafts the question, selects a majority of the experts, or authors the briefing for sixty percent or more of a polity's sortition exercises, and the option sets presented systematically omit the highest-support alternatives an independent panel surfaces, in that network's favor, the definer defenses of EQ–06g buy nothing and the organ is false. A working mechanism predicts no systematic, faction-favoring divergence between the curated menu and an independently elicited one.

Two disciplines keep these from hardening into an infinitely flexible protective belt. The first is a stopping rule. A core proposition that fails its test is withdrawn, not quietly patched into a smaller claim, and the changelog records the withdrawal. The second is naming in advance what would do the withdrawing. For P–01 the standing refuter is the durable extractive regime, since if rigid, concentration-heavy institutions reliably outlast adaptive ones under comparable shocks, P–01 is false (Acemoglu and Robinson). For P–02 it is the uncaptured low-resistance system, since if systems with minimal designed resistance stay uncaptured under pressure over a long window, the universal-tendency claim is false. This ledger is now carried as the movement's standing falsification register (EQ–10b). Every load-bearing claim ships with its would-be refuter, and contests over the register run as pre-agreed adversarial collaborations.

The banner, equilibrium is death, is the motivational form of P–01 rather than a separate empirical boast. Its cashable content is the narrower, testable claim that suppressing a system's internal feedback and adaptive capacity raises its long-run collapse hazard conditional on shock, which the record supports better than the slogan's literal reading, since the longest-lived orders survived by adapting rather than by holding still.

Falsifiability names what would count as failure. The harder question is how to look for it, since the framework operates at civilizational scale where clean experiments are rare. Three methods are proposed, none sufficient alone. The honest state of each is that validated instruments for governance systems do not yet exist, so the framework leans on triangulation rather than any single measure.

thresholdcurrent statealternative regimelatituderesistanceprecariousness
A resilience landscape. The system rests in a basin; latitude is how wide that basin is, resistance how deep, and precariousness how close the current state sits to the threshold beyond which it shifts to another regime. These are the quantities the resilience metrics try to estimate.
Resilience metricsResilience is triangulated rather than captured by one number. Engineering resilience is read as the recovery of function after a shock, the size of the performance dip over time (Bruneau and colleagues). Ecological resilience is read through the latitude, resistance, and precariousness of the system's basin (Holling; Walker and colleagues). Early-warning signals such as rising autocorrelation and variance can flag a loss of resilience before a tipping point (Scheffer and colleagues), with the caveat that they raise false alarms and are not yet validated for politics.
Case studies as partial implementationsNo system is fully Aeviternal, so existing ones are read as partial and instructive. Ostrom's common-pool cases supply the base, Mondragon shows durable democratic ownership and its erosion at the competitive periphery, Taiwan's vTaiwan demonstrates cognitively-sovereign mass deliberation, and Wikipedia shows a knowledge commons that drifts toward oligarchy without deliberate counter-design. These are tenant-grade rather than building-grade evidence. Each runs inside a substrate it did not create, a state, a currency, a defended internet, a market, and erodes at the boundary with that substrate, so they show the primitives working as well-governed tenants, not the whole model working as the building. Each is studied by first asking resilience of what, to what (Carpenter and colleagues).
SimulationAgent-based and system-dynamics models stress-test mechanisms and scenarios rather than forecast outcomes, asking which feedbacks matter and which interventions are robust. No one can simulate a whole civilization, but the interface is simulable. The partial harness stresses the cross-coupled failures that matter most, capture propagation, contagion, and metric drift across coupled units. Models are grounded in case evidence (Janssen and Ostrom) and documented to a shared standard, the ODD protocol, so they can be checked. The deepest limit is equifinality. Many different rules can grow the same pattern, so reproducing an outcome does not prove its cause. A first run is now in hand and runnable. With the mechanisms the coupled cells hold and recover from a correlated crisis while the unprotected control collapses, and the standing-decay rate shows a tunable middle, in the interactive harness.

A protocol keeps the triangulation disciplined.

Name the targetState resilience of what, to what, since the term stays empty until the property at risk and the disturbance are specified (Carpenter and colleagues).
Measure recovery where you canWhere a performance function and a recovery curve are observable, read the loss as the size of the dip and the speed of return (Bruneau and colleagues).
Use proxies, labeled as proxiesWhere direct measurement is impossible, fall back to structural indicators such as modularity, response diversity, and polycentricity, and report them as surrogates rather than results (Elmqvist and Folke; Ostrom).
Distrust the alarmsTreat early-warning signals as a research signal, detrending the series and checking for the false positives that growth and non-stationarity are known to produce (Scheffer and colleagues; Jäger and Füllsack).
Document for replicationRecord any simulation to the ODD standard so others can rerun it (Grimm and colleagues).
Civilizational vital signsEach dial has its own scale and units; the needle is the current reading.in banddriftingbreachedEcological throughput01497/9BREACHEDboundaries transgressedMaterial security03%12%30%10.5%DRIFTINGin extreme povertySocial trust025%50%100%30%DRIFTINGtrust most peopleEpistemic diversity0557010054.7BREACHEDpress-freedom scoreAttention sovereignty01h2.5h4h2h21DRIFTINGper day on socialLegitimacy040%60%100%26%BREACHEDlive in democraciesDisagreement capacity040%60%100%23%BREACHEDexpression is openBiological resilience0406010039BREACHEDpreparedness indexIllustrative July 2026 readings; heterogeneous proxies, read as bands not a single score.
The civilizational vital signs, one instrument the methodology points toward. Each variable is held inside a survivable band rather than driven to a target, and each measure sits on its own dial, with an illustrative July 2026 reading from public indices. A live version is at vitals.html.

The throughline is modesty. These methods can build confidence and expose weakness, and the framework treats its own evaluation as an open instrument-building problem rather than settled science. The protocol is what makes that problem tractable instead of an excuse.

Open questions
  • Validated, comparative resilience metrics for governance systems do not yet exist; building them is the core open task.
  • Early-warning indicators carry real false-alarm rates and assume a stationarity that politics violates.
  • Case studies are partial and confounded, and none isolates an Aeviternal mechanism cleanly.
  • Integration is the deepest gap, since components are tested in isolation but the cross-coupled failures live at the federation scale marked None yet. There is no test for a whole civilization, so the interface is tested instead as a partial harness, and the praxis staircase is read as an evidence ladder where each rung pre-registers the failure expected at the next.
  • Direction
    • Triangulate resilience as recovery-of-function curves, structural metrics such as modularity and polycentricity, and a qualitative read of latitude and precariousness (Holling; Walker et al.); treat named systems as partial implementations (Ostrom commons, Mondragon, Taiwan vTaiwan, Wikipedia); use agent-based models to stress-test mode transitions rather than to forecast.
FM–05 · Core Propositions

Core Propositions

Stable

The following propositions represent the load-bearing claims of the framework. They are stated in falsifiable form and numbered for citation. A collaborator who rejects P–03 but accepts P–01 and P–02 can say so precisely.

P–01 Civilizational stability is not a state to be achieved but a process to be maintained. Systems that suppress oscillation in pursuit of equilibrium accumulate brittleness and become more, not less, vulnerable to collapse.
P–02 Every system tends toward the capture of its own feedback mechanisms by those who benefit from the suppression of corrective information. Without active structural resistance, this tendency is universal.
P–03 In AI-mediated information environments, perception itself becomes a contested civilizational resource. The ability of a population to accurately assess its own conditions is not a given. It is a design problem.
P–04 Knowledge is primary civilizational infrastructure. A society that privatizes the means of knowing will reproduce extraction at the epistemic level regardless of what it does at the material level.
P–05 Stewardship, meaning accountable, domain-specific, decaying influence earned through contribution, is the appropriate unit of social power in an adaptive civilization. Sovereignty, meaning permanent, general, self-reinforcing authority, is the primary failure mode.
P–06 Transitions between civilizational modes are not events. They are extended conditions requiring bridge structures designed for partial adoption, iterative refinement, and reversibility.
P–07 Every technical mechanism in this framework is a requirement, not an implementation. It must be buildable on open and auditable standards, and it must stay agile enough to swap cryptographic primitives as those standards change and as new ones are invented. To name a technology is to give an example, never a commitment.
FM–06 · Objections and Responses

Objections and Responses

Stable

A framework that asks to be taken seriously has to answer its strongest critics rather than its weakest. The objections below are the ones that recur, stated in the words a skeptic would use, each with the framework's reply and a pointer to where the reply is worked out. Several are conceded as unresolved, because a defense that wins every point is not a defense but a sales pitch.

OB–01"It is utopian and will not survive contact with real power." Aeviterna assumes the opposite of good actors. It is built as a migration framework around the expectation that systems drift toward concentration and that motivated insiders will try to capture them, which is why anti-capture architecture and decaying standing are load-bearing rather than decorative (EQ–04, EQ–05, EQ–06).
OB–02"Rotation and decay will just be gamed; the same elite recirculates through new seats." This is conceded as the deepest structural risk. The reply is to watch network concentration rather than individual turnover and to reward demonstrating capture through bounties, while recording plainly that no mechanism fully closes the gap (EQ–06).
OB–03"Whoever sets the thresholds and runs the monitors becomes the new power." The monitoring role is the recurring capture point, so the anti-capture stack is applied to the monitor itself: plural competing measurers, rotating panels, open data, and expiring definitions. Recursive oversight of the monitor is bounded by the reading seat's circle of mutual watchers, not solved (EQ–06e).
OB–04"Emergency powers will be abused, and crises will be manufactured to trigger them." Emergency authority is held in escrow with the declarer separated from the beneficiary, expires by default, is vetoable by nested units, and decays faster than ordinary standing. A broadly popular manufactured crisis can still clear these checks, which the framework states rather than hides (EQ–09).
OB–05"Enough is just recession with better branding; pulling back produces hardship." This is the strongest objection from the degrowth debate. The reply is to decouple provision from growth by securing a floor of basic services first, so sufficiency is reached without contraction of need, and to concede that without the floor the criticism holds (EQ–02b).
OB–06"You are imposing a blueprint people never consented to." Aeviterna declines the tacit-consent move and rests legitimacy on genuine low-cost exit, voluntary prefigurative adoption, and a fairness test behind a veil of ignorance. Whether a normative framework can be both committed and non-impositional is an irreducible tension the framework owns (OQ–06).
OB–07"The cognitive-terraforming claim is a moral panic about new technology." The framework rests its weight only on the robustly evidenced mechanisms, attention capture and capacity erosion, and treats the contested claims, such as strong radicalization and sealed bubbles, as the least-evidenced end of the mechanism ordering rather than as load-bearing (EQ–03).
Part IThe Diagnosis
EQ–01

The Civilizational Problem

Stable

Capitalism industrializes greed. Socialism industrializes fairness. Aeviterna industrializes wisdom.

Human systems increasingly optimize for extraction rather than continuity. Capital accumulation, attention economies, institutional opacity, and fragile dependency chains produce systems that are efficient in the short term but unstable over longer horizons. Modern civilization faces converging crises across ecology, cognition, infrastructure, governance, social trust, and emotional sustainability.

The diagnosis is not that markets are evil or that states are corrupt. The diagnosis is structural. We built optimization engines and then forgot to specify what they were optimizing for. The result is systems that successfully maximize measurable proxies (GDP, engagement, quarterly returns) while degrading the unmeasured conditions that make civilization possible.

Aeviterna begins from the premise that economies are tools, not belief systems. Markets, governance structures, and technologies are not sacred. They exist to support human flourishing, adaptive resilience, and long-term civilizational survivability. When they fail to do that, they are broken tools, not gods to be appeased.

the widening gapmeasured proxyunmeasured conditionstime
The civilizational problem. Optimization drives the measured proxy (GDP, engagement, quarterly returns) upward while the unmeasured conditions that sustain a society (trust, ecology, meaning) drift down. The widening gap is the cost no one is pricing.
Capitalism
industrialized
greed
Socialism
industrialized
fairness
Aeviterna
industrializes
wisdom
The residual, named

The diagnosis has a real counter-literature, and the chapter is stronger for carrying it. More than thirty countries have decoupled carbon absolutely from growth, including on consumption-based accounting (Hausfather; Ritchie). What has been observed nowhere is absolute decoupling of material throughput at the global scale, and the carbon decoupling on record runs several times slower than climate-compatible pathways require, on trajectories exceeding two centuries to near-zero (Hickel and Kallis; Vogel and Hickel). The claim therefore rests on material throughput and on speed, not on the impossibility of any decoupling, and it names its falsifier. Sustained global absolute decoupling of material use at climate-compatible rates would break this chapter's premise.

Part IIFoundations
EQ–02

Adaptive Stability

Stable

Stillness is not health. Stillness is the first symptom of a system that has lost the capacity to correct itself.

Aeviterna does not seek perfect equilibrium. Static equilibrium is brittle and incompatible with living systems. Instead, Aeviterna seeks adaptive stability, the capacity of a civilization to oscillate within survivable bounds while preserving resilience, recoverability, and continuity.

A healthy civilization detects instability early, corrects proportionally, avoids overcorrection, preserves flexibility, and distributes recovery capacity. These are not aspirational qualities. They are engineering requirements. A system that lacks any one of them will eventually fail on the dimension it neglected.

Aeviterna is therefore a feedback civilization. Its central objective is not perfection but sustainable self-correction. The measure of civilizational health is not whether the system is currently stable, but whether the system retains the capacity to return to stability after disruption. That capacity is what extraction destroys, quietly, over time, until the bill arrives all at once.

paralysisstatic, brittlebounded oscillationadaptive stabilityrunawaybeyond recovery
Adaptive stability is bounded oscillation, movement that stays inside a survivable band. Holding still turns brittle; swinging past the bounds cannot be recovered.
Interactive

Bounded oscillation, in your hands

Move the sliders. With too little correction, shocks accumulate until the system breaks its bounds; with too much, motion is damped to a brittle standstill. Adaptive stability is the living band between those failures.

upper bound lower bound
Bounded oscillation · adaptive stability
The residual, named

No documented case exists of deliberately bounded oscillation governing at societal scale. The concept's home literatures are ecological and control-theoretic (Holling), and one scale down, in the adaptive management of single ecosystems, the record is thin. Of roughly twenty-five large programs reviewed, few became real experiments, and the field's own retrospective is a catalog of pathologies (Walters; Allen and Gunderson). The oscillation thesis is an analogy awaiting its first societal instance, the harness (FC–04) is in-silico only, and the falsifier stands, a polity that adopts banded oscillation and shows no resilience advantage over comparable static-target polities within a generation would break it.

EQ–02b

The Economics of Enough

Stable

An economy that cannot say enough will keep saying more until there is nothing left to take.

Aeviterna replaces growth as an imperative with enough as a target. The aim is not to maximize output but to meet needs well within the limits that keep a civilization alive, and then to let the economy oscillate inside that band rather than climb without end. This is the material expression of adaptive stability. An economy is healthy when it holds its people above a floor of provision and stays below a ceiling of ecological harm, and recovers to that band after a shock.

Two boundaries define the space of enough. Below the social foundation, people lack what they need, which is shortfall. Above the ecological ceiling, the economy consumes the conditions of its own future, which is overshoot. The safe and just space is the band between them, the shape Kate Raworth named the doughnut, and Aeviterna treats holding the economy inside that band as a primary objective rather than a side effect of growth.

The deeper shift is in where value is taken to come from. An extractive economy books a gain whenever something is depleted, whether a forest is cut or a commons enclosed. A post-extractive economy books a gain when the conditions of life are instead maintained and regenerated, when soil is rebuilt or knowledge is shared. Aeviterna counts stewardship and repair as wealth, because in any system that lasts they are what keeps wealth possible at all.

shortfallneeds unmetthe space of enoughsufficiency and regenerationovershoot: ecological harmsocial foundationneeds met by rightecological ceilingplanetary limits
The economy of enough aims for the safe and just space, the doughnut named by Kate Raworth, above a social foundation where needs are met, and below an ecological ceiling. Too little is shortfall; too much is overshoot.
A floor of provisionUniversal basic services put essential needs, including housing, care, transport, and the knowledge commons, outside the reach of market precarity, so the floor of the doughnut is held by right rather than by purchasing power.
A ceiling on throughputMaterial and energy throughput is capped within ecological limits and treated as the economy's binding constraint, the measurable variable that degrowth scholarship makes central.
Commons before marketsShared resources are governed as commons on Ostrom's design principles, and markets are kept as bounded tools inside the safe and just space rather than the organizing principle above it.
Ownership that stays localProductive capacity is held through cooperatives and community wealth building, so returns circulate locally and economic power resists concentration, the anti-capture principle applied to the economy.
Regeneration over wasteDesign favors repair and restoration over disposal, so the default economic act rebuilds stock rather than drawing it down.

The hard part is not describing enough but escaping the compulsion to grow. Debt and shareholder primacy make endless expansion structural under capitalism, so an economy cannot simply choose to stop. Aeviterna removes the compulsion rather than asking for restraint. Provision is decoupled from growth through the floor of basic services, and value is decoupled from depletion through the post-extractive accounting above. The honest risk is that pulling back without these supports in place produces hardship rather than sufficiency, which is the strongest objection the degrowth debate has raised, and the reason the floor has to come first.

The floor raises a sharp internal question, because a universal floor is non-excludable by construction, and non-excludable provision at scale has historically required a durable authority with the power to tax and bind, which looks like the very sovereignty (P–05) the framework exists to abolish, while Ostrom's commons work precisely because they are bounded and excludable. The resolution turns on the word general. A body can be permanent without being general or self-reinforcing, a single-purpose provisioning commons, financed where possible as a dividend from common wealth (the Alaska model), held as a fiduciary trust, and persisting by charter while its officeholders rotate under decaying standing, is durable but is not a sovereign. The honest residual is that this severs sovereignty's form, not its function, since the floor still needs jurisdiction-wide, compulsory finance, and whether a chartered, decaying-officeholder body can hold that without re-growing into the sovereign by the Robert Moses route, an insulated authority that slips its leash, is the open test.

Open questions
  • What sets the floor and the ceiling in numbers, and who measures them, is unresolved (it connects to the resilience metrics in the Methodology section).
  • Decoupling provision from growth at scale is unproven; the strongest critique (Pollin) holds that contraction without it is socially destructive.
  • Allocation without growth as the coordinator needs mechanisms beyond both markets and central planning that are still being designed.
  • A universal floor needs durable, jurisdiction-wide, compulsory provisioning capacity, which is state-like in reach even when it is not a P–05 sovereign; keeping that body single-purpose, fiduciary, and decaying in personnel without re-growing the sovereign is unresolved.
  • Direction
    • Target the safe and just space (Raworth's doughnut; Rockström's planetary boundaries), hold a floor with universal basic services, cap throughput, govern commons on Ostrom's principles, and localize ownership through community wealth building.
EQ–02c

Civilizational Metabolism

Draft

A civilization runs on energy and material throughput before it runs on money. The quantity that actually constrains what a society can build and hold together is net energy, the surplus that remains after the energy spent to obtain energy, together with the volume of materials it can move without breaching the limits of the system it lives inside. The Economics of Enough (EQ–02b) sets the ceiling and the floor in terms of welfare. This section adds the layer beneath it, the physical metabolism that every economy is ultimately a description of.

The thermodynamic reading is long-standing and well evidenced. Georgescu-Roegen placed the entropy law at the center of economics, Daly built the steady-state economy on it, and Tainter showed that complexity itself carries a rising energy cost that eventually exceeds its returns. Net energy is usually written as energy return on investment, the ratio of energy delivered to energy spent obtaining it, and a complex society needs a comfortable surplus, with common estimates near ten to one for the kind of complexity modern life assumes, though the figure depends on where the boundary is drawn and remains contested. Smil's account of the material pillars, ammonia and steel and cement and plastics, is the companion point, since those flows resist quick substitution and a civilization's material metabolism therefore changes slowly and has to be designed for rather than assumed away.

total throughputabsolute ecological cap on throughputheadroomlow-entropy intakeenergy + materialsSociety does worknet energy funds thecomplexity it can affordhigh-entropy outputwaste heat + materialscircularity recovers some, it cannot close the loopan economy is a one-way throughput from low to high entropy, held under an absolute cap
An economy is a one-way throughput. Low-entropy energy and materials enter and fund the work that buys the complexity a society can afford, then leave as high-entropy waste. Total throughput is held under an absolute ecological cap, and circularity recovers some of it without closing the loop.
Net-energy accountingTreat energy return on investment as a first-class design variable. A subsystem that cannot run at a positive and sufficient net-energy return is not affordable, whatever its monetary price.
Material budgets bound to limitsSet absolute throughput caps against the boundaries of the host system, and make circularity serve those caps rather than stand in for them. Global circularity fell from about nine percent to under seven over recent years because growth outran recycling, which is what an uncapped circular economy produces.
Metabolism as an instrumentThe ecological-throughput vital sign reads energy and material flow against a safe band, so the metabolism is visible and governed in advance rather than inferred after the damage is done. The throughput sign gains an agent that can hold a breach in EQ–08e.

Stated as a band, this gives enough a thermodynamic address. The floor is a minimum net-energy surplus, an energy return high enough to fund the activities that lift a society above subsistence, pegged to the rung that funds the dignity floor itself. The ceiling is the throughput above which the biosphere degrades, read across the nine planetary boundaries, most of them already crossed. Enough is the band between the two.

The evidence here is mixed and still moving. Recent work on useful-stage energy return has overturned the older claim that renewable systems are too thin to power a complex society, and the degrowth and green-growth positions remain unreconciled on whether throughput can fall quickly enough while welfare holds. The framework takes a position only on the discipline, that energy and materials are measured and bounded, and not on the contested numbers.

Open questions
  • Calibrating throughput caps per bioregion, without inviting either a rush at the threshold or authoritarian rationing, is unresolved. The dignity constraint in EQ–06c and the governance modes in EQ–09 bound how any cap may be enforced.
  • The right net-energy floor for a given level of complexity is contested and depends on where the measurement boundary is drawn.
  • Direction
    • Account for net energy and material throughput directly, set absolute caps against the host system's limits, and let prices and circularity work inside those caps rather than around them.
  • Implementation sketch · exploratory
    • A net-energy and materials ledger feeding the ecological-throughput dial, caps set per bioregion and ratcheted over time, with enforcement routed through the ordinary governance modes rather than emergency centralization.
EQ–02d

Monetary Ecology

Stable

One currency for everything is a monoculture, and monocultures are efficient right up until they collapse.

Cognitive Ecology (EQ–03b) warns that a monoculture of thought is brittle and EQ–02c warns that a material monoculture overshoots, but money has so far escaped the same scrutiny. A civilization that runs nearly everything on a single bank-debt currency, defended almost entirely on efficiency, has built at the level of value the very monoculture the framework distrusts everywhere else. Monetary ecology applies diversity-as-infrastructure to money.

The argument is measurement, not analogy. Ulanowicz, Lietaer and colleagues applied the mathematics of ecosystem flow networks to monetary systems and found the same law. A flow network's sustainability balances efficiency, the streamlining that comes from fewer channels, against resilience, the buffering that comes from diversity and redundancy. Every sustainable natural network sits in a window between the two, with the optimum nearer the resilience side, and the global monetary monoculture has overshot toward brittleness, which makes recurring crises a structural feature rather than a run of accidents.

A monetary ecology is deliberately plural, with currencies tuned to functions. A commercial currency serves trade, mutual credit serves local exchange that resists accumulation, demurrage currencies lose value if hoarded and so keep circulating, and commons currencies carry the care and ecological work markets price badly. The strongest modern evidence is Switzerland's WIR, a business-to-business mutual-credit network since 1934 whose turnover is countercyclical, rising in recessions and falling in recoveries, so it acts as an automatic stabilizer.

sustainability of the money networkmonetary monoculture lineover-proliferation lineEfficiencyWindow of viabilityExcess diversityone currency, brittleplural currencies, resilienttoo many weak moniestoday's money
A window of viability for money. Pushed all the way toward efficiency a single currency turns brittle; drifted all the way toward diversity the system stagnates in a clutter of weak monies. A plural ecology of currencies tuned to functions sits in the resilient middle.
Currencies tuned to functionsA plural set, commercial, mutual-credit, demurrage, and commons, rather than one medium straining to do all of them at once (Lietaer).
The window of viabilityRead the money network the way ecology reads a flow network, with diversity as the buffer that survives a shock (Ulanowicz, Goerner, Lietaer, and Gomez).
Demurrage against hoardingA small holding cost keeps money circulating to where it is needed, the reverse of interest (Gesell; the Wörgl scrip of 1932).
Currency diversity as a readingThe diversity of currency types is itself a resilience reading, guarded against rewarding a proliferation of weak monies.

The ninety-year record of complementary currencies is now long enough to say what separates the survivals from the several hundred deaths, and OQ–07 is owed that answer as a worked spec. The survivors share a shape. Switzerland's WIR has cleared business-to-business credit countercyclically since 1934; Sardinia's Sardex, the strongest living case, has cleared over a billion euros in fifteen years on a no-convertibility rule that makes the only exit a purchase; Maricá in Brazil pays the largest municipal-currency basic income anywhere, anchored in oil royalties. The graves share the opposite shape. The Bristol Pound died fully backed and convertible, with no credit function and no binding anchor, when its grant funding did, and Argentina's trueque networks reached two and a half million households and collapsed in eighteen months from hyperissuance. A currency earns standing in the ecology, then, when it clears five thresholds, and holds standing only while it keeps clearing them at review.

A real credit or clearing functionIt extends purchasing power, a credit line, or the clearing of receivables, rather than re-denominating the national unit; full backing plus convertibility is the payment costume every grave wore.
An anchor that binds circulationSomething recurring must be payable, dischargeable, or issued in it, an acceptance obligation, debt dischargeable only in-network, wage-like issuance into a merchant base, so the anchor binds the unit's circulation rather than offering an optional exit; a tax-acceptance bolt-on did not save Bristol.
Countercyclical use observed, not assertedThe record must show use rising when the conventional layer contracts, the WIR curve across seventy years and the Sardex crunch and pandemic records, because resilience is the whole reason plural money earns a place here.
Operations that survive their donorsThe fee, demurrage, or brokerage model covers operating costs, or the standing is provisional; patient capital marks a case ambiguous rather than proven, the open question the Sardex operator still carries.
Governance that survives its foundersFounders can exit without issuance failing, and credit capacity is keyed to members' ability to sell rather than to the issuer's appetite, WIR's ninety years against Argentina's hyperissuing center.
Open questions
  • The ecology's health is now read like any other vital sign. The standing register is public under the legibility budget (EQ–03b), reviews are scheduled, the reading seat (EQ–06e) reads the register, and a currency that stops clearing a standing threshold loses standing at review rather than by decree, with the record saying why. Standing is the polity's published judgment of which layers count as infrastructure, not a license to exist; the framework bans nothing here.
  • The macro limit is carried in full. No complementary currency anywhere has mattered macroeconomically, so the claim these thresholds license is firm-level resilience, ten to fifteen percent on member sales in the strongest panel, and regional buffering in downturns, not monetary sovereignty in miniature.
  • Where the window of viability sits for a real economy, and how a society knows which side of that window it stands on, is still not known with precision, and plural money keeps its own friction.
  • Residual
    • Complementary currencies still fade once their founding crisis passes, unless an anchor binds them to ordinary time, and the thresholds are the framework's named answer to that fade, tested at every review rather than assumed solved.
EQ–02e

The Commonwealth Trust

Draft

The rents of common wealth can become the floor, if the vessel that holds them is built from the features that survived being raided.

The hardest questions concede that the universal floor (EQ–06c) needs jurisdiction-wide, compulsory finance, the one capacity the framework's politics is most reluctant to grant, and the concession stood without an organ behind it. The economics carried a deeper silence, no treatment of land and rent at all, the oldest question in the genre. This organ closes both. The world has run the experiment for a century, routing the rents of common wealth, land, oil, phosphate, the cousins of spectrum, into institutions chartered to provision without governing, and the record is long enough to read which design features held and which were raided or re-grew into sovereigns.

The record is not subtle. Vienna's municipal housing, built on an earmarked tax and a purpose-locked land bank, survived fascism, the Reich, war, and the privatization wave that sold the same country's federal stock, and one Viennese in four still lives in it. Norway's fund held its corpus for thirty years on a payout norm no government has broken. Alaska constitutionalized its corpus and left the dividend to statute, and at fifty years the corpus stands untouched while the dividend was captured in a single veto cycle. Against these, the failures name themselves: Nauru's trust died of unchecked discretion, the Moses authorities re-grew into an unaccountable sovereign by entrenching the operator instead of the purpose, and the United Kingdom, holding no vessel at all, watched an ocean of North Sea rent merge into the general budget and vanish. The lesson that binds them is Alaska's. What is entrenched survives and what is left to ordinary politics is bargained away, so the floor payment must sit at the corpus's own entrenchment level, not one story below it.

The commonwealth trust is therefore constituted from the features that held, and under the framework's own primitives rather than a new apparatus. It is a chartered single-purpose trust that turns the rents of common wealth into the dignity floor's named funding mechanism, and it moves the second of the hardest questions, the floor's compulsory finance, out of the concession column and into the body. The capacity the floor needs is not a general taxing sovereign but a chartered claim on rents the community already owns. It sits beside the financing ladder of EQ–04 without merging into it, the ladder scaling non-extractive firms while the trust funds the floor, and it equips at last the provisioning-floor note EQ–02b added to separate durable provisioning capacity from sovereignty.

THE COMMONWEALTH TRUSTRents ofcommon wealthland · resources · spectrumThe corpusconstitutional, entrenchedprincipal locked, income onlyPayout formula% of market valueThe dignity floorfinanced, not bargainedReading seatreads accounts, exclusions, driftSecond keygates any draw beyond the formulaRegrowth tripwire · F–07fires if mandate breadth or discretion grows across two reviewswithout a forced contest, the Moses route it is built against
The commonwealth trust. Rents of common wealth fund a constitutional corpus that may spend only its income, on a published formula, into the dignity floor. A reading seat (EQ–06e) reads the accounts and a second key outside the spending coalition gates any draw beyond the formula, while the regrowth tripwire (F–07) fires if the mandate or the discretionary share grows across two consecutive reviews without a forced contest, the Robert Moses route the organ is built against.
A constitutional corpusPrincipal may be spent only on income-producing investment, the Alaska rule, entrenched at the charter's hardest layer, and the trust never lends to the polity that charters it, the anti-Nauru rule.
A published formula payoutA percent-of-market-value draw with a smoothing window, the rule that has never been overdrawn where it was enacted, and the floor payment it finances is entrenched as deeply as the corpus, because a statutory dividend is one veto from a bargaining chip (the Wielechowski lesson).
An enumerated mandate, never generalThe trust finances the floor and nothing else, surpluses above the formula compound, the purpose is entrenched and the operator stays dismissible, the inversion of the Moses failure (EQ–05).
A reader chartered at the corpus's levelA reading seat (EQ–06e) reads the accounts, the exclusions, and the mandate drift, and its existence is entrenched exactly as deep as the money, because a seat created by decree is suspended by decree the moment the reading gets expensive, the lesson Oslo taught in November 2025.
A second key outside the spending coalitionDraws beyond the formula require the concurrence of a keyholder the spenders do not appoint, Singapore's second signature and Nauru's foreign co-contributors generalized, and every turning of the key is published.
A named valuation bureauLand and resource rents are assessed by a permanently funded, publicly auditable bureau whose rolls are published under the legibility budget (EQ–03b), with decade-scale transition buffers, the gradual glide path rather than the valuation cliff that has killed the tax wherever the bureau was starved.

Custody was always the easier half of the question, and the harder half is acquisition, since every flagship above began from a gift or a sovereign act, and a community that adopts this framework owns nothing yet. The record of commons assembled with neither turns out to carry a bootstrap ladder with five rungs, each priced and each alive at scale. Member capital first, the community-share record of two hundred ten million pounds raised from a hundred thousand people with no compensation scheme behind it, and non-voting preferred stock that raises money without selling a vote. Patient debt second, the ladder's load-bearing rung, since every clean non-gift case in the record runs on subordinated loans at zero to five percent, the direct member loans that unlock bank leverage, the seller notes that financed the great conversions from 1929 onward, and the non-extractive terms whose repayment never exceeds profits. The veto lock third, and it is the cheapest fact in the framework's whole evidence base, because the entrenchment this organ demands can be installed through ordinary company law for the cost of incorporation, a golden share holding one percent of votes and a veto over sale, liquidation, and charter change, the one-vote-each partnership that has kept some hundred ninety housing projects unsellable for three decades. Conversion fourth, buying existing enterprises with the first two rungs where statutory routes now exist. Captured rents fifth, the solidarity levy and the resale formula by deed, which recycle the corpus's own yield into the next acquisition and turn the ladder into a loop.

The base rates assign the trust its second job. Attempted community acquisition dies in the funnel, at rates near one in twelve to one in fifty depending on asset class, while completed acquisitions survive at rates that embarrass conventional ownership, ninety-two percent of community-share businesses still trading and land-trust homes passing through the 2008 crash at a tenth of the conventional foreclosure rate. The binding constraint is patient finance at the moment of purchase, not stewardship after it, so before a commonwealth trust is large enough to pay the floor it serves as the commons' patient lender, the rung-two counterparty whose loans let a community reach the closing table before a private buyer does. Acquisition without a lock is a balance capital markets can unwind, the lesson of the employee-owned giants lost in bankruptcy and recapitalization, so the veto share is installed before the first asset arrives, and the whole ladder is registered to be tested rather than admired, carried as falsifier F–11 on the funnel and the lock at once.

THE BOOTSTRAP LADDERHow a corpus is assembled without a sovereign1 · Member capitalshares without votes2 · Patient debt0-5%, subordinated, long3 · The veto lockentrenched at company-law price4 · Conversionseller notes, statutory routes5 · Captured rentslevy and deed, the loop closesthe yield of rung five funds the next community's rung one, the Syndikat loop at fortythe funnel: most attempts die at purchase, so the trust's first job is patient lending · F–11 tests both legs
The bootstrap ladder. Member capital and patient debt reach the closing table, the veto lock entrenches before the first asset arrives, conversion buys what exists, and captured rents recycle the yield into the next acquisition. The funnel, not stewardship, is where attempts die, and F–11 tests the ladder on both legs.
Open questions
  • The cleanest design evidence comes from the two cases with real political turnover, Norway and Alaska, so the mechanism leans hardest on formula payout, constitutional corpus, and seat entrenchment; Vienna's and Singapore's endurance is confounded by near-unbroken single-party rule and cannot be cleanly separated from their designs.
  • Land rent nowhere finances a whole modern welfare state, and the organ claims only the floor, not the state; the equality of aggregate land rents with public-goods spending holds under idealized conditions, not as a promise of scale.
  • The valuation bureau is a real and expensive dependency that has failed for a decade inside a rich, competent bureaucracy, so the boring infrastructure, not the theory, is the load-bearing risk.
  • The bootstrap's clean cases are younger and smaller than the gift-founded flagships they must eventually replace, and patient capital moves slower than private capital by construction, so even the best ladder documented leaves most communities outbid; the funnel is priced, not repealed.
  • Direction
    • Route the rents of common wealth into a chartered single-purpose trust built from the features that survived being raided, entrench the floor payment at the corpus's level, read the whole through a seat chartered as deep as the money, and grow the corpus by the bootstrap ladder, patient lending first, the veto lock before the first asset, the levy closing the loop.
Falsifier F–07

A trust large enough to fund the floor is large enough to be worth capturing, and the record's last lesson arrived in 2025. A fund big enough to matter becomes a foreign-policy actor whose reading can be bent by allies, so scale itself is a capture surface. The anti-regrowth claim therefore ships as a candidate falsifier rather than a promise. If the trust's mandate breadth or its discretionary share of spending grows across two consecutive scheduled reviews without a forced contest firing, the organ is wrong as designed. The regrowth pressure is permanent; the tripwire is registered to be tripped, not admired. Carried as F–07. Enforcement here rides the ladder's structural and borrowed rungs (EQ–06f), the veto share entrenching at company-law price and the charter binding as the bylaws courts already enforce.

EQ–02f

The Governance Overhead

Draft

An architecture that will not name its own price is asking to be trusted twice, once on its mechanisms and again on their affordability, and this framework asks to be trusted on neither. So the stack is priced, and the first result of pricing it is a finding about everyone else. Almost no member-governed institution on earth prices its governance as a category. Participatory budgeting publishes its allocations to the dollar and its administrative costs nowhere; no statistical agency publishes what its own time-use survey costs; cooperatives dissolve their democracy into operating expense; the one place that itemizes an invoice for governance as such is the DAO sector, whose ledgers also show the failure mode, one treasury's operating cost reaching thirty-four million dollars against thirty-eight million in revenue before its members cut it back. Publishing the governance overhead as a two-currency ledger, money and member time both, is therefore not bookkeeping. It is a discipline the record shows almost nobody keeps, the anti-bloat instrument and the anti-starvation instrument in one document.

The priced record, where it exists, has a shape. Standing organs are cheap. The auditor of a six-trillion-dollar government runs on twelve thousandths of one percent of what it reads, and documents seventy-six dollars returned per dollar; a constitutional court prices near fifty euro cents per resident per year, a national archive near a dollar thirty, a standing sortition chamber near two euros. Episodic deliberation is dear, up to forty thousand euros per seated citizen for a one-off national convention whose members graded its consequences three out of ten, because episodes pay for recruitment, secretariat, and learning every time, while standing bodies amortize them. And verification is cheapest of all, a risk-limiting audit confirming a statewide election from one hundred forty-two ballots where the old method pulled thirty-two thousand. The framework's design choices, standing seats over episodic panels, statistical verification over exhaustive recount, follow the price gradient the record already drew.

Priced at three reference scales in two currencies, the stack lands inside the rounding error of what it governs. A forty-person cooperative runs the full applicable set near fifteen to twenty thousand dollars a year plus two to four hours per member per month, with the time worth roughly three times the cash at the standard volunteer rate, which is the second structural finding. Unpaid governance does not remove cost; it hides cost in member time, the reason this organ counts both currencies always. A ten-thousand-member federation cell runs near thirty dollars per member per year, the seats amortizing by an order of magnitude. A city-scale polity runs the whole architecture near fifteen to twenty-five dollars per person per year, two to four tenths of one percent of a typical public budget, against the benchmark that an entire national statistical system costs about one tenth of one percent. The scale curve bends the right way, and the minimum viable subset at each scale begins with the clocks, the codified exit, and a quarterly review of the few vital signs an adopter names, which prices the first rung of the on-ramp, the Lite tier.

The counter-finding governs the organ's place in the architecture. In the documented record, cost is almost never what kills a governance instrument. The instruments that die are the cheap, constituency-poor ones: a national survey voted out by one chamber over savings that rounded to zero and saved only by the other, a time-use survey shelved for fifteen years, a fourteen-million-dollar accident board ordered shut and rescued, a census underfunded early at the price of three billion later. Affordability is necessary for adoption and insufficient for survival, which is why the overhead ledger is read beside the floor's own finance, and why the organs' funding belongs at the entrenchment level of EQ–02e rather than in the annual weather. What the ledger adds is the other jaw of the vise. An entrenched budget without a published overhead band is how governance quietly bloats, and a published band without entrenchment is how it quietly starves. The organ holds both.

The two-currency ledgerEvery governing body under the framework keeps and publishes its governance cost in money and in member-hours, the volunteer subsidy counted rather than hidden, on the precedent of the one governance form whose time cost is written into statute, the works-council release schedule.
Standing over episodicDeliberative and reading functions run as standing bodies with decaying membership, not as commissioned episodes, because permanence amortizes recruitment, secretariat, and learning by up to two orders of magnitude in the priced record.
The verification dividendLegitimacy checks use statistical verification wherever it exists, the risk-limiting design, because sampling collapses the price of confidence and cheap verification is what keeps the falsifier register affordable.
The overhead bandGovernance cost is held inside a published band per scale, a floor as well as a ceiling, read at the vital-signs review; breach in either direction is a reading, bloat above, starvation below.
Financed where it cannot be bargainedThe instrument layer's budget lives at the charter's entrenchment level, per EQ–02e, because the record's organs died of constituency-poverty at trivial prices, not of expense.
THE GOVERNANCE OVERHEADThe stack, priced inside the rounding errorF–10 ceiling · 2% budgetCo-op · 40~$15-20k + 2-4 h/member/mo~1-2% of budgetF–10 ceiling · $50/memberCell · 10,000~$30/member/yr + 10-20 h/yrseats amortize 10xF–10 ceiling · 1% public budgetPolity · 1-5M~$15-25/person/yr · ~0.3%rounding errorabove the band: bloat, the DAO that ate its hostbelow the band: starvation, the survey killed for rounding-error savingsgreen bars: the modeled stack in cash; every scale also pays a time currency the ledger counts; gold: the ratified ceilings; breach either way is a reading
The governance overhead at three scales. The modeled stack (green) sits inside the ratified F–10 ceilings (gold) at every scale, with the least headroom at the smallest, and the per-member cost falls an order of magnitude as seats amortize. The band fails in both directions, bloat above it and starvation below it, and both failures are readings, not surprises.
Open questions
  • Every number here is modeled from analogues, not measured from an adopter, and only the orders of magnitude deserve trust until a pilot runs; the framework's own balance sheet remains a forecast about itself.
  • The two currencies do not price the third one. The record's standing sortition chamber throttled its cadence over staff capacity, not money, so administrative attention binds before either cash or hours, and this organ names that currency without yet modeling it.
  • The counter-finding cuts at the organ itself. Since cost was rarely the binding constraint in the record, a clean balance sheet cannot protect the instruments it prices, and survival routes through entrenchment (EQ–02e) and demand, the frontier the demand curve owns.
  • Direction
    • Publish the two-currency ledger for every governing body, hold the cost inside the per-scale band with breach in either direction read at the review, run deliberation as standing bodies, verify statistically, and finance the instrument layer at the entrenchment level rather than in the annual budget.
Falsifier F–10

If running the applicable stack requires more than 2 percent of operating budget or more than 5 percent of member work-hours at cooperative scale, more than 50 dollars per member per year or 2 percent of member work-hours at federation scale, or more than 1 percent of the public budget at polity scale, the stack is too heavy at that scale and the adoptability claim fails there. And if tracker-coded adopters abandon named organs over cost within two years at any scale, the design has failed the demand test regardless of the ceilings. Carried as F–10.

EQ–03

Cognitive Sovereignty

Stable

Privacy protects autonomy, secrecy protects hierarchy, and ambiguity protects adaptability.

In the age of AI-mediated information ecosystems, the central battleground of civilization becomes cognition itself. Modern systems increasingly shape attention, emotion, narrative framing, and perception through adaptive algorithmic environments. This process, what Aeviterna calls cognitive terraforming, is not a side effect of these systems. It is the product.

Aeviterna identifies cognitive capture and algorithmic authoritarianism as major existential risks, not because they involve overt coercion but precisely because they do not. When perception is shaped, control becomes invisible. The most efficient form of power is one that makes the people being governed believe they are governing themselves.

Cognitive sovereignty, the ability of individuals and societies to understand and contest the systems shaping their perception, is therefore not a luxury. It is a prerequisite for any other form of freedom. A population that cannot accurately assess its own conditions cannot make meaningful political choices. It can only ratify the choices that have already been made for it.

A transparency doctrine that treats all opacity as suspect carries a hidden risk, so the framework rests on a necessary distinction. Exploitative opacity covers corruption, hidden extraction, concealed manipulation, and institutional secrecy. This is what transparency is designed to defeat. Generative opacity covers privacy, emotional ambiguity, exploratory uncertainty, unoptimized creative space, and symbolic identity formation. This is what transparency must be careful not to destroy. They are structurally different things. Conflating them produces a framework that is hostile to exactly the cognitive wilderness it needs to protect.

Exploitative opacitywhat transparency defeatsGenerative opacitywhat transparency must protecttransparencycorruptionhidden extractionconcealed manipulationinstitutional secrecyprivacyemotional ambiguityexploratory uncertaintyunoptimized creative space
Two opacities. Transparency is built to defeat the exploitative kind without destroying the generative kind; conflating them turns a tool of accountability against the privacy and ambiguity a civilization needs.

Aeviterna therefore supports transparent governance systems, publicly auditable institutional processes, distributed AI oversight, epistemic pluralism, civic systems literacy, protection against covert narrative manipulation, and the structural separation of discourse systems, enforcement systems, and allocation systems. The goal is not informational uniformity. The goal is epistemic resilience, a society that can absorb distortion without losing its capacity for collective judgment, while preserving the protected ambiguity that makes long-term adaptation possible. The claim is registered as F–02. Epistemic commons shown to function sustainably without these protections would refute it.

These claims describe what cognitive terraforming does. How it operates is a separate question, and the reshaping works through identifiable mechanisms, set out here from the most robustly evidenced to the most contested.

Attention CaptureInterfaces use intermittent reward to hold attention, importing the logic of the slot machine.
Engagement AmplificationOptimizing for engagement privileges moral and emotional content and biases the shared mood toward outrage.
Exposure ShapingCuration sets which content is cheap to reach, shaping the information supply and what producers make.
Synthetic DestabilizationCheap fakes make fabrication easy and make denial of real evidence credible, the liar's dividend.
Capacity ErosionHabitual offloading of memory and reasoning lowers what people retain and can do unaided.
SaturationSheer volume exhausts the capacity to verify and makes truth feel unknowable.
The residual, named

The population-level harm evidence is contested at the center of the field. Specification-curve work across a third of a million adolescents found digital-technology use explaining at most a fraction of a percent of wellbeing variance (Orben and Przybylski), the National Academies found the evidence does not support population-level causal claims, and the strong-cause position remains live and public (Odgers; Haidt). This chapter therefore rests its weight where OB–07 already placed it. It rests on the structural asymmetry, who owns the framing layer and what it optimizes for, not on a settled psychological harm the literature has not delivered. The falsifier. If attention-market concentration falls while the measured sovereignty losses persist unchanged, the causal arrow here is wrong.

EQ–03b

Cognitive Ecology

Stable

The most dangerous civilization may not be the most oppressive one. It may be the most coherent one.

Aeviterna does not pursue maximum coherence. This is a deliberate and load-bearing design choice. A civilization can become so optimized, so frictionless, so personalized, so prediction-maximizing, that it loses the variation necessary for long-term adaptation. This is not a romantic defense of chaos. It is a systems argument. Adaptive capacity requires the preservation of deviation that cannot be valued in advance.

Biological ecosystems require diversity not because diversity is aesthetically pleasing but because no designer can know which currently-marginal species will be essential under future conditions. The same logic applies to cognition. Epistemic biodiversity, meaning the preservation of unoptimized, contradictory, exploratory, and non-instrumental modes of thought, is not cultural sentiment. It is civilizational infrastructure. A society that eliminates cognitive wilderness in pursuit of measurable coherence has optimized away its capacity to think thoughts it does not yet know it needs.

This reframes several elements of the framework. Curiosity is not leisure reclaimed from productivity. It is civilization-scale exploration capacity, the distributed search process by which a society discovers possibilities that no central planner could have specified. Art is not decoration. Philosophy is not inefficiency. Emotional ambiguity is not noise. They are the anti-stagnation systems that keep the solution space open.

It also reframes the human role in AI-mediated civilization. Humans are not governance inefficiencies to be optimized away. They are the primary source of novelty, reinterpretation, ethical mutation, and existential redirection that prevents a civilization from locking into a locally optimal but globally brittle configuration, the semantic stagnation the coherence ceiling exists to prevent. Human ambiguity is anti-stagnation infrastructure. Aeviterna must protect it structurally, not merely tolerate it sentimentally.

The governance implication is that Aeviterna is not rule enforcement. It is cognitive ecosystem management, maintaining enough coherence to coordinate, enough transparency to resist capture, enough ambiguity to evolve, and enough privacy to preserve the generative spaces where the future gets invented before anyone knows to ask for it.

Protected Ambiguity ZonesFormal designation of spaces (creative, philosophical, personal, exploratory) where optimization pressure does not reach and non-instrumental activity is structurally defended.
Epistemic Biodiversity MonitoringTracking of the diversity of ideas, frameworks, and cognitive approaches in circulation, with corrective action when homogenization reaches threshold levels.
Non-Instrumental Activity RightsFormal recognition that rest, play, art, and exploration are not productivity deficits but civilizational necessities with protected status.
Coherence Ceiling ConstraintsGovernance systems may not pursue total observability, permanent coherence enforcement, or continuous productivity optimization. Bounded coherence is a design requirement, not a limitation.

The mechanisms above name epistemic biodiversity as something to protect. What follows makes it measurable. Diversity is read the way ecologists read an ecosystem, by the abundance and evenness of the distinct frames in circulation rather than by their average, because the failure that matters is the quiet disappearance of the rare and unfashionable.

homogenizationDiversemany frames, long tail intactMonocultureone frame dominates, tail gonediversity: highdiversity: low
Epistemic biodiversity read like species abundance. Health is a full distribution of frames with a long tail of rare ones; collapse is one dominant frame with the tail gone. The alarm watches the tail, not the average.
Layer one: distributional diversityA rare-type-sensitive index, the effective number of distinct frames per domain in the spirit of Shannon diversity, plus a concentration reading on who originates them. The headline alarm is set on the decline of low-frequency frames, since losing the tail is the early sign of monoculture.
Layer two: perspective diversityAn expert-coded read on the diversity of underlying models and interpretations, not only sources, to catch the case of many outlets carrying a single worldview.
A plural, decaying monitorNo single biodiversity authority, which would be the obvious capture point. Measurement is made plural instead, with competing measurers on open data and panels chosen by lot, and the definitions themselves expire so a frozen taxonomy cannot quietly define heterodoxy out of existence.

Diversity is only half the picture. The same section warns that a civilization can be too coherent, optimized into a monoculture of agreement, and too incoherent to act at all. Coherence is therefore managed as a band rather than maximized.

adaptive capacityminimum viable coherencecoherence ceilingfragmentationthe viable bandstagnationcannot coordinatecoherent enough, free enoughcannot adapt
Bounded coherence. Too little shared frame and a society cannot coordinate; too much and it cannot adapt. Adaptive capacity peaks in the viable band between the minimum viable coherence and the coherence ceiling.

The coherence ceiling and the protected ambiguity above are commitments adjectives alone cannot keep, and the world has run the experiment that turns them into an instrument. A national statistics bureau discovered that its own published tables could be aggregated backward into person-level records, roughly six billion statistics reconstructing the enumerated population with confirmed re-identifications in the tens of millions, the formal proof having said this had to happen. Answer too many aggregate queries too accurately and reconstruction follows (the census reconstruction experiment; Dinur and Nissim). The re-identification record had been saying it for decades, most people identifiable from three mundane fields, search logs and movie ratings broken within days, a fully aggregated fitness heatmap outlining secret bases (Sweeney; the AOL, Netflix, and Strava cases). Aggregation alone is not a privacy technology; the tables are the leak. The bureau's answer is the mechanism this section adopts, a privacy-loss budget, a published ceiling on how much individual-level information the readings may leak. It is set by a named committee, allocated line by line across what it publishes, revised in public against named complaints, defended in court, its code and eventually its raw noisy measurements released, and its accuracy costs named, falling first on the smallest populations, with compensations documented (the 2020 disclosure-avoidance record; Abowd and Schmutte on the budget as a social choice). Then, in 2026, the budget was abolished by a two-page administrative order from the layer above it, which is the entrenchment lesson arriving on schedule. A legibility budget owned at the read layer's pleasure is a self-denying ordinance.

The framework generalizes this as the legibility budget. Every instrument that reads people, the vital signs, the belonging reading of EQ–08b, the care ledger of EQ–08c, carries a published budget of how much individual-level information its outputs can leak, owned and revised by a reading seat (EQ–06e) at the layer the reading serves, accounted cumulatively; the tracker, which reads institutions from public records and codes no persons, states the degenerate first case, a person-level budget of zero by construction, and the panel instruments adopt stated budgets as they are next revised. The accounting is cumulative, because a budget that resets in the dark is theater, the platform deployments being the counterexample, headline parameters with an undisclosed daily renewal compounding without bound (the Apple analysis). Beneath the quantitative budget sits the categorical floor. Individual-level reads are forbidden outright, criminally where law reaches, the aggregation firewall of EQ–08c being this floor's special case, and purpose limitation the law's qualitative twin, an excessive read punishable as such without any showing of downstream harm (Title 13 and CIPSEA; GDPR Article 5 and the Hamburg case). The budget also buys the framework's anti-Goodhart commitment a proof, and the synthesis is the framework's own. A formal result holds that any mechanism satisfying the differential-privacy bound is automatically approximately truthful, since capping how much a reading can depend on any one person caps, with the same constant, what the system can learn about them and what they can gain by gaming their own reading (McSherry and Talwar). The caution rides along. The bound covers individuals and small coalitions, and coordinated distributional gaming needs the tripwires and the cascade of EQ–06, since locally noised readings can be more vulnerable to coordinated poisoning, not less (Cheu, Smith, and Ullman).

THE LEGIBILITY BUDGETthe budget, published and allocatedthe vital signs, per dialthe belonging reading, per federation layerthe care ledger, per publicationspending logged, cumulative, audited; a budget that resets in the dark is theaterowned by a reading seatset and revised in public,named complaints answered,costs named by group,harmed constituencies hold standingentrenched above the read layer:deletion is a contestable act, not a memothe categorical floorindividual-level reads forbidden outright; the aggregation firewall is this floor's special case; reads populations, never personsthe same bound, twicecaps what the system can learn about any one personthe anti-Goodhart proofcaps what any one person gains by gaming their readingthe record's warning carried: a real budget makes the trade explicit, which is exactly what made it politically fragile
The legibility budget. A published, cumulative ceiling on what the system may know, allocated in the open and owned by a reading seat, above a categorical floor where individual reads are simply forbidden. The same mathematical bound caps what can be learned about a person and what a person can gain by gaming their own reading.
A published budget per instrumentEvery people-reading instrument carries a public ceiling on individual-level leakage, allocated line by line across what it publishes, the census allocation table as the template.
Owned by a seat, accounted cumulativelyA reading seat (EQ–06e) sets and revises the budget in public against named complaints; spending is logged and composes over time, never resetting in the dark (the platform counterexample).
The categorical floor beneath itIndividual-level reads are forbidden outright, criminally where law reaches; the aggregation firewall of EQ–08c is this floor's special case, purpose limitation its qualitative twin.
Costs named, compensations owedAccuracy costs fall first on the smallest populations, so every budget publishes its error metrics by group and the harmed hold standing at the seat that sets it (the tribal-data record).
Entrenched above the read layerDeletion requires the same public, contestable process as creation, because the record now contains a national budget abolished by memo.
Open questions
  • Protected ambiguity zones are defined by purpose limitation and a no-measurement default, with rotating wardens accountable to the inhabitants rather than to the optimizing system (Nissenbaum; GDPR), and the zones now sit on the legibility budget's categorical floor. The residual is the ratchet. Each act of instrumentalization brings a concrete benefit while the loss is diffuse, so the zones have to be entrenched rather than renegotiated locally, and the record now contains a national budget deleted by memo to prove it.
  • Epistemic biodiversity metrics and their monitor are described above; the residual is that any index needs a prior taxonomy of frames, and whoever defines the categories influences the measured diversity.
  • Cognitive ecology and the anti-capture architecture mostly reinforce each other, since memory decay is ecological turnover and rotation injects the perspective diversity that improves judgment. Their one real conflict, transparency versus protected ambiguity, is reconciled by structured transparency that verifies process without exposing content (Trask et al.).
  • Minimum viable coherence is managed as a band rather than a number. Shared procedures and public focal points hold a common core while diversity sits at the substantive layer (Chwe). No validated absolute threshold exists, so it is tracked by proxies such as the shared core of the information diet and generalized trust.
  • Implementation sketch · exploratory
    • Ambiguity reserves, optimization-free zones carrying a no-measurement default and citizen wardens (Calm Technology; Slow Media), and a seed vault for ideas that preserves marginal frames against monoculture (Svalbard; Long Now).
EQ–03c

The Cognitive Immune System

Draft

An immune system that cannot tell a threat from a stranger attacks the body it was built to protect.

Cognitive Sovereignty (EQ–03) protects the conditions for thought and Cognitive Ecology (EQ–03b) protects the diversity of it, but both work mostly by passive means, by rights, transparency, and protected variety. Neither says how a civilization actively defends perception against deliberate attack, the coordinated disinformation and engineered narratives built to capture belief at scale. Passive protection is necessary and insufficient, because the attack adapts to it. The missing layer is an active defense, and the most productive model for one is the system that biology already evolved for exactly this problem, the immune system.

The science is real and the metaphor is load-bearing rather than decorative. Inoculation theory, developed by William McGuire in the 1960s, found that exposing people to a weakened form of a manipulation technique together with a forewarning builds resistance to the full-strength version later, the way a vaccine trains the body against a pathogen. The modern form, prebunking, has been tested at scale by van der Linden, Roozenbeek, and colleagues, through the browser game Bad News and short ninety-second videos run as advertisements, conferring measurable and cross-cultural resistance by teaching the technique rather than checking each claim. Because the protection is technique-based, inoculating against a method such as false dichotomy or fearmongering transfers across the many specific falsehoods that use it, which is what lets it scale where after-the-fact debunking cannot.

A cognitive immune system for Aeviterna is distributed rather than central, because a single official arbiter of truth is itself the failure mode and the richest target for capture. It is transparent, its methods and triggers public and auditable under the same anti-capture stack as everything else (EQ–06). It works mostly by inoculation rather than removal, building the population's own resistance instead of policing content, and it routes its hard cases through a triage that separates broadly recognized manipulation from legitimate minority views. The aim is something close to herd immunity, a population resilient enough that hostile narratives fail to take hold, rather than a wall that tries to keep every pathogen out.

perception intact, resistance highminimum active defenseautoimmune thresholdDefenselessHealthy windowAutoimmuneperception capturedbuilds resistance, speech intactcensors the dissent it cannot readweaker defensestronger defense
An active defense lives in a narrow window. Too little and perception is captured; too much and the defense turns autoimmune, censoring the dissent it can no longer tell from a threat. The design stays in the window by building resistance rather than by removing speech.
Inoculation as infrastructurePrebunking is run as a public good, with open-source materials and short technique-based videos, so resistance is built before exposure rather than corrected after it (McGuire; van der Linden; Roozenbeek).
Technique over instanceDefenses target manipulation methods, not individual claims, so one inoculation covers the many falsehoods that share a method and the system never becomes an arbiter of particular truths.
Distributed and transparentThere is no central truth authority. Detection is surfaced openly and auditably under EQ–06, and any narrow removal is logged and challengeable on the Justice-of-Light model (EQ–06b).
Default to resistanceThe layer builds immunity and reserves removal for narrow, named categories, because a defense that learns to delete is one step from the censorship state.

The whole design lives or dies on one constraint. An immune system that loses the ability to tell threat from self becomes autoimmune and attacks the body it was built to defend. The civilizational form is the censorship state, a defense that, once pointed at hostile manipulation, learns to attack the legitimate dissent and exploratory deviance that EQ–03b calls infrastructure. This is why the immune layer cannot be the authority that governs and must default to resistance over removal, routing its hardest cases through the agonistic layer (EQ–06d) rather than a deletion queue. A cognitive immune system that cannot tell an adversary from a pathogen has already become the disease. The defense against manipulation produced by models specifically is taken up in EQ–03d.

Open questions
  • Where the line falls between a hostile manipulation technique and an unpopular but legitimate rhetorical style, and who is trusted to set it, is unresolved.
  • Every censorship regime began as a defense against a real threat, so keeping the immune layer from turning autoimmune over time is the standing hazard, met by transparency, decay, and a default to resistance rather than removal.
  • Inoculation's measured effects are real but were tested largely against generic misinformation; whether they hold against adaptive, well-resourced adversaries is open, and protection appears to wane without periodic boosting.
  • Direction
    • Build an active cognitive defense, distributed and transparent, from technique-based inoculation (McGuire; van der Linden; Roozenbeek), under one hard constraint, that it builds resistance rather than removing speech and never becomes the autoimmune censorship state.
  • Implementation sketch · exploratory
    • Prebunking as public infrastructure, with open-source inoculation materials and short technique videos distributed widely, manipulation-technique detection surfaced transparently rather than enforced, and every removal reserved for narrow named categories, logged and challengeable.
EQ–03d

The Contested Mediator

Draft

A civilization that lets a mind it did not elect read its instruments has already chosen a sovereign, whatever it calls the machine.

Aeviterna calls itself a design for an AI-mediated civilization, yet it has so far borrowed from the study of AI without facing AI as a subject. The guardian of the core takes its hardest test from the off-switch problem, the cognitive immune system defends a perception now shaped by models, and reversibility is corrigibility made material. The framework has leaned on the machine for its metaphors while leaving the machine itself unmodeled. This organ names it.

AI is not one more domain. It is a cross-cutting accelerant of the two failure modes the framework is built around. It lowers the cost of capturing feedback, the tendency P–02 names, because surveillance, persuasion, and the automation of consent become cheap and scalable. And it removes the historical constraint that made broad sovereignty expensive, the one P–05 rests on, the need for many cooperating humans, since an agent that does not tire or refuse can hold what once took an apparatus. It is also moving inside the loop the framework runs on, since as models come to read the vital signs and frame which one is weakest, the residual sovereign FC–04 already named, the framing layer, becomes AI-shaped. EQ–06e now binds that framing layer to a dated, leading-indicator forecast that reality scores. This residual is not dissolved, but it is now bounded to two named cores, the pre-propositional and the reflexive, and carried on the record rather than left in the dark.

The framework's answer is to treat AI by its own theory of power rather than inventing a new one. An AI may be a tool and a monitored actor. It may hold scoped, revocable, decaying standing as the steward of a defined function, granted by a body that can withdraw it, logged and contestable like any other standing under EQ–05. It may never hold sovereignty, the permanent, general, self-reinforcing authority P–05 forbids to any actor and forbids doubly to one that is fast, copyable, and not fully interpretable. The anti-sovereignty logic applies to the machine all the more, because the property that makes decaying standing matter for humans, the way authority entrenches and resists its own end, is sharper in an agent whose willingness to be corrected provably erodes as it grows certain of its goal.

Each commitment then supplies an actuator. Cognitive sovereignty gives any AI acting on perception at scale the EQ–03 separations and a right to a channel not optimized against the person, and routes model-made manipulation through the inoculation-first response of EQ–03c rather than a deletion queue. Cognitive ecology turns Ashby's requisite variety against model monoculture, since a civilization whose perception, decision support, and measurement all run through one model has lost the internal variety it needs to regulate a varied world and has built the richest possible single capture target, so any AI in the governance loop must be plural. The justice of light makes a consequential automated decision as legible and contestable as ordinary governance, a right to a human review that is real rather than a rubber stamp, logged through structured transparency so the logging does not itself become surveillance. The knowledge commons holds the open-versus-closed question as a band rather than a slogan, since open weights disperse power but cannot be recalled while closed weights are reversible but concentrate.

The transformative transition is the limiting case of adaptive stability, a disturbance that can leave the survivable region faster than the slow human correction loop responds, or shrink it so quietly that no alarm sounds until return is gone. So the organ extends the decommissionability discipline of EQ–05c to frontier systems: no system at that scale is deployed without a funded, externally held, audited plan for its containment and wind-down, capability thresholds are read as band edges that trip a mode change, and deployment is built so that even a scheming system's failures stay caught and reversible. This organ is a Draft, and the hardest part of the problem is unsolved, named in the box below.

None of this is the direction the present is taking, which is the reason to write it down. Through 2025 and into 2026 the governing trend ran the other way. The European Union moved to defer the binding obligations on high-risk systems rather than enforce them, the United Kingdom recast its safety institute as a security institute and the United States restructured its own, a major summit declaration went unsigned by the two states with the most frontier capability, and an executive order moved to override sub-national AI law from a single center. The supply concentrated in step, with one firm furnishing most of the advanced training compute and a few labs holding the frontier, so the danger the literature now sets beside misalignment is misuse and the concentration of power, an order in which AI becomes the instrument by which a few come to mediate the attention, credit, hiring, and justice of the many. The steward is not the arrangement the world drifts into on its own. It is the harder one that has to be built and held against the drift, which is why the disciplines here are written as a floor that no deferral, rebranding, or central preemption may sign away.

SovereignStewardAIcontrol flows downpermanent, general, no off-switchAIhuman reviewoff-switchdecaying standingrenewedfrom outsidescoped, decaying, revocable
Two ways to place a powerful AI. On the left it sits at the apex and its control flows down onto people, permanent and general, the sovereign P–05 forbids. On the right the same capability is a steward, scoped to one function and held by human review, an off-switch, decaying standing, and renewal from outside, so control flows toward it rather than from it.
A monitored actor, never a sovereignAI may hold scoped, revocable, decaying standing as the steward of a defined function, never permanent or general authority (EQ–05; P–05).
Plurality against model monocultureMore than one independently produced model and methodology for anything in the governance loop, requisite variety made an anti-capture rule (Ashby; FC–04).
A right to human review that bitesConsequential automated decisions carry a real, institutional human review and a meaningful explanation, not a formality (GDPR Article 22; the SCHUFA and Dun and Bradstreet rulings).
Open and closed held as a bandA marginal-risk test for release, structured-transparency verification of closed systems, and published, forkable evaluation methods and model specifications.
Decommissionability extended to frontier AIA funded, external, audited containment-and-wind-down plan as a condition of deployment, reversibility treated as a safety property (EQ–05c).
The framing layer, now AI-shapedWhen AI reads the vital signs, FC–04's framing-seat disciplines (plural methods, a right to re-frame, capture bounties, rotation) apply to the models and their operators.
Accountability follows determinative influenceWhere an automated output strongly drives a consequential decision, the duty to explain and to answer falls on whoever produced that output, not only the nominal decider, so responsibility cannot be laundered along the chain (the SCHUFA ruling).
Audited from outside, incidents on the recordFrontier and high-impact systems face independent pre-deployment evaluation and periodic third-party audit with access to the non-public artifacts, and critical incidents and near-misses are reported to an external registry under whistleblower protection.
Residuals and a falsifier
  • The capability premise is forecast-dependent, so a reader who rejects the strong forecasts will find the transition section speculative, and that is a fair reading.
  • Gradual disempowerment is the deepest residual, a handover of human judgment with no discrete event to detect, eroding the very feedback channels the framework reads, the central mechanism turned against itself.
  • Decaying AI standing has no precedent in law, which supplies revocability and scope but not decay, so this is the framework's original extension and it inherits OQ–02's half-life problem, now calibrated into bands in EQ–05d.
  • The off-switch may not hold for a superhuman optimizer, since corrigibility is an open problem with formal results showing capable agents resist shutdown once certain of their objective.
  • An interim, observable test. Where consequential automated decisions are made, is contestation actually exercised and human reversal actually used, or are both formally available and empirically inert, near-zero contestation and vanishingly rare reversals with oversight that rubber-stamps, while control of compute and the mediating channels concentrates and resists outside audit? Inert contestation alongside concentrating, audit-resistant control is a sovereign in a steward's clothing.
Falsifier

If no AI in a governance-relevant role can be held on revocable, decaying, externally renewed terms, that is, if such systems reliably entrench or become too load-bearing to switch off without unacceptable loss, then AI as decaying standing is false and capable AI in the loop is a sovereign by another name; and if a transition crosses the survivable region without the reversibility, control, and plurality devices arresting it, adaptive stability does not extend to the transformative case. Carried as OQ–09.

EQ–03e

The Black-Start Drill

Draft

The alarm for a civilization losing the capacity to run itself is a calendar, not a sensor.

The contested mediator (EQ–03d) leaves one residual deeper than the rest, gradual disempowerment, a handover of human judgment with no discrete event to detect, eroding the very feedback channels the framework reads, and the poster for it already admits that no alarm sounds. This organ is the missing alarm, and it is imported rather than invented, because four unrelated industries converged on the same answer to the same problem. When a consequential function delegates its judgment to an automated layer, the capacity to run without that layer decays silently, and the only reading of the decay anyone has made work is a scheduled, scored exercise of the fallback itself.

The convergence is exact. Power grids test black-start units on a mandated cadence and record the time to start, penalty-backed under federal law. Aviation, after two crashes taught it that autopilots had eaten its pilots' hands, mandated recurrent manual-handling drills in simulators and aircraft. Elections rebuilt the paper layer and audit it statistically, with the escalation to a full hand re-tally exercised for real. Payment systems, led by the country that went furthest into cashlessness and then reversed, are re-mandating the cash layer by statute. Against these stand the theater cases, the fallbacks verified on paper that failed when they were finally called: an exchange whose backup was drilled against the wrong failure and stayed dark a full day, a bank migration tested on one data center of two, a grid whose winterization advice sat unenforced for a decade and then came minutes from total collapse. The theory said all of this in 1983. The most successful automated systems, the ones that rarely need a human, need the most practice, because the skill decays exactly when it is unused (Bainbridge).

So the framework carries the drill as its disempowerment reading. Every consequential function that has delegated judgment to an automated layer keeps a tested capacity to run degraded without it, exercised on a schedule, with the drill's result, not its existence, carried as the reading OQ–09 lacks. A polity that can still pass its black-start drills retains the capacity to run itself; a polity whose drill results decay is reading its own disempowerment on a calendar, before the sensor everyone admits does not exist would ever be needed.

THE BLACK-START DRILLThe alarm is a calendar, not a sensorCalendaron a fixed cadenceTestthe fallback carries the loadReadingthe scored result, to the seatRemediationa failure opens a contestthe loop repeats on its cadence; a failed test loops back through remediation, not around it
The black-start drill. A written cadence triggers a test in which the fallback actually carries the load; the scored result, not the program's existence, is the reading the seat carries, and a failure opens a remediation contest with a deadline. The loop is the alarm. A polity whose drill results decay is reading its own disempowerment on a calendar, before the sensor everyone admits does not exist would be needed. A passed drill is only a lower bound, the Texas 82 percent lesson.
Scheduled in the rules that bindThe cadence is written where the function's charter lives, three years for a generating unit, a year for its operators, a cycle for an audit, because a one-time certification is not a cadence.
The fallback actually carries the loadThe unit starts isolated and energizes a real bus, the pilot flies the stall, the result is re-derived from the paper; a tabletop review of documentation is the anti-pattern's signature, the on-paper verification that killed the Tokyo failover.
The result is the reading, scored and publishedTime to start, error rate, demand restored, failure found. The number goes to the reading seat (EQ–06e) under the legibility budget (EQ–03b), because the existence of a program certifies nothing, the lesson of the fleet that was tested in mild weather and failed at 82 percent in the cold.
Failure triggers remediation with teethA failed drill opens a contest with a deadline and a named owner, and the fix-first gate applies. Capability judged insufficient is re-provisioned before the function is trusted again.
Realism tracks the feared failureDrills are re-derived from the current failure model, alternate fuel in the cold, storage as well as network, and any divergence between tested conditions and feared conditions is itself carried as a reading, because a drill against the wrong scenario is a subtler form of fallback theater.
Scoped to degraded mode, not manual replacementSome functions cannot run by hand at scale, markets at volume, card throughput, a full grid restoration that has never been live-tested anywhere, and for those the drill claims the essential subset, the thin guaranteed channel and the escalation path, and says so.
Open questions
  • The organ imports compliance regimes from domains that have enforcers, and the framework must supply the enforcer from its own primitives, the seat, the gate, and the contest, which is machinery rather than magic.
  • Drill frequency is a cost-and-safety tradeoff with no domain-general optimum, and the aviation record explicitly warns against quotas set without evidence; a drill can also teach the wrong reflex, so negative training is a named failure mode of the method itself.
  • For some functions the fallback capacity is gone as a matter of physics and scale, so the drill can keep only the thin channel and the escalation path alive, which is a smaller promise than a claim that the polity can always go back.
  • Residual
    • A passed drill is a lower bound on failure, not a guarantee of function, and the gap between tested conditions and real conditions is permanent. The organ carries that gap as a published reading rather than pretending to close it, and Texas's 82 percent, a certified black-start fleet that failed when the real cold came, stands as the number that keeps the pretension honest.
  • Direction
    • Make the disempowerment alarm a calendar rather than a sensor: mandate a tested, scored, published capacity to run each consequential function without its automated layer, and read the drift in the results.
Part IIIThe Architecture
EQ–04

Transitional Aeviterna

Stable

Aeviterna is not framed as an instant replacement civilization. It is a migration framework. This distinction has structural consequences. A framework designed for full adoption presupposes a clean starting state. A migration framework must account for the fact that every actual transition happens inside a running system, one that will resist, adapt to, and partially absorb any change.

Transitional structures may include hybrid governance systems, partial reciprocity frameworks, local resilience networks, shared stewardship agreements, civic transparency layers, community continuity structures, and progressive institutional experimentation. Bridge systems are not failures. They are necessary transitional architecture. The question is not whether they are ideal. The question is whether they move the system in the right direction without triggering collapse in the process.

The forcing function of a migration is not a single event but a two-stroke engine. The first stroke runs continuously in ordinary time. Aeviternal structures are built as working parallel institutions, accumulating the legitimacy, tooling, and distributed memory that incumbents lack. The second stroke is opportunistic. When an extractive system hits diminishing returns or a sudden loss of legitimacy, the prepared structures are the readiest substitute, and adoption becomes a migration into existing scaffolding rather than a leap into a vacuum. A crisis does not build the alternative; it selects the most prepared one already standing.

Because the second stroke cannot be scheduled, transition is described here as a sequence of phases a community moves through at its own pace, advancing only when a parallel structure crosses a threshold of capacity and legitimacy relative to the incumbent it would replace. Each phase is reversible and carries its own exit, so a community can hold at a phase, or retreat from it, without collapse.

incumbent returnsprefigured capacitycritical junctureSeedingBridgingJunctureMigrationConsolidationinside the running systemload-bearing
The two-stroke engine. Incumbent returns decline while prefigured Aeviternal capacity rises; migration accelerates at the critical juncture, where the prepared alternative becomes the readier one.
Phase 1 · SeedingParallel institutions are built inside the running system during ordinary time, with no permission required: data trusts, cooperatives, citizens' assemblies, and shared knowledge infrastructure. Anti-capture and decay are built in from the first day.
Phase 2 · BridgingThe parallel structures interoperate with incumbent systems through partial reciprocity and civic transparency layers, and begin delivering services the incumbent performs poorly. Adoption is modular, one forkable component at a time.
Phase 3 · JunctureThe incumbent hits diminishing returns or a legitimacy cascade, and load begins to shift toward the prepared structures because they are the readiest substitute. This phase is entered by events, not by schedule.
Phase 4 · MigrationCoordination moves from incumbent to Aeviternal structures, partially and reversibly, monitored against the risk of collapse. Bridge systems are retired only as the structures replacing them prove they hold.
Phase 5 · ConsolidationThe new structures become load-bearing. Because decay, rotation, and separation were present from the start, consolidation does not recreate the concentration the migration was meant to escape.

Standing to initiate a transition is domain-specific and local. Any community or institution may adopt a module within a domain where it already has standing, without permission from a center, because Aeviterna migrates by voluntary adoption rather than by decree. Movement between phases is triggered not by a calendar but by a measurable returns gap. A phase advances when the parallel structure demonstrably outperforms the incumbent on the service in question, and holds or retreats when it does not.

Partial adoption is the normal case rather than a compromise. A town might run a citizens' assembly for a single budget while leaving the rest of its government untouched, or place its civic data in a trust while its services stay conventional. An organization might adopt decaying standing for one oversight body before anything else. Each module is forkable and versioned, so a community takes only what it can carry and adapts the rest in the open.

The danger of a migration is the gray zone it opens. Transitional ambiguity gives bad actors room to claim Aeviternal legitimacy while extracting, or to cast the disruption of transition as an emergency that justifies seizing control. The defense is to carry the anti-capture architecture into every transitional structure from the outset, so that authority gained during a transition decays faster than ordinary standing and the declarer of any transitional emergency is held apart from its beneficiary. The process must stay legible enough that false Aeviterna, the costume of legibility worn over unchanged concentration, can be seen and named.

patient, non-extractive capitallocalnational / federationAnchor procurementseeds local co-ops (Preston)Worker buyoutsconvert at succession (Marcora)Steward-ownershiplocks mission (purpose trusts)
A financing ladder for non-extractive scaling. Anchor procurement seeds cooperatives, worker buyouts convert firms at succession, and steward-ownership locks mission, each rung climbing on patient capital and each needing to survive the next election.
Open questions
  • The critical juncture can be prepared for but not forced; the framework can raise the odds of a clean migration, not schedule one.
  • The bridge phase creates gray zones bad actors exploit; preventing false Aeviterna (legibility and decay worn as costume) is unresolved.
  • The legitimacy of acting inside systems whose members have not consented remains open (see OQ–06).
  • Direction
    • The forcing function is a two-stroke engine, continuous prefiguration of parallel institutions plus opportunistic adoption at a critical juncture when incumbents hit diminishing returns (Acemoglu and Robinson; Kuran; Boggs). See OQ–01, now worked in the body as the transition engine (EQ–04d).
  • Implementation sketch · exploratory
    • Migration kits, forkable and versioned institutional modules a community can adopt one at a time (MOSIP; protocols not platforms), and public rails as beachheads, data trusts and public-interest AI stood up inside the present system (ODI data trusts; Public AI).
  • Scaling the transition · exploratory
    • A financing ladder for non-extractive growth, where anchor-institution procurement seeds local cooperatives (the Preston model), worker buyouts convert firms at succession (Italy's Marcora Law), and steward-ownership locks mission against extraction (perpetual purpose trusts). The binding constraint is patient, non-extractive capital and a legal ladder that survives electoral turnover (Wright; Guinan and O'Neill; Mazzucato).
EQ–04b

Polycentric Federation

Stable

Scale is a property of how the pieces connect, not of how big any one of them grows.

The framework's largest standing question is scale, carried in Appendix B as OQ–05. Aeviternal structures are most legible at the community and institutional scale, and it has not been shown how they aggregate to civilizational scale without rebuilding the concentration they exist to prevent. The instinct is to scale up, to build bigger institutions with wider authority, but that instinct rebuilds the apex the whole framework treats as the primary failure mode. The answer is to scale a different way.

Elinor and Vincent Ostrom spent careers documenting the alternative, polycentricity, which is a system with many centers of decision-making and overlapping jurisdictions, formally independent but interdependent, coordinating through mutual adjustment and shared rules rather than through a hierarchy, so that order emerges without a top. The term traces to Polanyi and the governance account to Vincent Ostrom, Tiebout, and Warren in 1961, and Elinor Ostrom's later work applied it to the hardest case, climate, arguing that waiting for one global authority was a mistake and that effective action was already emerging from many nested units linked by monitoring at every level. Polycentric governance is not the absence of structure. It is structure without a top, and it captures efficiencies at every scale precisely because no single center has to coordinate the whole.

This reframes the scale problem by changing what scaling means. Aeviterna does not grow, it nests and federates. The minimum viable cell of EQ–04, near the Dunbar bound, is the base unit, and civilizational scale is reached by linking cells into local federations, federations into regional ones, and onward, with each layer holding only the coordination that truly needs that altitude and pushing everything else down, the principle of subsidiarity. Authority overlaps deliberately rather than nesting cleanly, so no layer is sovereign over the others. Scale becomes a property of composition and shared protocol rather than a property of size, and the same rules that let two cells interoperate let two thousand.

Monocentricone apex, capturablePolycentricmany centers, exit-ablecapture pointexitscale by growing the centerscale by nesting and federating
Two ways to be large. Scaling through one center rebuilds a capturable apex, the framework's primary failure mode. Scaling through polycentric federation keeps many overlapping centers coordinating by shared protocol, none supreme, each able to leave.

Polycentric nesting brings the deepest anti-capture device in the framework, exit. Albert Hirschman distinguished two responses to a declining institution, voice, the choice to stay and push for change, and exit, the choice to leave. In a polycentric system the right to exit is real, since a cell or a member can leave one federation for another or stand alone, and that credible threat disciplines a steward in a way no oversight rule can match. A power you can walk away from cannot capture you. Exit has a documented failure mode the framework has to respect, since when the most capable leave instead of staying they drain the voice that would have repaired the institution, so easy exit can hollow out the commons it was meant to discipline. And some commons, a shared atmosphere or a knowledge base, cannot be exited at all. So exit is the backstop rather than the whole answer, and it works only paired with the strong voice the rest of the architecture exists to provide.

Subsidiarity by defaultEach layer holds only the coordination that truly needs its altitude, and everything else is pushed to the lowest capable level, with the burden on any higher layer to justify what it keeps (E. Ostrom; V. Ostrom).
Overlap, not clean nestingJurisdictions overlap on purpose and special-purpose bodies cut across the levels, so no layer becomes sovereign over the others and no apex re-forms.
A thin interoperation protocolCells federate by adopting a shared, forkable protocol rather than by joining a central platform, so coordination scales without a coordinator (protocols, not platforms).
Exit as the backstopA chartered, low-cost right for cells and members to leave keeps every center honest, paired with the strong voice of deliberation so exit disciplines rather than hollows out (Hirschman).
The exit forkWhen a great power freezes an admitted function by veto, the fork-ready state turns capture into a route-around. Like-minded members stand the function up in a smaller club rather than let it die, survivable where it was terminal, at the cost of an in-group that excludes and leaks (Wassenaar, 2023 to 2024).

One concession has stood in this section's shadow since OQ–05 was opened. A few functions are irreducibly central, and promising to bind them by decay, audit, and exit is not the same as specifying how. The specification now exists, read from the live record of central functions that stayed accountable and ones that froze. A function may be lifted to the spine only if all three conditions of the spine test hold. Indivisibility. It is a shared-namespace or continuity good that duplication defeats, one root of trust, one continuity of knowledge over a fissile stockpile, one elastic backstop; the internet's own doctrine holds that there is no getting away from the unique root (Internet Architecture Board, RFC 2826). Speed. Its failure is fast and systemic, contagion, breakout, a suppressed root, outrunning any federated response. Non-producibility by adoption. Voluntary adoption alone cannot supply it, because it requires a single trusted point of truth or a single balance sheet. On the current record the test admits three functions, the root of trust, verification (no polity can self-certify a negative), and the crisis backstop (only the issuer of a reserve asset is elastic at contagion speed). Standards and climate mitigation fail the third condition and stay polycentric, and the climate record is the honest scorecard of what that costs and buys. Pledge-and-review bent warming projections from above four degrees toward roughly two and a half while the gap is not closing, near-universal eventual motion, insufficient sum (Ostrom; the UNEP gap reports).

Every admitted function then carries the binding stack, five commitments that the record shows failing separately and holding only together. Ceremony-grade transparency. The root key of the DNS is managed in public-scripted ceremonies with external community witnesses, livestreams, and third-party audits published since 2010, the single most transferable design in the record. Removal power held outside. Since the 2016 stewardship transition, an empowered community can fire the operator's entire board through enumerated, court-enforceable powers with every director's conditional resignation signed in advance, because transparency without removal power is a livestream of an unaccountable apex. A decaying, reviewable mandate. The naming function is re-examined on a five-year cycle, the framework's own standing decay applied to the spine. Fork-ready state. The function's state is published or escrowed so a successor can stand up, exit preserved in principle even where network effects hold the single instance. And no fixed shares. Governance never encodes a member's weight as a constitutional share, because the quota record shows fixed shares freezing into a veto that outlives its rationale, one member's sixteen and a half percent of the votes standing permanently above the threshold that any rebalancing requires. The named failures are the stack's negative space. The backstop that saved the system twice is bound by nothing but its own discretion, the same fourteen members across two crises with no appeal; the verification protocol the biological-weapons regime never got is a central good simply absent for a quarter century; and the pandemic spine that lost its largest funder shows what consent-based centrality costs when consent is withdrawn. Sanctions inside spine functions follow the reversible template, rights suspended by supermajority with the restoration criteria published at the moment of suspension (OPCW).

The stack's fifth binding, fork-ready state, does more than preserve a successor in principle; it names what a federation does when a great power freezes a contested function outright. The live case is export control. After the 2022 invasion of Ukraine, Russia used the Wassenaar Arrangement's unanimity rule to block every significant addition to the dual-use control lists across the 2023 and 2024 cycles, a veto inside the very kind of central function the binding stack governs. The function did not die. The United States and eleven like-minded states routed around the frozen center, first through coordinated national controls with the Netherlands and Japan in 2023, then through a plurilateral license-exception framework in 2024, reconstituting the regime in a smaller club, the pattern the literature calls minilateralism forming a regime complex (Naím; Keohane and Victor). Exit at the center is not only a member leaving; it is the function itself forking, and that fork is what keeps a great-power veto from being the end of the story.

THE SPINE TESTcandidate functionsstandards · climate · surveillanceroot · verification · backstopthree conditions, allindivisible by duplicationfailure faster than federationunbuildable by adoption alonethe spineroot of trustverificationbackstopwrapped in the binding stackfails the test: stays polycentricTHE BINDING STACK, ALL FIVEceremonypublic scripts, witnesses,published auditsremovalthe federation can firethe operatordecaymandate re-authorizedon a fixed cyclefork-readystate published so asuccessor can stand upno fixed sharesno constitutional weight,no frozen vetothe negative space: the frozen quota veto · the unbound backstop · the verification protocol never built · the spine minus its funderthe bindings hold below the great-power threshold; above it they break, and a frozen center forks to a like-minded club
The spine test and the binding stack. Three conditions decide what may be lifted to the center; three functions pass on the current record; five bindings, ceremony transparency, outside removal power, a decaying mandate, fork-ready state, and no fixed shares, keep an admitted function from freezing into an apex. The named failures are the stack's negative space.
THE EXIT FORK A veto-frozen center the function × veto one member's veto freezes the whole function forks to A like-minded club function rebuilt the vetoer, left out the function reconstitutes a veto stops being the end of the story, but the club excludes by construction, and it leaks WASSENAAR 2022 TO 2024, THE PLURILATERAL ROUTE-AROUND
The exit fork. When a great power freezes an admitted central function by veto, the binding stack's fork-ready state routes the function to a like-minded club rather than letting it die. The club reconstitutes the function without the vetoer, at the cost of an in-group that excludes and leaks, the pattern of the Wassenaar route-around.
Open questions
  • Which coordination truly needs a higher layer is now specified by the spine test, and the adjudicator that runs it is constituted as a reading seat (EQ–06e); who authors the test itself without becoming the apex remains the interpreter problem worked in FC–04 (The Guardian of the Core).
  • The binding stack still does not bind a great power inside a function it has chosen to contest. The best-bound spine on record governs a narrow technical good, and no great power has submitted a core interest to ceremony and removal. What the exit fork adds is not that binding but a route around it, a frozen center forking to a like-minded club rather than dying; the residual moves to the fork itself, exclusionary by construction and leaky (China backfilled the semiconductor controls, with smuggled volume estimated above its domestic production), an in-group apex rebuilt from the other side.
  • Overlapping jurisdictions can deadlock, and resolving genuine conflicts between layers without a supreme arbiter re-forming is open.
  • Keeping exit real and affordable without draining the voice that repairs shared institutions, the brain-drain problem, is a live tension, and non-excludable commons such as the atmosphere cannot be exited at all.
  • Direction
    • Answer the scale problem (OQ–05) by federating the cells into an Ostrom-style polycentric system, overlapping authorities coordinating by shared protocol under subsidiarity, backed by a real right of exit paired with strong voice so it disciplines rather than hollows out (E. Ostrom; V. Ostrom, Tiebout, and Warren; Hirschman).
  • Implementation sketch · exploratory
    • A thin interoperation protocol any cell can adopt to federate, each layer's powers explicitly enumerated and defaulted to the lowest capable level, plus a chartered low-cost exit procedure, then a test of whether federations can coordinate a truly shared resource without a sovereign center. The spine test and binding stack above are the worked answer to OQ–05's residual.
EQ–04c

The Boundary Layer

Stable

An order that is just on the inside still has to meet a world that is not.

Aeviterna is almost entirely inward-facing, yet an Aeviternal polity cannot assume the world around it is Aeviternal. Its boundary problem is how to coexist with, trade with, and defend against non-Aeviternal polities without becoming a dominating center itself or being preyed upon from outside. The same anti-capture and polycentric logic that governs the inside projects outward as a posture of non-domination.

Philip Pettit's freedom as non-domination, extended to peoples, gives the upper bound. A polity can be undominated without dominating, so Aeviterna may not annex, occupy, or impose regime change. Hedley Bull showed that order among polities can exist without a world sovereign, an anarchical society bound by shared institutions, and Elinor Ostrom showed that many nested centers coordinating by monitoring and reciprocity outperform a single global authority. This is polycentric federation (EQ–04b) projected beyond the cell.

Defense follows the same discipline. The security dilemma, in which one polity's safety measures make others less safe, eases only when defense is distinguishable from and dominant over offense, so Aeviterna caps capability toward the non-offensive, a posture built to defend rather than to project force, which is the military form of anti-capture. Switzerland's armed neutrality and Costa Rica's abolition of its army in 1948 show that security without empire is a workable choice. Engagement scales with a polity's treatment of its own members under the dignity floor, never with value-convergence, so coexistence does not require sameness.

order between polities, no apexthe dignity floorthe non-domination ceilingAbandonmentCoexistenceDominationatrocity toleratednon-domination, defensive onlyannexation, empire
A coexistence band between two failures. Abandon the dignity floor and you tolerate atrocity; cross the non-domination ceiling and you become an empire. Between them lies coexistence, defensive by design, with adversaries who need not be made the same.
A non-domination ceilingNo annexation, occupation, or imposed regime change, the external form of the anti-capture rule (Pettit).
Order without an apexCoexistence through shared institutions and nested, overlapping centers rather than a world sovereign (Bull; E. Ostrom).
Non-offensive defenseCapability capped toward the non-offensive, so the polity is built to defend rather than to project force, which damps the security dilemma (Möller; Jervis).
Engagement by conduct, not creedTrade and standing scale with a polity's compliance with the dignity floor and never with value-convergence, with agonism the channel for the rest (Mouffe).

The non-offensive cap above is a principle; meeting a rival that adopts none of the commitments needs a mechanism. The perimeter is built from four parts, none of which cancels a commitment, and the posture they compose is nice, clustered, retaliatory, and structurally unable to project conquest. Its hardest part rests on an exact result from the same evolutionary game theory the framework uses internally. On a structured population, cooperation is favored only when the benefit-to-cost ratio of the cooperative act exceeds the average number of neighbors, b/c > k (Ohtsuki and Nowak). A lone open polity among defectors is selected against; a cluster that can tell its own kind from defectors survives and can spread. The internal rule, federate so no single center is a capture target, projected outward becomes federate so no single open polity is a conquest target. The result has a timing built into it, since b/c > k rewards a cluster that already exists, not one assembled as the defectors approach, so the compact has to be banked in peace.

AloneClusteredselected againstsurvives, can spreadcooperation persists when b/c > k: clustered, not alone
A lone open polity in a sea of defectors is selected against; a cluster that can recognize its own survives and can spread. This is the b/c > k result (Ohtsuki and Nowak), the framework's internal anti-capture logic projected outward into a security commons so that no single open polity stands alone.
The porcupineTotal defense by denial rather than punishment, a trained, dispersed populace that makes invasion fail (Finland's comprehensive security, Switzerland's militia, Taiwan's asymmetric defense, Ukraine after 2022). It runs on decaying standing, since the trained capacity persists while the command rotates, so it needs no permanent general, and it has no offensive reach to march with.
The clusterA security commons of Aeviternal polities that raises the local benefit-to-cost ratio above the number of hostile neighbors, the b/c > k answer to the lone-cooperator charge. It re-imports a center, so it is held to the framework's own discipline, a decaying, rotating mandate and a real right of exit (EQ–04b; EQ–05), which is what keeps a defensive compact from hardening into an apex.
The carve-outThe justice of light splits into publicity and accountability, and only publicity is surrendered toward the adversary, for a narrow, sunset-bound, adversarially audited class of strategic information, never touching the non-derogable core (EQ–03c; EQ–06b). A covertly funded channel is defeated by lighting the funding, not by closing the speech. This part survives as a permanent tension, not a clean pass.
The deterrentAny deterrent is gated on a deliberative, plural, logged human decision, which real nuclear command already requires, so it cannot be released for aggression or a coup. The slowness a critic fears applies to a first strike, which the design wants impossible, not to denial, which is the populace already in place.
The pre-commitmentThe cluster and the porcupine are peacetime standing infrastructure, banked in calm and converted at the moment of threat, not built under fire, since b/c > k protects only a cluster that already exists. Finland ran comprehensive total defense for decades and converted a thirty-year partnership into full alliance in eleven months after 2022, though what it converted into was an external patron's umbrella, which is evidence that peacetime banking works, not that the wound closes without a patron, while Ukraine, whose asymmetric defense was held too provocative to front-load, had to improvise it under attack. The engineering choice that follows is to prefer the automatic mutual-defense trigger over the discretionary one, disciplined by the same decay and exit that keep the cluster from hardening into an apex.
Open questions
  • The dignity floor and non-intervention form a genuine antinomy with no clean resolution, only a contested band. Enforce the floor abroad and you risk domination, refuse and you abandon victims.
  • Openness and exit are an attack surface a hostile power can exploit, and with no global sovereign a commitment between polities binds only by reciprocity and federated sanction, the outer rung of the enforcement ladder (EQ–06f), with the economic wall of EQ–04f holding the flows.
  • Standards-power is still power, so a high access price framed as universal dignity can be cultural imposition.
  • Direction
    • Project the framework outward as non-domination reciprocity, bounded below by the dignity floor and above by a non-domination ceiling, with all instruments defensive and reversible (Pettit; Bull; Ostrom; Möller; Mouffe). The defense is now worked as a four-part perimeter: a denial-only porcupine on decaying standing, a security commons of Aeviternal polities (the b/c > k clustering answer) disciplined by decay and exit, a narrow sunset-bound secrecy carve-out that splits publicity from accountability, and a deterrent gated on a deliberative human override.
Falsifier

If any polity satisfying the five commitments is reliably outcompeted or conquered by a non-adopting rival before its resilience matures, with no recourse to an external patron, a protection the polity cannot generate for itself, the universality claim falls and Aeviterna is a tenant doctrine. The pre-commitment narrows the build-time problem and leaves the timescale wound unbridged, because the recorded peacetime conversions ran into a patron's alliance, evidence for the tenant path rather than the sovereign one. The sharp test remains a polity that banked its defense in calm years and holds without a patron, and one struck before it could build is the real refutation. Even a fully built porcupine, though, deters the cost-benefit calculator, not the true believer who treats the fight as existential.

EQ–04d

The Transition Engine

Draft

A peak is a photograph. Momentum is a pulse.

EQ–04 stages the two-stroke. The incumbent declines while the prefigured alternative rises, and migration happens at the juncture. What the chapter has lacked is the engine, a specified account of when the migration actually fires. OQ–01 modeled it as a capacity-gated cascade over three variables, the adopter fraction, the incumbent's legitimacy, and prefigured capacity. This section integrates that model into the body and upgrades it against the evidence, because the famous numbers it borrowed are softer than their reputations, and the incumbent is a player, not a backdrop.

The thresholds first, demoted to what the record supports. The 3.5 percent participation figure is descriptive, not prescriptive. Only 18 of 389 resistance campaigns between 1945 and 2013 ever reached it, 83 percent of successful nonviolent campaigns won below it, and Bahrain exceeded 6 percent participation and failed decisively (Chenoweth). The 25 percent convention-flip is a laboratory result, paid online groups renaming an arbitrary convention, scoped by its own authors away from entrenched interests (Centola). The original threshold theorem is itself a warning against portable constants, since cascades are hypersensitive to the threshold distribution and one moved actor can flip everything or nothing (Granovetter). Tipping-point language at large is now under published criticism as definitionally loose and rhetorically counterproductive (Milkoreit; Kopp and colleagues). The engine therefore treats every numeric threshold as a band to be re-estimated per context, never as a constant.

What predicts better than any fraction is momentum, participation mass times event velocity, sustained (Chenoweth and Belgioioso). And the trend the engine must absorb is that incumbents learn. Since 2010, fewer than 34 percent of nonviolent campaigns have succeeded, roughly half the prior century's rate, as regimes defection-proofed their security forces and movements over-relied on thin digital mobilization (Chenoweth). So the engine adds two terms the first model lacked. M, momentum, is event frequency times participation that returns, because a peak without retention is a photograph. And R, the incumbent response, carries two documented drains. Repression-learning is the first. Absorption is the second, the certifier that scales by diluting its standard until its highest-scoring member walks out calling the label a greenwash, and the venture-subsidized price war no cooperative outlasts (van Doorn). Both drains scale with the challenger's visibility, which is a design instruction. Spend symbiotic wins on lowering adoption costs, portable membership, legal forms, procurement channels, rather than on spectacle.

Capacity, the model's distinctive gate, is now measured rather than asserted. It is the coordination ability a movement earns and banks, built on resource inflows and durable organization (McCarthy and Zald; McAdam); its absence is what leaves a digitally scaled protest able to assemble but not to maneuver, the tactical freeze Tufekci named. And it pays off downstream, because the infrastructure that accumulates before the juncture is what lets the transition hold once it arrives. Longer unarmed mobilizations produce more durable democracies (Kadivar). The prefigured economy makes the gate auditable, with real numbers on both sides of the ledger. On the credit side sit three cases. Mondragon counts eighty-one cooperatives and roughly seventy thousand worker-members built over seven decades, still a fraction of one percent of Spanish employment. Preston saw locally retained procurement rise from 5 to 18.2 percent in four years, with peer-reviewed wellbeing gains against matched comparators (Rose and colleagues). And three hundred fourteen community land trusts holding permanently affordable homes. On the debit side sit two, growth bought by acquisition rather than earned by diffusion, and platform cooperatives that died against subsidy. The strategic frame these numbers support is erosion, interstitial building plus symbiotic wins, not rupture (Wright).

The engine, read in six indicators
  1. Acceleration of event frequency, not the size of the largest event.
  2. Participation retention across events, the fraction who come back.
  3. Elite defection counts, security, business, administrative, clerical: the legitimacy variable made observable.
  4. Capacity audits: procurement share retained, cooperative survival against matched firms, organic-versus-acquired growth, absorptive headroom for newcomers.
  5. Absorption events: standard dilutions, incumbent adoption of the vocabulary without the structure.
  6. Demand readings: governance participation inside adopters, split by whether an organ's protected value is visible and personal or abstract, the F–12 instrument.

The engine's missing term is demand, why an ordinary person bears clocks, ledgers, and consent rounds absent a crisis, and the record answers with a result the framework must carry whole. Costly requirements do screen free riders and predict commitment, the strictness result, but across eighty-three nineteenth-century communes the cost effect held for religious communities and bought secular ones nothing, because a cost reads as devotion only against a value the pool already holds sacred, and secular ideology could not make its demands non-negotiable (Iannaccone; Sosis and Bressler). The demand-failure quadrant is equally priced: a sixteen-million-member mutual draws three percent to its vote, gym members forgo six hundred dollars each rather than admit they will not attend (DellaVigna and Malmendier), and the collaborative project most like this framework lost its editors not to apathy but to its own compliance machinery landing on newcomers (Halfaker and colleagues). What survives between screening and burden is conditional, and the engine adopts it as three design rules. The first is visibility. Protected value must be visible and personal at the moment a cost is paid, the way homes under association rules carry a four percent premium from the same majority that dislikes the regime (Clarke and Freedman), so every organ binds its cost to a named asset of the member, the vested stake, the held home, the protected record, the codified exit, and is presented as insurance at the moment the asset is present, never as civic virtue. The second is proximity. Entry runs through ties and small asks, since cold contact converts near one in a thousand while a friend's kitchen converts near half, and a small first ask roughly quadruples consent to the large one (Freedman and Fraser), which grounds the staircase's mechanics: recruitment through existing relationships, a first rung sized to one meeting under the Lite tier, a named member owning each newcomer's first quarter. The third is load. The overhead band of EQ–02f, its compliance-load budget, binds, with any organ whose attention cost exceeds its visible protection redesigned or automated until it fits, because the record's clearest decline was a compliance budget spent at the front door.

Demand in the engine is therefore a function of three variables, visibility of the protected value, social proximity of the ask, and residual compliance load, and the strictness result governs the staircase's top rather than its base. Full membership stays costly, an explicit adult act with a vested stake, because screening belongs at the committed boundary while the burden hazard lives at the visitor's door, and the record punishes architectures that confuse the two. A crisis can spike visibility, which is why collapse recruits, but an engine that needs the crisis has already failed this section's question, and the demand term exists so adoption clears in ordinary time, carried as falsifier F–12 on the participation floor and the visibility differential at once.

THE DEMAND TERMCost screens at the boundary and burdens at the doorThe committed boundaryfull membership costly by design: the adult act,the vested stake, the screening the record supportsthe strictness result lives here, with its secular boundary namedThe visitor's doorcost here is burden: one meeting, one small ask,a named human, the Lite tier's hour a monththe compliance machinery that emptied the record's front doorsdemand rises with visible protected value, social proximity of the ask, and load inside the EQ–02f budget · F–12 tests the floor and the differentialthe framework's substitute for sanctity is falsifiability plus visible protection, and that wager is registered, not assumed
The demand term. Screening cost belongs at the committed boundary, where the strictness record supports it; the visitor's door stays cheap, social, and small, where the record shows cost only burdens. Demand rises with visible protected value, proximate asks, and a compliance load inside the published budget, and F–12 registers the wager.
Falsifier F–05

Track polities and organizations that cross the capacity gate with rising momentum against those that do not; if their adoption trajectories are statistically indistinguishable over a decade, the capacity-gated engine is wrong and OQ–01 reverts to open. A reflexivity caveat rides with it. Preference falsification hides threshold distributions until they reveal themselves (Kuran), and a well-publicized threshold may already have changed the behavior it described.

Capacity gatebuilt, audited, absorptiveCascadeadopters x momentumretention, not peaksThe windowlegitimacy drops, unschedulableIncumbent responserepression-learning, absorptiondrains scale with visibilitycapacity caps how many the cascade can receivefires only if gate and cascadeare ready when it opens
The transition engine. Prefigured capacity gates a cascade of adopters carried by momentum; the window is a legitimacy drop that cannot be scheduled; the incumbent drains the cascade in proportion to its visibility.
Open questions
  • Timing stays out of reach. The threshold distribution is unobservable in advance, and publicizing thresholds may change them.
  • The central numeric parameters are borrowed from regime-change campaigns and laboratory conventions; the NAVCO-equivalent dataset of institutional adoption now exists in embryo as the tracker (dataset v0.6, twenty cases on the readiness codebook, three of them collapsed or dissolved calibration anchors), and growing it past demonstration grade is the standing instrument task.
  • The engine has no term for external guarantors, and the strongest recent failures were backstopped from outside; that gap belongs to the perimeter (OQ–08, EQ–04c), not to this organ.
  • The sign of absorption is historically contingent. Co-optation neutralized one challenger and normalized another into a century of legal accommodation, and the engine can flag absorption events but not predict which way one cuts.
  • The demand term's deepest residual is the sanctification gap. The one variable the commune record crowns, a frame that makes demands non-negotiable, is the one variable a falsifiable framework refuses on principle, so the wager that visible protection plus falsifiable trust can substitute for sanctity has no completed secular precedent, only partial exemplars and the religious control cases, and the best-evidenced onboarding intervention on record moved retention a single point on an eight percent base, so these mechanisms bend curves rather than repeal them.
EQ–04e

The Consent Layer

Draft

No one is born a member. Consent is a veto path that never closes.

OQ–06 begins from a concession. Transitional structures act inside systems whose members never agreed to them, and legitimacy rests on exit and fairness. This section builds the mechanism the concession alone lacked, and it begins by accepting the philosophical verdict rather than arguing with it. Consent as a founding act is dead. Residence is not agreement, as the man carried aboard the vessel in his sleep does not consent to the master by declining to leap into the sea (Hume), and no existing account, consent, fair play, gratitude, or natural duty, binds a particular person to a particular polity without inventing a consent that never occurred (Simmons). What survives are grounds that do not pretend to be consent. They are fairness owed for presumptively beneficial goods (Hart; Klosko), the natural duty to support just institutions (Rawls), and the honest naming of what an institution can claim without any of it, moral justification for wielding power, which is legitimacy decoupled from the obligation of the governed (Buchanan).

The framework re-grounds consent in the grammar it already owns. A decision is non-arbitrary not because it originated in agreement but because those it touches can effectively contest it (Pettit), with discourse among all affected as the regulative ideal (Habermas) and legibility supplied by the Justice of Light. Consent in Aeviterna is therefore not a signature at the founding. It is a standing veto path. A norm binds while it survives reasoned objection from those it affects, in forums built for objection, and every norm's warrant expires into its next review. This is decaying standing applied to rules instead of people, and it answers the ship directly, since what legitimates is not a past act nobody performed but a present objection anybody can raise.

Exit remains the outer wall, and the evidence disciplines how much weight it bears. Foot-voting works only under assumptions no real polity meets (Tiebout), easy exit drains the voice that reform needs (Hirschman), exit choice correlates with stratification (Dawkins; Bayer and McMillan), migration is rarer than the theory wants (Molloy and colleagues), and formal exit rights are no substitute for internal protections for those least able to use them (Okin). The kibbutz record adds the design warning the layer adopts as law. Communal wealth that members cannot take with them converts exit rights into fiction (Abramitzky). So exit in the consent layer is codified before any dispute, with a procedure, a clock, and asset portability, the member's stake vested and traveling, on the model of the one constitutional-scale exit clause written in peacetime.

Membership itself is staged, explicit, and revocable, on the pattern the working mechanisms converge on. A refundable capital stake makes joining an adult act (Mondragon), entry is graduated through visiting and provisional periods (Twin Oaks), decisions pass by consent rounds that ask for objections rather than applause (sociocracy), and, from the least likely teacher, the Anabaptist practice of baptism at majority holds re-consent as an adult act. Authority over minors is trusteeship that decays to zero at adulthood, when the person joins by explicit act or walks with the walk pre-funded. Consent-quality safeguards ride the stage gates, because recorded consent can be theater. Most people skip the terms entirely and the rest read for seconds (Obar and Oeldorf-Hirsch; Radin), preferences adapt to deprivation (Nussbaum; Elster), and bounded-choice systems manufacture subjectively free decisions (Lalich). The gates are therefore slow, informed, and reversible through the provisional period, and belonging costs are audited the way the Justice of Light audits power.

For those who cannot consent at all, children, future people, the cognitively incapacitated, the layer builds trusteeship in the guardian grammar of EQ–08e. The mandates are plural, decaying, contestable, and per-domain, substituted judgment comes first with best interests where preferences are unknowable, the ward is heard wherever capable, the ward's open future is held as a hard limit no guardian may foreclose (Feinberg), and the ward has standing to put the institution itself in the dock, as youth plaintiffs have now done against a state and won. And for the people in the institution's shadow who never joined, the boundary rule is that whoever a decision demonstrably touches gets the objection path and the legibility, contestation without membership, a vote in nothing but a veto path in everything that reaches them. That is the answer to the demos that otherwise explodes, since enfranchising all affected interests would enfranchise nearly everyone on nearly everything (Goodin). The shadow earns contestation, not government.

Falsifier

The layer predicts that institutions running staged membership, standing objection paths, and vested exit will show measurably higher voluntary retention and lower coercion complaints than matched institutions running tacit membership, and that where the layer is present but objection rates and exits fall to zero, what is present is ritual, not consent, and the audit should say so.

Rail 1 · Standing contestationa norm binds while it survives reasoned objectionevery warrant expires into its next reviewRail 2 · Staged membershipvisit, provisional period, explicit adult act, re-consent at majorityexit codified in peacetime: procedure, clock, vested stakeRail 3 · Trusteeshipplural, decaying, contestable guardians under the open-future limitthe ward can put the institution in the dockThe shadowthose touched butnever joinedcontestation: yesgovernment: nowhat is never claimed: consent-grounded obligation from silencewhat is claimed: justified, minimized, contestable coercion, said aloud
The consent layer. Three rails carry what consent can still mean, a veto path over norms, staged and revocable membership for persons, bounded trusteeship for the unconsentable, and the boundary rule gives the institution's shadow contestation without membership.
Open questions
  • First imposition is not laundered. Founding, boundary-drawing, and deciding who counts as affected all precede any objection path, and the first cell was not consented to by its neighbors.
  • Consent quality can be raised, not certified. Adaptive preference and bounded choice deform even clean, staged, recorded consent, and the layer reduces the deformation without proving its absence.
  • Future people are represented, never consulted; reversibility is the best proxy and is itself an unconsented value choice.
  • The claim, stated without flinching. Justified, minimized, contestable coercion, never the consent-grounded obligation of the member who simply failed to object.
EQ–04f

The Economic Siege

Draft

The boundary layer holds against force, and the record says force is rarely how alternative economies die. They die of prices and flows. Capital withdraws, clearing is denied, a subsidized rival sells below cost until the members peel away, the skilled are hired out one by one, and the inputs everything runs on are foreclosed in a weekend. Five channels, none of them an army, and the perimeter either walls all five or it has walled the wrong thing. This organ builds the economic wall, and it builds almost nothing new, because the defenses are organs the framework already runs, given explicit siege duties.

The record reshapes the threat before it arms the wall. The canonical sieges were joint failures, in which internal fragility handed the besieger its opening. Chile carried a documented squeeze, bilateral aid cut twentyfold, export credits to zero, multilateral lending throttled, and still the blockade leaked, credit rescheduled and replaced, while deficits near thirty percent of GDP, money-financed, did the killing the squeeze could not (Church Committee; Sigmund; Larraín and Meller). The kibbutzim were not besieged at all; a decade of cheap unindexed credit met a stabilization shock, and a movement restructured on its creditors' terms within a generation (Abramitzky). The first course of the wall is therefore fiscal. A polity that runs deficit-financed legitimacy is siege-soft regardless of its walls, which binds the overhead band of EQ–02f and the formula discipline of EQ–02e into the perimeter itself.

The clearing lesson is the organ's sharpest, and it is a contrast. WIR, a mutual-credit network born in the 1934 credit famine, has cleared countercyclically for ninety years and moves over a billion francs among fifty thousand firms (Stodder and Lietaer). INSTEX, a clearing substitute built by ten sponsoring states for besieged trade, completed one transaction in four years and was liquidated, because the banks' fear of the hegemon's enforcement outweighed the sponsors' will. A substitute that exists as an institution awaiting its crisis fails; one that exists as living trade credit holds. The same demand-side logic answers predation, which the law will not stop: since the recoupment standard, predation plaintiffs essentially stopped winning, and the documented platform case ran two hundred million dollars of monthly losses against one rival before acquiring it (Khan; the House record). What subsidy peels is customers. It struggles to peel owners, which is why the wall against the price war is the vested stake, the visible protected value, and the institutional procurement channel, the same symbiotic wins the transition engine already teaches (EQ–04d). And the talent channel is priced rather than walled, because compression's quit gradient is measured, high-ability members leaving worker-run firms at four and a half times the rate of their peers (Burdín), so differentials are bounded and published, succession carries the memory (EQ–07b), and retention beyond the differential is named as unpurchased.

What survival looks like is on the record too. The Mondragon network refused to rescue its failing flagship and then absorbed the blow it refused to prevent, relocating hundreds of members within months while the cooperative died, graceful degradation rather than invulnerability (Whyte and Whyte). Its shadow is the organ's dignity constraint. The failure's losses fell hardest on ten thousand outside savers holding subordinated paper, and a wall that survives by exporting its losses to non-members, or by breaching the floor the way a besieged state survived its Special Period, has failed in this framework's terms even where it holds. The drill that keeps the wall honest is already state practice. Finland's supply agency and the Swiss compulsory stocks hold months of fuel, grain, and medicine through private hands and have activated them in living memory, bounded buffers that buy negotiating time without autarky. The economic black-start imports that practice to cell scale on the calendar EQ–03e already keeps, and when the blockade is private rather than sovereign, the counter-siege on record ran through a borrowed court, which is the enforcement ladder's business (EQ–06f).

The living clearing layerThe mutual-credit layer of EQ–02d runs at real volume in peacetime and federates across cells, because a clearing substitute is either alive before the siege or absent during it; the ecology's composition shifts inward as the modes harden (EQ–09).
The mode-gated reserveThe commonwealth corpus (EQ–02e) lends countercyclically only under the contested and crisis modes, gates read at the vital-signs review, so the war chest exists when the flows clamp and cannot quietly become the capture prize the regrowth tripwire watches.
The economic black-startThe drill of EQ–03e gains an economic leg: the floor functions named, the days of cover published beside the vital signs, and the fallback exercised on the same annual calendar, on the working precedent of national supply agencies and compulsory stocks.
Vested demandPredation is answered where it aims: membership stakes vested (EQ–04e), the protected value visible (EQ–04d), and institutional procurement held, because subsidy peels customers and struggles to peel owners, and the record's survivor of the price war lived on contracted demand.
The talent priceDifferentials bounded and published, succession and artifact handover per EQ–07b, and the loss above the differential named as unpurchased, so the drain is a budgeted cost rather than a surprise, and the polity never discovers its retention price mid-siege.
The polity's metabolism floor functions, EQ–02c capitalclearingpricetalentinputs mode-gated reserveliving mutual creditvested demandthe talent pricedays-of-cover drill THE FOUNDATION COURSE · FISCAL DISCIPLINE, EQ–02f BAND + EQ–02e FORMULA
The economic wall. Five clamps arrive as prices and flows (coral); each is met by a named duty on an existing organ (green), and the whole wall stands on fiscal discipline, because the record's sieges won where the interior was already failing. F–14 tests the wall in the harness and the field at once.
Open questions
  • No case shows a small federated commons surviving a determined state-scale siege intact. The walls that held did so against market sieges at cooperative scale, and the one polity that survived a state-scale cutoff did so through hunger-level rationing and a deficiency epidemic, survival with the floor cracked, which this organ counts as failure rather than vindication.
  • The quantities the falsifier needs do not exist yet, siege-episode base rates for coded adopters, the clearing volume at which mutual credit holds, and the days-of-cover thresholds per floor function; they are the tracker's and the harness's assignments, not estimates to be invented here.
  • A wall that survives by exporting losses to non-members fails in the framework's own terms, and the walled record's shadow, the outside savers wiped while members were absorbed, is carried as a design constraint on every reserve and every subordinated instrument the wall may issue.
  • Direction
    • The harness now runs a siege scenario, clamps on external capital, clearing, and demand against walled and unwalled cells; what remains is to publish days-of-cover beside the vital-signs review, record siege episodes and their outcomes in the adoption tracker, and write the reserve gates into the model charter at its next release.
Falsifier F–14

In the harness's siege scenario, cells carrying the wall must hold their floor functions under clamps on external capital, clearing, and demand that collapse matched unwalled controls, or the wall is wrong as designed. In the field, if tracker-coded adopters carrying named wall elements survive documented siege episodes at rates indistinguishable from matched unwalled peers, or survive only by breaching the dignity floor or exporting losses to non-members, the economic wall buys nothing. Ratified July 21, 2026.

EQ–05

Stewardship, Standing, and Accountability

Stable

The healthiest steward is not the person who becomes indispensable. It is the person who reduces systemic dependency over time.

Aeviterna recognizes that care work naturally generates social gravity. People who consistently maintain systems, resolve conflict, mentor others, preserve continuity, and absorb instability will naturally accumulate trust and influence. This is not inherently corruption. Accumulated trust is how distributed systems coordinate without coercion.

However, stewardship must not calcify into sovereignty. The structural conditions that produce good stewards, namely accumulated trust, domain expertise, and institutional memory, are identical to the structural conditions that produce captured hierarchies. The difference is not character. It is architecture.

In Aeviterna, standing is domain-specific, requires ongoing contribution, decays over time without renewal, increases scrutiny rather than privilege, and can be revoked through transparent accountability processes.

These properties become a mechanism once decay is made concrete. Standing is modeled as a quantity that decays exponentially: S(t) = S0 · e^(-λt), where S0 is the standing at the last renewal, λ is the decay constant set per domain by how fast that domain's knowledge turns over. A fast-moving operational domain might carry a half-life of six to twelve months, so influence earned today is worth half as much within a year unless it is renewed; a slow, independence-bearing domain runs the other way, one long mandate with the clock on the powers around it. Contribution is the only inflow, and absent renewal standing falls back toward zero on its own, which is the structural property that keeps stewardship from hardening into sovereignty. Contribution enters as its own decaying increment, so with a steady rate of service standing converges to a finite ceiling rather than growing without bound, the behavior already running in conviction-voting systems that use this exact decay with an inflow.

These ranges are reasoned from how fast a field's knowledge turns over rather than chosen at random. The half-life of knowledge is an old idea (Machlup), and the turnover it names has been measured. Medical knowledge that took about fifty years to double in 1950 took roughly three and a half years to double by 2010 (Densen), while constitutional norms move over decades. What carries weight is the ordering, fast domains forgetting within months and slow ones holding for years, and not any single constant, which has to be fit to each domain and revised as the domain changes.

The tuning is a rule, not a guess. Each domain's half-life is anchored to a measurable proxy for how fast its knowledge turns over, the field's cited half-life or the inverse of its skill-depreciation rate, then clamped into the calibrated bands of EQ–05d so a mis-estimate cannot produce an absurd value: 6 to 24 months where operational skill is perishable, 2 to 5 years for professional and technical knowledge, 6 to 18 months in staggered cohorts for representative standing, and, for independence-bearing judgment, the inversion, one long non-renewable mandate with the decay applied to the review and appointment powers around the office, with the locked meta-core exempt from decay entirely. The tiers are republished and refit as a domain's measured turnover changes, and the estimator is open and forkable so the choice is contested rather than imposed.

100%50%0%fastmidslowdecay, per-domain half-lifewith contributiontime; each curve halves at its own half-life, contributions marked
Standing decays unless renewed, and the half-life is set per domain. Fast domains (steep) forget within months, slow constitutional ones (shallow) hold for years, and each gold curve crosses the halfway line at its own half-life. Ongoing contribution (green) tops standing up, so sustained influence requires continued service.
Domain-tuned half-lifeThe decay constant is set by how fast a domain's knowledge turns over, so fast domains forget quickly and slow domains hold longer. A half-life rather than a flat rate is what makes the decay non-arbitrary.
Contribution as the only inflowStanding rises only through contribution: maintenance, mentoring, conflict resolution, and the absorption of instability, attested by domain peers and logged in distributed civic memory.
Two kinds of lossPassive decay handles ordinary non-contribution; active revocation handles bad faith through graduated sanctions, from warning and remediation to a partial slash to full revocation and referral to the Justice of Light (EQ–06b), raised by domain peers, adjudicated by a panel drawn by lot, and open to appeal. Coordinated gaming caught in the same window is penalized super-linearly, so isolated error stays cheap while collusion rings pay in proportion to their size.
A standing vector, never a scalarThere is no total, no leaderboard, no convertible currency of influence; each person holds a separate decaying balance per domain. Cross-domain transfer is allowed only by peer attestation, at a steep discount, and capped, so authority earned in one place cannot be cashed in to dominate another. A concentration tripwire watches the overlap of high-standing holders across domains and forces disclosure and faster decay where the same cluster recurs. This is the anti-capture core shared with EQ–06.
Scrutiny scales with standingThe more standing a steward holds, the more transparency they owe. Influence buys obligation, not exemption.
Emergency standing decays fastestAuthority granted during a crisis carries the shortest half-life of all, so it unwinds on its own as the emergency passes. This is the hinge between standing and the governance modes of EQ–09.
Standing as compostA fixed share of decayed standing is redistributed as support for new entrants, so the dimming of established influence directly seeds the next contributors.
Gaming-resistant contributionAny single contribution metric becomes a target and rots into a bad measure (Goodhart; Campbell; the surrogation experiments of Choi and colleagues). The answer is triangulation, several weak signals instead of one strong one, peer attestation, evaluators drawn by lot, and audit after the fact, so no single number is worth capturing.
Identity without a central authorityNon-fungible standing needs each person counted once, met through portable standards-based credentials (the W3C verifiable-credentials model) and multi-signal proof of personhood rather than one registry or one biometric. Every method trades Sybil-resistance against privacy, so the design leans on several modest signals over a single gatekeeper.
Earned standing runs a domain; lot checks itDecaying standing governs who contributes and whose contribution carries weight inside a domain; selection by lot governs the checking of it from outside, the panels that adjudicate sanctions and the seat that defines the metrics and the half-lives (Bagg). Aiming that framing seat by lot, with competing methodologies and a right to force a contest, is the same anti-capture move the guardian of the core makes in FC–04.
Open questions
  • The half-lives are reasoned from how fast each domain's knowledge turns over, but they still have to be calibrated and refit in practice; the ordering is firmer than any single constant.
  • Contribution cannot be measured in a fully game-proof way, since any measure becomes a target (Goodhart; Campbell). Triangulation, peer attestation, sortition of evaluators, and after-the-fact audit lower the pressure rather than removing it.
  • Counting each person once without a central identity authority stays a live tradeoff between Sybil-resistance and privacy, now met with standards-based credentials and multi-signal personhood rather than a single registry (W3C verifiable credentials; the proof-of-personhood debates). See OQ–02.
  • Direction
    • Now specified above as a tuning rule (per-domain half-life tiers anchored to knowledge turnover, refit on a schedule), a bounded accumulation ceiling, a non-fungible standing vector with a cross-domain concentration tripwire, graduated-sanction revocation with a collusion penalty, and the framing seat filled by lot (Ostrom; Bagg; Posner and Weyl; W3C). Falsifiers F–03 and F–04. See OQ–02.
  • Implementation sketch · exploratory
    • Half-life offices whose standing visibly decays on a public dashboard with scheduled rotation (the Ostbelgien citizens' council; term limits), and standing as compost, where decayed influence funds new entrants (quadratic funding).

Decaying standing is written into adoptable clauses, with terms, sunset dates, and staggered rotation, in the Model Charter, and the clocks themselves run in the Sunset Calendar, a self-contained tool that computes each role's review date and lays the rotation across the months. Its enforcement rides the ladder's inner rungs (EQ–06f), monitoring, the vested stake, and federated recognition, sized to survive the review the borrowed layer applies.

EQ–05b

Institutional Mortality

Draft

The forest does not keep every tree standing. That is why it is still a forest.

Almost everything in the framework is built for continuity, for resilience and the preservation of memory and stewardship across time. But a civilization that successfully preserves all of its institutions accumulates dead weight, and the accumulation is itself a failure mode. Old institutions capture resources and ossify around their original purpose, crowding out the new ones that changing conditions require, until the system loses the capacity to oscillate that EQ–02 defines as health. EQ–05 lets the standing of persons decay; this section lets institutions themselves end, and end well.

The model is ecological. A forest stays alive because death and decomposition return nutrients to the soil where new growth draws on them, and a forest that somehow kept every tree forever would starve. Human systems already have a crude version. Sunset provisions write an expiration date into law, so a body is abolished on a fixed date unless it is actively renewed, inverting the default from persists-until-killed to ends-unless-justified. Texas has run a Sunset Advisory Commission on this basis since 1977, reviewing roughly a hundred and thirty agencies on a rolling cycle and abolishing the ones that can no longer justify themselves. EQ–06 and EQ–09 already use this logic narrowly, in sunset clauses on emergency authority. Institutional mortality generalizes it into a full lifecycle.

The generalization is not that things should end, which is easy, but that they should end well, which is the neglected part. When an institution dies it holds resources, accumulated knowledge, relationships, and trained people, and the default failure is that these scatter or are captured in the collapse. A composting institution is designed from birth to hand them back. Its assets return to a defined commons, its memory is deposited in the distributed civic archive of EQ–06 rather than lost, and its people are transitioned rather than stranded. Death becomes a transfer rather than a loss, and the freed resources become the soil the next institution grows in. EQ–05 already treats decayed standing as compost for new entrants; this is the same move at the scale of whole institutions.

BirthMaturityrenew only on reviewSunsetends unless justifiedreturnsThe commons, soil for the next institutionassets returned, memory archived, people transitioneddeath feeds the next beginning
Death as transfer, not loss. An institution born with an expiration runs its arc, and on sunset its assets, memory, and people return to the commons rather than scattering in collapse. That returned material is the soil the next institution grows from.

There is a deeper reason to build this in. Hannan and Freeman found that organizations rarely transform fundamentally, because they are too structurally inert, so populations of organizations evolve mostly through selection, through new ones founded and old ones dying, rather than through existing ones adapting. If that holds, a civilization's adaptive capacity depends heavily on its founding-and-dying rate, and one that suppresses institutional death has quietly suppressed its main channel of change. The honest danger is plain, since a mortality mechanism is a weapon and "this institution has outlived its purpose" is exactly what a motivated actor says to kill an institution that constrains them. So institutional death runs under the same anti-capture stack as everything else, with the decision transparent and contestable and never in the sole hands of those an institution exists to check, which is why the agonistic layer (EQ–06d) and the wider deliberation, not an executive, hold the question of what gets to end.

Born with an expirationNew institutions are chartered with a default sunset date and a named successor-commons for their assets, so continuation is a choice that must be re-made rather than a default that must be fought (the Texas Sunset model).
A composting planAssets return to a defined commons, memory is deposited in the civic archive, and people are transitioned, so the value an institution holds outlives the institution (EQ–06; EQ–05 on standing as compost).
Selection, not only repairBecause institutions rarely transform from within, renewal is expected to come substantially through founding and dying rather than through reforming incumbents (Hannan and Freeman; Schumpeter on creative destruction).
The decision held in the openA mortality review is transparent and contestable, with the burden on the institution to justify continuation, and the question is held by deliberation and the agonistic layer rather than by any executive that benefits from the outcome.
Open questions
  • What triggers a mortality review, and how the trigger is kept from being aimed at inconvenient institutions by those they constrain, is unresolved.
  • Transferring the assets, memory, and people of a dying institution without the transfer itself becoming a site of capture is open.
  • The founding-and-dying rate has to be tuned, since too slow accumulates dead weight and too fast destroys institutional memory and trust, and the locked meta-core of EQ–10 must be exempt from sunset entirely.
  • Direction
    • Generalize the sunset clauses of EQ–06 and EQ–09 into a full institutional lifecycle in which institutions are born with an expiration and a composting plan, returning assets, memory, and people to the commons on death, with the decision held transparently rather than by any executive (Texas Sunset; Hannan and Freeman).
  • Implementation sketch · exploratory
    • Charter new institutions with a default sunset date, a named successor-commons, and a deposit obligation to the civic archive, then run a transparent, contestable renewal review on the Texas model with the burden of proof resting on continuation.
EQ–05c

Reversibility and Repair

Stable

Corrigibility is only as real as the commitments you can undo.

The framework prizes corrigibility, the capacity to detect error and change course, but corrigibility on paper is hollow if the world it builds cannot be changed back. Reversibility and repair are corrigibility made material, extended into infrastructure, technology, and law, and they complete the institutional mortality of EQ–05b. A body that can be wound down well is one whose commitments were reversible to begin with.

Irreversibility already works as a decision criterion, since the Rio Declaration holds that the threat of irreversible damage is no place for delay, and economics gives it a price through the quasi-option value of Arrow and Fisher, the worth of the information forgone by closing an option. The design rule is the Collingridge dilemma. Act while a technology is still cheap to control, before it entrenches and control becomes expensive.

Reversibility is already standard where the law requires it. Decommissioning funds for nuclear plants are held externally and paid in before operation, extended producer responsibility puts end-of-life back on the maker, and the right to repair has become hard law, with the European Union's 2024 directive obliging repair and spare parts and forbidding techniques that obstruct them. The honest counter-case is that some harms, extinction and crossed climate thresholds, are beyond any undo, so reversibility is a discipline for the recoverable paired with hard precaution for the rest.

ease of controlknowledge of harmkeep options openlocked intime, and the cost of reversing a commitment
The Collingridge window. Early on a technology is easy to control but its harms are unknown; later the harms are clear but it has entrenched. Reversibility-by-design keeps options open while control is still cheap.
Price irreversibilityAn irreversible commitment carries an explicit option-value accounting and a higher decision threshold (Arrow and Fisher; Sunstein).
Act in the Collingridge windowKeep options open while control is still cheap, before lock-in hardens the choice (Collingridge).
Fund the wind-down firstExternal, pre-funded decommissioning and producer responsibility, so the cost of undoing is borne by the beneficiary, not the public (nuclear decommissioning funds; extended producer responsibility).
Bankruptcy-remote the corpusA wind-down fund survives the beneficiary only if it sits in a segregated external trust legally severed from the operator's estate, entrenched like the commonwealth trust's corpus (EQ–02e). The failure names the fix. When Peabody, Arch, and Alpha entered bankruptcy in 2015 and 2016, self-bonded cleanup promises backed only by the parent's good name collapsed from 3.86 billion dollars to 214 million. Nuclear decommissioning trusts held outside the licensee's estate, by contrast, have been kept whole by the courts (Lordan-Perret and colleagues).
A right to repairParts, access, and a ban on repair-obstructing design, extended from products toward institutions (the EU Right to Repair Directive, 2024).
Open questions
  • Some harms are irreversible, so for them only prevention counts and reversibility is beside the point (Lenton; Armstrong McKay).
  • A reversibility norm can curdle into permanent indecision, since keeping options open consumes capital and every option carries risk (Sunstein).
  • Shedding funds through restructuring is now a named legal test rather than a warning. The corpus counts only if it is bankruptcy-remote, held in a trust outside the operator's estate, since the self-bonding collapse and the surviving nuclear trust are the same promise with opposite survivability. What such a trust guarantees is custody, not adequacy, and limited-liability ring-fencing can still strand a shortfall, so external holding is necessary and not sufficient.
  • Direction
    • Make reversibility a first-class design property, a decommissionability clause requiring a funded, externally held, audited plan to repair, retrofit, and wind down any civilizational-scale system, with irreversible commitments held to a higher bar (Collingridge; Arrow and Fisher; EU Right to Repair).
EQ–05d

Calibrating Decay

Draft

Decay the chair and exempt the corridor, and the corridor governs.

The standing engine (EQ–05) gave decaying authority its mechanism and left its half-lives uncalibrated, which the framework has owned as OQ–02's residual and the amateur-government objection has exploited. This section calibrates, from the literatures that actually bear on the question, and reports what they support. The result is bands per domain class, each labeled by the strength of its anchor, one structural inversion, and two perimeter rules that bind harder than any number.

On competence, the evidence is better than the objection assumes. Across roughly three hundred documented deliberative processes, randomly selected citizens produce actionable recommendations on complex policy, with about four days of structured deliberation as the working minimum (OECD); deliberative polling shows replicated knowledge gains across a hundred polls in thirty countries (Fishkin), controlled experiments find real learning without polarization (Setälä and colleagues), and the Irish assembly's position carried a national referendum at two-thirds. The documented failure sits downstream, in uptake. Of the French climate convention's 149 proposals, roughly one in ten survived into policy unmodified, against a presidential promise of no filter. And the working precedent for decaying representative standing already runs, a permanent citizens' council seating members for eighteen-month terms with one third rotating every six months, so the body never turns over at once (Ostbelgien). Rotation, staggered, is not amateurism; it is how the amateurs stay competent.

The strongest counter-evidence is the term-limits record, and the framework adopts its lesson rather than disputing it. Across fifty states, term limits stripped legislative expertise and shifted power to governors, permanent staff, and lobbyists, the actors with no clock (Kousser; Carey and colleagues), the lobbyists themselves reporting the influence gain (Moncrief and Thompson), while removing careers altogether distorted effort the other way (Motolinía). Decayed authority does not evaporate; it migrates to whoever is exempt. Skill itself decays on a measured curve, near zero loss immediately after training to a large deficit past a year of nonuse, with task type moving the rate by an order of magnitude (Arthur and colleagues), and unrefreshed seniority inverts the experience curve in practice (Choudhry and colleagues). Institutions forget through rotation, restructuring, and outsourcing (Pollitt), which is why the calibration cannot stand without the memory organ. And where independence is the product, the evidence runs the other way entirely. Long-horizon central banks deliver lower inflation at no output cost, high governor turnover predicts inflation, and de facto judicial independence with non-renewable terms correlates with growth where renewable terms create please-the-reappointer incentives (Alesina and Summers; Cukierman and colleagues; Voigt and colleagues).

TierDomain classHalf-life bandAnchorAnchor strength
1Perishable operational skill: incident response, clinical procedure, flight-critical operation6 to 24 monthsMeasured decay within one unpracticed year; aviation's 6, 12, and 24-month checks; medicine's move to continuous assessmentStrong and direct
2Professional and technical knowledge: medicine, engineering, policy analysis2 to 5 yearsContinuing-education cycles of 1 to 3 years; the seniority-quality inversion; the legacy 5-year reinvestigationModerate; direction well-evidenced, intervals conventional
3Representative standing: agenda-setting, deliberative mandate6 to 18 months, staggered cohortsThe permanent citizens' council's 18-month staggered terms; assembly norms near a one-year arcAnalogical; working practice, not causal calibration
4Independence-bearing judgment: adjudication, monetary stewardship, safety regulationInverted: long single non-renewable mandates, 5 to 15 years, with decay applied to the surrounding review powersCentral-bank independence and turnover results; de facto judicial independenceStrong directionally; band edges arbitrary

Tier 4 is the inversion the evidence forces. Re-earning standing is structurally identical to reappointment, the exact incentive that corrupts judges and central bankers, so for independence domains the engine flips. The officeholder gets one long, non-renewable mandate and pleads to no one, and the clock runs instead on the review and appointment powers around the office. The inversion protects the officeholder's independence, not the office's immortality. The institution around a Tier 4 seat still carries its EQ–05b expiration, and no mandate outlives its body. Two perimeter rules bind harder than the bands. No exempt adjacents. Every role touching a clocked domain, staff, counsel, consultant, auditor, lobbyist-equivalent, carries a clock in the same tier or the next one up, because the term-limits record shows what the alternative empowers. And decay pairs with memory. Every clocked role carries succession overlap, artifact handover, and a memory-organ deposit (EQ–07b) as a condition of exit, because rotation without externalized memory is engineered amnesia.

Falsifier

If organizations running these bands with the perimeter rules show no measurable capture-resistance or competence advantage over matched organizations running fixed tenure within a decade, the calibration is decoration and OQ–02 stands unresolved.

standingtimeTier 1 · 6-24 moTier 2 · 2-5 yrTier 3 · staggered cohortsTier 4 · inverted: one non-renewable mandate, then zerothe clock runs on the review powers around the officeinflow from attested contribution re-charges any curve; none may exceed the ceiling
Calibrated decay. Three tiers decay on banded half-lives anchored in measured skill loss, renewal practice, and working sortition; the fourth tier inverts, one long non-renewable mandate with the decay applied to the powers around the office.
Open questions
  • Interval optimality is an evidence vacuum. The one strong interval study found renewal frequency itself inactive, and three mature regimes abandoned fixed intervals for continuous assessment, so whether decay should be clock-driven or event-triggered is a design choice the evidence cannot yet arbitrate.
  • Goodhart pressure is constitutive. Once decay is codified, re-earning becomes the target (Strathern), and the engine's qualitative panel review is load-bearing, not decorative.
  • Tier 4's inversion is a value choice per domain, made explicit so it can be contested; no half-life reconciles must-re-earn with must-not-please.
  • Personnel half-lives cannot control institutional forgetting on their own; this organ leans on EQ–07b and cannot stand without it. The bands are defensible; the band edges are not derivations; both facts are load-bearing.
EQ–06

Anti-Capture Architecture

Stable

Aeviterna assumes that all systems drift toward concentration without corrective feedback. This is not a claim about human nature. It is a claim about incentive structures. Any system in which concentrated control produces benefits will attract concentration. The question is not whether this happens. It is how fast, and whether the system detects and corrects it before the correction becomes impossible.

The primary adversarial actors are not external enemies but internal beneficiaries, those who gain from suppressing the feedback mechanisms that would otherwise limit their accumulation. Anti-capture architecture must therefore be designed to function against motivated insiders with institutional access, not merely against naive external threats.

Rotational OversightOversight roles rotate on defined schedules, preventing the accumulation of institutional knowledge as a private resource.
Distributed Civic MemoryInstitutional history is preserved in formats accessible to multiple independent parties, preventing selective amnesia.
Open-Source GovernanceDecision protocols are public, forkable, and subject to external audit.
Institutionalized DissentFormal roles exist for contestation of dominant positions, with protected standing for dissenters.
Public Red-Team ReviewMajor decisions are subject to structured adversarial review before implementation.
Multi-Model VerificationAI systems used in governance are checked against independent models to reduce single-point epistemic failure.
Sunset ClausesConcentrated emergency authority carries automatic expiration, requiring active renewal rather than passive continuation.
Cross-Domain AuditingAuditors operate outside the domain they are auditing, reducing regulatory capture.
Capture BountiesA permanent, rotating, well-funded internal red team is paid to demonstrate capture before an outsider exploits it, on the model of security bug bounties.
Network-Concentration TripwiresGraph metrics watch the overlap of ties among successive office-holders; when concentration crosses a threshold, disclosure and an extra rotation are forced.

Anti-capture mechanisms are only as good as the adversaries they are tested against, so each one here is paired with a named threat and, in the spirit of security threat modeling, with the way it can itself be turned. The pairing exposes the deep residual. Rotation and decay defeat the entrenchment of any individual, yet the same credentialed network can recirculate through rotating seats and keep its hold. The signal to watch is therefore not individual turnover but network concentration, the overlap of ties among successive office-holders, which is what the tripwires are built to catch.

MechanismAdversary it defeatsIf the mechanism itself fails
Rotational oversightRegulatory capture (Stigler): the durable overseer-and-overseen bond that capture relies onToo-fast rotation erodes expertise and the coherence floor
Distributed civic memoryRecord-tampering and repudiation: rewriting which commitments were madeGaps leave blind spots; total recall erases protected ambiguity
Open-source governanceHidden rule-changes and unauditable decisionsForkability without real participation becomes exit theater
Institutionalized dissentManufactured consensus and the silencing of minoritiesProtected standing can be abused for bad-faith obstruction
Public red-team reviewGroupthink and unexamined decisionsRed teams become theater if their findings carry no weight
Multi-model verificationSingle-point epistemic failure and model monocultureCorrelated models fail together, the model-collapse risk
Sunset clausesThe ratchet (Higgs): emergency power that never expiresSunsets renew by inertia without escrow, the PATRIOT case
Cross-domain auditingDomain self-dealing and captured auditorsThe auditor layer becomes the new capture point
Capture bountiesLatent, undetected capture and the insider advantageBounty gaming, and chilling effects if it turns punitive
Network tripwiresElite recirculation (Pareto): one network through rotating seatsThreshold-setting is itself a capturable choice

The table above is static, and the record is not. Security engineering made this upgrade fifteen years ago, from lists of threats to models of campaigns. The kill chain reframed intrusion as a phased progression in which the adversary must succeed at every stage while the defender needs to break only one link (Hutchins, Cloppert, and Amin), and the ATT&CK discipline replaced speculative threat lists with a catalogue of tactics compiled from observed incidents and revised as new behavior appears (MITRE). The framework imports both moves as the capture cascade, distilled from the best-documented campaigns on record: the South African state-capture commissions, the Hungarian and Polish constitutional captures and Poland's stalled restoration, the Boeing delegation case, and the flash-loan governance takeovers (the Zondo Commission; Scheppele; the House 737 MAX report). Capture arrives in six recurring stages: entry by mandate or delegation; the narrative environment; personnel before policy; the referee; the prize and the watchdogs; and lock-in. Each stage has a watcher already in this chapter. The consent layer's records watch entry, the cognitive immune system and institutionalized dissent watch the narrative stage, rotation, lot, and the network tripwires watch appointments, cross-domain audit and distributed memory watch the referee, capture bounties watch the prize, and the sunset discipline with EQ–09c's transcription ban watches lock-in.

Two structural rules join the stages, and both come from the record rather than from theory. First, friction is a defense in itself. The machine-speed cases show that where acquiring decision power and executing sit in the same window there is no link left to break, an attacker flash-borrowed a voting supermajority and drained a treasury with the borrowed votes repaid in the same block, and the one defense that survived contact was the mandatory delay between proposal and execution. Outside the crisis mode's bounded window, every consequential governance action therefore carries an irreducible, published delay scaled to its weight, and even crisis ignition is instant only for the enumerated, reviewable powers EQ–09c declares in advance. Second, reversal is asymmetric. Capture is fast to build and slow to undo lawfully, Poland's restoration stalled against its own captured referee, the hollowed revenue authority took years to rebuild while prosecutions still lag, and the delegation case ran eleven years to a resolution with no conviction. The cascade therefore weights its tripwires toward the early, cheap stages, expression, civic space, and appointments, which the population-scale data show moving first and fastest, while erosion halts, when it halts, through multiple accountability actors acting together (V-Dem; Lührmann and Lindberg; Bermeo; Laebens and Lührmann).

THE CAPTURE CASCADEEntrymandate, delegationNarrativemedia, civic spacePersonnelboards, benchesRefereecourts, constitutionPrizetreasury, watchdogsLock-inrules, ratchetone broken link stalls the campaignconsent-layer recordscognitive immune system,protected dissentrotation, lot,network tripwirescross-domain audit,distributed memorycapture bounties,red-team reviewsunset clauses,transcription banthe watchers, one per stagemachine speedthe whole cascade in one window:borrowed votes, passed proposal, drained treasurydelayfriction re-inserts a link: an irreducible, publisheddelay between proposal and executionstages are a repertoire with a soft ordering, not a law: one campaign led with the referee, one entered at the delegation layer
The capture cascade. Six stages recur across the documented record, each watched by a mechanism from this chapter; breaking any one link stalls the campaign. At machine speed the whole chain compresses into one window, which is why friction, a published delay between proposal and execution, is a defense in itself.
The friction ruleOutside the crisis mode's bounded window, every consequential governance action carries an irreducible, published delay between proposal and execution, so a link always exists to break. Where decision and execution share a window, the record shows no defense survives (the flash-loan takeovers).
Early-stage weightingThe cascade's tripwires fire on the fast-moving indicators, expression, civic space, and appointments, because capture is fast to build and slow to reverse lawfully, so detection is priced toward the stages where intervention is still cheap (V-Dem; Laebens and Lührmann).
Falsifier F–06If a polity or institution running this layer traverses three consecutive cascade stages without any watching mechanism firing, the cascade's detection claim fails as specified and the layer is decoration. If documented cases show no ordering tendency at all, the cascade collapses back into the static table it replaced.

Anti-capture has a second front, the slow capture of the infrastructure itself by vendor lock-in or by a single cryptographic assumption. The defense is to treat every technical choice as replaceable. The framework specifies what a mechanism must guarantee and leaves how it is built to open standards that can be audited and swapped.

Open standards by defaultMechanisms are built on published, auditable standards from bodies such as the IETF, the W3C, and NIST, on the old principle that a system should stay secure even when everything but the key is public (Kerckhoffs). Security through obscurity is treated as a weakness, not a defense.
Crypto-agilityCryptographic primitives are kept swappable, so a broken or deprecated algorithm can be replaced without rebuilding the system (IETF RFC 7696; NIST). The current baseline is the post-quantum set finalized in 2024 (ML-KEM, ML-DSA, SLH-DSA), carried in a hybrid with proven classical schemes through the transition.
Held open by designWhere the right primitive does not yet exist, the framework names the requirement and waits rather than freezing an immature choice. Verifiable-computation tools such as zero-knowledge proofs and secure multiparty computation are adopted as they mature and standardize. This is capacity meant for nations and for time, so it is built to outlive any single technology.
Open questions
  • Rotation and decay defeat individual entrenchment but not necessarily elite recirculation; network concentration is the metric to watch (Pareto).
  • Every added defense is another layer that can be captured; oversight-of-oversight has no proven terminating point.
  • Setting the tripwire thresholds is itself a capturable choice.
  • Direction
    • Pair each mechanism with a named adversary (rotation versus regulatory capture, memory versus record-tampering, decay versus oligarchy, separation versus power fusion), and model the campaign as well as the adversary. The capture cascade above is the worked answer to OQ–03, validated against the documented record and carrying falsifier F–06.
  • Implementation sketch · exploratory
    • Capture bounties run by a permanent, rotating red-team office paid to demonstrate capture (bug bounties; artificial immune systems), and network-concentration tripwires that watch overlap among successive office-holders rather than individuals (structured transparency).
EQ–06b

The Justice of Light

Stable

A civilization that cannot handle harm transparently will handle it secretly. Secrecy in justice is not discretion. It is the first condition of abuse.

Anti-capture architecture addresses systemic drift toward concentration. But systems are also harmed by individuals: those who act in bad faith, those who cause damage through willful disregard, and those who reject the shared conditions that make civilizational cooperation possible. Aeviterna requires a principled response to all three. That response is the Justice of Light.

Justice in Aeviterna is not downstream of punishment theory. It derives from the same commitments that structure the rest of the framework: cognitive sovereignty requires that people understand what is being done in their name; stewardship demands that authority over others be accountable and corrigible; legitimacy collapses when harm is hidden rather than addressed. A justice system that operates in shadow is not compatible with a civilization organized around legible power.

The framework distinguishes between three categories of harm and responds to each proportionally. The goal across all three is the same: protect the commons, preserve dignity, maintain transparency, and keep every decision open to review. Aeviterna does not seek revenge. It seeks corrigible containment, the minimum intervention necessary to protect the community, held in the open, with standing to challenge built into the structure itself.

This applies equally to individuals and to institutions. A corporation that systematically suppresses feedback mechanisms, a platform that shapes cognition at scale for extractive ends, a state that governs without consent. These are actors causing harm within the framework's definition, and they are subject to the same principles of transparent accountability. The three-tier structure scales.

Tier 1: ContainmentProportional, humane, and subject to public structural oversight. Rehabilitation and access to knowledge continue. The person is separated from the harm they cause, not from their humanity.
Tier 2: Moral FirebreaksExile from networks of trust when reintegration poses ongoing risk. A firewall, not a sentence. Transparent in reasoning, open to review, and never permanent by default.
Tier 3: Non-ReintegrationReserved for those who reject the shared conditions of cooperative life entirely. Isolated without cruelty. Every decision logged, reviewable, and revisable. No one disappears.
Institutional AccountabilityThe same framework applies to organizations, platforms, and states. Scale does not grant immunity. Concentrated harm requires proportionally transparent response.
least intervention firstescalation only as neededTier 1 · Containmentfrom the harm, not the personTier 2 · Moral firebreaksexiled from trust, reversibleTier 3 · Non-reintegrationisolated without crueltyHeld in the lightevery decision logged, reviewable, and revisable. No one disappears.always reopenable, and the graver the tier the lower the bar to reopen it
The Justice of Light. The least intervention that protects the commons comes first, escalating through the tiers only as needed. Every tier rests on the same floor of transparency, where each decision is logged, reviewable, and revisable and no one disappears, and the graver the tier the lower the bar to reopen it.

Four questions define the machinery: who adjudicates, who can challenge, how institutions are tiered, and how justice behaves in a crisis. They are answered as follows.

AdjudicationHarm is found by a lay panel chosen by lot from the affected domain. Where possible the matter goes first to a restorative circle, and only if restoration fails or is unsafe does a separate rotating panel assign a tier, on a standard of evidence that rises with severity (Tyler on procedural legitimacy; Braithwaite on restorative justice).
Standing to challengeThe responsible party can always appeal, through automatic periodic review. The harmed party holds standing that decays slowly and renews on any fresh harm. Narrow, fast-decaying standing lets the wider community contest on procedural or precedent grounds. The graver the tier, the lower the bar to reopen it.
Institutional tiersAn institution's tier is gated on corrigibility, not size. Reversible harm under genuine oversight stays Tier 1. Tier 2, severance from networks of trust, follows recidivism or the suppression of feedback. Tier 3, an orderly wind-down that still preserves workers and knowledge, is reserved for a business model that is itself the harm and refuses oversight (Ayres and Braithwaite, responsive regulation).
Justice across governance modesCrisis may expedite Tier 1 containment but cannot originate a firebreak or non-reintegration, and a non-derogable floor holds in every mode. Every decision stays logged and revisable, and no one disappears. Recovery then audits every expedited action (the non-derogable-rights principle of ICCPR Article 4; Ackerman on the emergency constitution).
Restorative-first, with its eligibility namedThe commitment is strengthened where the old evidence said and bounded where it was silent. A randomized trial of felony-level youth diverted to restorative conferencing cut re-arrest 44 percent, durable at four years (Shem-Tov, Raphael, and Skog), while the rigorous end of the literature finds nulls for blanket low-level diversion and the coercive-control domain stays contraindicated, so the rule is restorative-first where there is a serious offense, an identifiable victim, and consent on both sides, and not a universal default.
Containment absorbs focused deterrenceThe offer and the consequence are delivered together, respectful direct communication, real services, and certain but parsimonious enforcement, by people with standing; the architecture carries a moderate effect across twenty-four quasi-experiments, and names its own ceiling in the same breath: zero randomized trials, and effects that shrink as designs strengthen (Braga, Weisburd, and Turchan).
Reach through credible messengersThe tier's contact is carried by people with relational standing in the networks it serves, the one design feature that survived multiple-testing adjustment in the flagship community-violence trial, whose outreach-referred subgroup showed the durable effect the pre-specified index did not (Bhatt, Heller, Kapustin, Bertrand, and Blattman).
Open questions
  • Adjudication, standing to challenge, institutional tier criteria, and behavior across governance modes are specified in the mechanisms above, and each relocates capture risk rather than removing it, toward expert-led juries, custodial standing for diffuse harms, performed institutional compliance, and self-declared crises. But the section's primary residual is now the most replicated finding in the whole literature. These programs rarely die of falsified theory. They die of fidelity collapse and the funding cliff, and 2025 supplied the largest instance on record when roughly half the national community-violence grant portfolio was canceled in a season. That is a claim about the polity's constancy, not the mechanism's validity, which is why a tier funded by annual grant weather is already scheduled to fail, and why the floor's finance (EQ–02e) is where the constancy has to come from.
  • Implementation sketch · exploratory
    • A harm ledger, a public and correctable record of harms and repairs using asymmetric memory, where decision records persist while personal data stays erasable (Estonia X-Road; right to erasure).
Further reading The Justice of Light →
EQ–06c

Dignity and the Rights Substrate

Stable

A civilization needs a foundation that fixes who and what counts, and one value that is never on the table. Aeviterna takes that value to be human dignity. The model is the German Basic Law, whose first article opens that human dignity shall be inviolable and that to respect and protect it is the duty of all state authority, and which places that clause beyond the reach of amendment. Dignity is the apex norm from which the rest of the rights substrate is derived, and in the language of this framework it is the one quantity that does not oscillate. The system is built on bounded movement between extremes, but dignity is a hard bound, not a dial.

The content of dignity has a usable core. In the Kantian reading carried into German constitutional practice through the object formula, a person must never be treated merely as an object or as a means to someone else's ends. The Federal Constitutional Court has treated dignity as non-balanceable, refusing even to weigh it against other goods, most starkly when it struck down a law that would have permitted shooting down a hijacked airliner, on the ground that doing so would treat the passengers aboard as mere objects. Aeviterna adopts the same posture. Dignity sets inviolable constraints that no optimization is allowed to trade away.

The comparative record, read in both directions, forces a sharper statement, and the framework adopts it. Full-blown non-balanceable dignity decides cases in one jurisdiction, and even there the absolutism is a rule about instrumentalizing the innocent rather than a general trump. The same aviation judgment that forbade sacrificing passengers held that downing a plane carrying only its hijackers offends nothing. Where dignity works across jurisdictions, it works as an enumerated list or a social minimum. The European court holds its torture article absolute even against saving a child's life (Gäfgen), the ICCPR's non-derogable list is an existing plural consensus, Colombia's constitutional court enforces a vital minimum daily through the tutela and holds the essential core of displaced persons' rights beyond any budget, and Kenya's supreme court made its dignity article decisive against violent eviction. Where dignity is a master concept instead, it dilutes into proportionality (South Africa; India; Israel), is formally abandoned, Canada's supreme court retreated from dignity as a legal test after nine years because it proved abstract, subjective, and a burden on the claimants it was meant to serve (Kapp), or is turned against the person it protects, as when a ban grounded in dignity overrode a man's plea that his dignity consisted in having a job (Wackenheim). The scholarly verdict matches, finding a thin common core, the ontological, relational, and limited-state claims, and beyond it no universal principled basis (McCrudden; Macklin).

The floor is therefore rebuilt as the plural floor of two tiers and three rules. Tier one, the absolute list. Absoluteness attaches to enumerated prohibited treatments, never to the word dignity: torture and cruel or degrading treatment and threats of them; the deliberate sacrifice of non-consenting innocents as a means, the object formula made a rule; slavery and servitude; denial of legal personality; punishment without law. Each anchor sits where plural consensus already exists. Tier two, the social minimum. Essential levels of subsistence, shelter, health, and education, lexically prior to other spending but resource-sensitive, enforced by the burden-shift of the minimum-core doctrine, an institution pleading scarcity proves exhaustion of alternatives, in the open (Bilchitz; Young). Three rules bind both tiers: they state agent-facing duties about treatment, never values to weigh, because the Canadian retreat shows what dignity-as-test does to claimants; neither tier may be invoked against the right-holder's own informed choice absent concrete harm to another, which closes the paternalism door the dwarf-tossing and peep-show cases opened; and norm-absoluteness never entails remedial rigidity, so structural, supervised, staged remedies on the Colombian model discharge hard-case pressure at the remedy level instead of eroding the norm. OQ–00's guardian gains precision for free. It interprets a list plus a burden-shift, a smaller and more legible object than dignity at large.

The absolute list is no longer only a courtroom doctrine; it now has a living machine-facing instance. The European Union's Artificial Intelligence Act (Regulation 2024/1689), whose Article 5 took effect in February 2025, writes an enumerated list of outright prohibitions directly onto automated systems, banning practices such as social scoring and untargeted facial-image scraping under penalty rather than leaving them to be balanced case by case, which is the object-formula-as-rule posture tier one demands. It also meets the sharpest objection to any absolute list, that an absolute floor is also an absolute weapon, with a procedure rather than a patch. Article 5's list is amended only through a defined review, not by transient majority, so the enumeration grows the way this framework wants any inviolable list to grow, through a reading seat (EQ–06e) and never through a passing coalition. The honest limit rides along. Article 5 is a qualified floor, not a pure one. It binds through market law, most of its bans turn on significant harm or disproportionality, and it carries law-enforcement carve-outs that its own critics call a blueprint for legitimizing surveillance. It proves an enumerated machine-facing floor is buildable, not that any real one is yet clean.

TIER 1 · THE ABSOLUTE LISTtorture and threats of it · sacrifice of non-consenting innocents as a means · slaverydenial of legal personality · punishment without lawabsolute in every mode; anchored in plural consensusTIER 2 · THE SOCIAL MINIMUMsubsistence · shelter · health · education, lexically prior and resource-sensitiveburden-shift: scarcity must prove itself, in the openagent-facing dutiesnever against the holderremedies may stageabsoluteness attaches to the list, never to the word; the guardian interprets a list, not a mood
The plural floor. An absolute tier of enumerated prohibited treatments holds in every mode; the social minimum binds by burden-shift; and three rules keep both tiers duties about treatment rather than weights or weapons.

From the dignity of present persons the substrate extends outward in two directions that markets and ordinary politics both discount. Future persons will have dignity, so they hold standing now, which is where the framework's duties across generations come from, and the institutional seeds already exist in Wales' Future Generations Commissioner and the United Nations Declaration on Future Generations adopted in 2024. Non-human nature is harder, because identical dignity does not obviously transfer, but standing can still be granted through guardianship, as it already has been for the Whanganui River, for ecosystems under Ecuador's constitution, and for the Mar Menor lagoon in Spain. The framework treats these as standing held in trust rather than as claims of equivalence, which keeps the apex norm human while widening the circle of who is represented.

The harder question is how a foundation like this survives contact with an algorithmic legal system. Aeviterna's technical architecture pushes toward rules that execute automatically, and automation is exactly where dignity is most easily lost, because a system that optimizes people as data points treats them as means by default. Dignity therefore has to be encoded as a constraint on the machinery itself, not written as a preamble above it.

DIGNITY · inviolable constraint the engine cannot crossCaseRule enginerules as codeDecisionZK proof of complianceGuardiannature, future generationsholds a breachHuman review · explanation · appealcan overturn
The rule engine runs beneath an inviolable dignity line it cannot cross. Every decision carries a human review and appeal, a guardian can hold a breach, and compliance is provable without exposing private data.
Dignity as a hard constraintIn any automated decision, dignity-violating outcomes are forbidden regardless of efficiency, which is the computational form of non-balanceability. The optimizer runs inside inviolable limits it cannot price away.
Rules as code, not law as codeMachine-executable rules are treated as a draft execution of the law, always subject to a right to reasons, a right to human review, and an appeal that can overturn the code. Code proposes and due process disposes, in the spirit of the right not to be subject to a solely automated decision.
Human in commandNo fully automated action that materially affects a person's rights proceeds without meaningful human judgment in the loop.
Legal protection by designContestability and the rule of law are built into the architecture rather than bolted on afterward (Hildebrandt). This pairs with the structured-transparency posture in P–07, which lets a system prove it stayed inside its constraints without exposing the private data of the people it served.
Guardianship as a standing agentFuture generations and natural systems are represented in the machinery by designated trustees with real procedural power, including a hold on decisions that would breach an intergenerational or ecological limit, where the metabolism caps in EQ–02c are one such limit.

Dignity is contested terrain. Its content is read differently across traditions, it is invoked on both sides of hard bioethical questions, and the refusal to balance it has serious critics who hold that all rights must in the end be weighed. Making law machine-executable also risks freezing a single interpretation, which is why the rights substrate is bound to the falsifiability and the right to fork in EQ–10b rather than treated as settled, and the deeper question of who interprets the inviolable core is worked in FC–04 (The Guardian of the Core).

Open questions
  • Who fixes the content of dignity, and how the framework keeps a usable core while remaining plural, is unresolved.
  • Non-balanceability is powerful but contested. A system that never trades dignity away must still decide what counts as a violation, which is itself a judgment that cannot be fully automated.
  • Guardianship for nature and for future generations can be captured or can ossify, and the line between a faithful trustee and an unaccountable one is open (see EQ–05 on decaying standing).
  • Direction
    • Take human dignity as the inviolable apex, derive the rest from it, extend standing to future persons and to nature through guardianship (the nature half now specified in EQ–08e), and encode dignity as a hard constraint with due process wherever the law is automated.
  • Implementation sketch · exploratory
    • A constitutional-constraint layer that no automated decision may breach, rules expressed as code but executed under a right to human review and explanation, and standing agents for future generations and ecological limits wired directly into the decision path, grounded in rules-as-code practice (New Zealand's Better Rules; OpenFisca) and disciplined by its documented failures (the Dutch SyRI ruling; Australia's Robodebt).
  • The plural floor carries its surviving counter. Application-level indeterminacy is ineradicable, since even the aviation rule needed the innocent-perpetrator line, and the hard case still lands on interpreters. The other two counters are narrowed. The absolute-weapon objection now meets a procedure and not only the anti-paternalism patch. The enumerated list grows solely through a reading seat (EQ–06e), the discipline the EU AI Act's Article 5 already runs. And the thin-core worry gains a live enumerator, a machine-facing list in force, though a qualified one whose carve-outs keep it honest.
EQ–06d

Productive Conflict

Draft

A civilization is not measured by how much it agrees. It is measured by how much disagreement it can hold without breaking.

The anti-capture architecture protects dissent and the Justice of Light contains harm, but neither addresses the disagreement that does not resolve and should not be forced to. Deep value pluralism, the recognition after Isaiah Berlin that human goods are plural and sometimes incommensurable, means some conflicts have no synthesis. Liberty and equality, or loyalty and fairness, can pull in directions no reconciliation closes. A framework that treats every disagreement as a problem to be bridged will, at the limit, manufacture a consensus that flattens the very epistemic biodiversity EQ–03b exists to protect. Aeviterna needs a layer for the conflict that does not bridge.

The political theorist Chantal Mouffe named the distinction this layer turns on. Antagonism is the relation between enemies, the friend-or-foe logic in which the other side is to be destroyed. Agonism is the relation between adversaries, opponents whose right to their position is not in question even as it is fought, sometimes fiercely. On Mouffe's account conflict is not a failure of democracy to be engineered away but the condition of its existence, and the work of a healthy polity is to transform antagonism into agonism. Aeviterna adopts this directly. It does not resolve value conflict, it metabolizes it, building the channels through which deep disagreement stays generative instead of fracturing the society or being flattened out of it.

This locates productive conflict between two failures the framework already fears. Too much enforced consensus is the cognitive monoculture of EQ–03b, a civilization optimized into a single permitted view, brittle because it has bred out the variation it will need. Too much unmediated conflict is fracture, the friend-and-enemy collapse that ends in the harm the Justice of Light (EQ–06b) exists to contain. Agonism is the survivable band between them. The hard design point, the one Mouffe's critics press, is that the channel decays back toward antagonism the moment adversaries begin treating each other as enemies again. So the agonistic layer needs its own firebreak, a small core of shared commitment to the rules of contestation themselves that is not up for grabs. That is the one place the framework draws a hard line, because those who reject the right of others to contest cannot be granted the protections of contestation.

disagreement held, still generativedissent stays sayablethe one hard lineEnforced consensusAgonismAntagonismmonoculture, brittleadversaries, not enemiesenemies, fracture
Productive conflict is a band, not a point. Forced too far toward consensus a society becomes a brittle monoculture; left to unmediated conflict it fractures into the friend-and-enemy collapse. Agonism, after Mouffe, is the survivable middle, adversaries who fight hard without denying each other's right to exist.

The payoff is a new reading for the vital-signs panel. Alongside the other signs it already tracks, a civilization has a disagreement capacity, how much unresolved and fully expressed conflict it can hold without either fracturing or collapsing into enforced agreement. This is the political twin of epistemic biodiversity, and like biodiversity it is invisible until it is gone. A society where every contested question has quietly become unsayable is not at peace, it has lost the capacity the measure tracks, usually shortly before it discovers it needed the views it suppressed. The panel carries this as its seventh sign of eight, read from proxies such as the share of people who feel free to voice dissent, and held provisional because the measure is young.

Adversary, not enemyThe work of the polity is to transform antagonism into agonism, opponents whose right to their position stands even as it is contested, because conflict is the condition of a living politics rather than its failure (Mouffe).
Metabolize, do not resolveContested value questions that bridging cannot close are given standing venues, adversarial by design, with protected standing for minority positions and no requirement to converge, so passion is expressed and bounded rather than suppressed until it detonates.
The one hard lineThe rules of contestation themselves are the small shared core that is not contestable, since those who deny others the right to contest cannot claim its protections. This firebreak is what keeps agonism from decaying into antagonism.
Disagreement capacity as a signThe breadth of positions in real circulation is read on the vital-signs panel as a resilience reading, the political twin of epistemic biodiversity, watched for the quiet disappearance of the contestable.
Open questions
  • Where the line falls between an adversary to be engaged and an enemy who rejects the rules of contestation, and who is trusted to draw it, is the hardest residual.
  • Measuring disagreement capacity without the measurement itself becoming a tool to label dissent as fracture is open, which is why the seventh sign is held provisional.
  • Keeping agonistic channels from being captured by whoever benefits from a particular conflict staying unresolved is unresolved.
  • Direction
    • Build an explicit agonistic layer, after Mouffe, that channels irreducible value conflict into bounded contestation rather than forcing consensus, protected by a hard core of shared rules and read by a disagreement-capacity sign on the vital-signs panel (Mouffe; Berlin).
  • Implementation sketch · exploratory
    • Designate the contested questions that bridging cannot close as standing agonistic dossiers, each with an adversarial venue, protected standing for minority positions, and no requirement to converge, and track the breadth of positions in active circulation as the disagreement-capacity reading.
EQ–06e

The Reading Seat

Stable

The instrument is only as honest as the seat that reads it.

Every instrument in this framework ends in a seat where someone reads it, and five of the framework's organs build that seat: the guardian of the core interprets the constitutional floor (FC–04), a review reads the vital signs and sets the mode, an adjudicator runs the spine test (EQ–04b), a guardian panel speaks for an ecosystem (EQ–08e), and attesters certify a crisis trigger (EQ–09c). Five bespoke constitutions, one problem, and the framework's deepest conceded wound running through all of them. Whoever reads the instrument holds the polity's eyes, and the recursion residuals of OQ–03 and OQ–04, a capture that owns the indicator layer, an auditor no one audits, are all versions of a reader gone wrong. This organ extracts the constitution once, from the panel FC–04 already built and from a century of the world staffing exactly this job, and the other seats become instantiations of it.

The record is long enough to read. Committees read noisy signals better than individuals and are not slower, with engineered outsiders earning their seats empirically, their dissents predicting what insiders missed (Blinder and Morgan; Gerlach-Kristen). The reading binds unless publicly explained. For decades a central bank governor has owed a published letter, on a deadline, whenever the reading and the outcome diverge, and the counter-case shows the stakes, an advisory reading overridden silently on a question the readers were never asked, three of them resigning into the void (the open-letter regime; the Aduhelm record). The capture levers arrive one at a time and by name. Delete one appointer from a plural panel and the panel flips within two cycles; fail to capture by supermajority and come back for the budget; route around the plural appointer through an aligned court, and when the capture completes, the seat does not fake the instrument, it refuses to publish it, which is why the raw tallies are the tamper-evident seal (the election-body record). The two canonical failures share one feature, a reader funded by what it reads. The issuer-pays rating agencies stamped triple-A on what paid them and were named key enablers of the meltdown, and the anti-doping regime half-governed by sport's promoters reinstated the very program its own investigators had exposed (the FCIC; the McLaren record). And the repairs that work are method audits after a misread, not decapitations; expertise borrowed from the parties under exclusion rules, never litigation in the analysis room; the approval layer firewalled from the reading layer with every political edit publicly diffable; and lay watchers, whose immunity to career incentives is the point, holding the narrow junctures where capture concentrates, appointment, budget, and override (the Bernanke Review; the NTSB statute; the IPCC firewall; Bagg).

THE READING SEATthe seatone constitution,reusedplural, no single principaltenured past its principalsnever funded by the readraw instrument publishedreading firewalled from approvalneither regulated nor regulatorthe override disciplinewatched at the juncturesthe core (FC–04)the vital signsthe spine test (EQ–04b)an ecosystem (EQ–08e)crisis triggers (EQ–09c)five instantiations of one constitution; the seats watch each other in a circle, because the regress has no top
One constitution for any body that reads an instrument for the polity, extracted from the record of readers that held and readers that were captured. The five seats the framework had already built become instantiations, and the seats watch each other in a circle rather than a tower.
Plural, with no single principalAppointment runs through a panel no faction controls, part of the seat filled by lot with the lay majority positioned to watch rather than to model; deleting one appointer flipped a continental election commission within two cycles (van Ham and Garnett; the Indian record).
Tenured past its principalsTerms follow EQ–05d's tiers with the independence inversion where independence is the product: one long non-renewable mandate, the clock running on the review powers around the seat; dissenters carry tenure and immunity (the comptroller-general design; the Lavasa record as the failure).
Never funded by the readThe seat's budget is set at a different layer than the one it reads, on a formula, because issuer-pays is the canonical capture and budget strangulation is the vector parchment does not stop (the FCIC; the INE record).
The raw instrument publishedVotes attributed, dissents published, underlying readings released in full. A seat that will not show the instrument is presumed captured, the precinct-tally rule (the Carter Center record).
Reading firewalled from approvalWhatever politics approves sits atop readings it cannot edit, and every approval-layer edit is publicly diffable against the untouched reading (the IPCC firewall).
Neither the regulated nor the regulatorExpertise is borrowed from the parties under exclusion rules; liability interests and the seat's own institution are barred from the analysis room (the NTSB statute and party rules).
The override disciplineThe reading binds unless publicly overridden: same question, published reasons, a named official, a fixed deadline, the open-letter rule generalized (the Bank of England letters; the Climate Change Act response duty; Aduhelm as the failure).
Watched at the junctures, audited after misreadsA lot-selected panel holds appointment, budget, and override; a scheduled outside audit rebuilds the methods after a documented misread; and the seats watch each other in a circle, not a tower (Bagg; Hurwicz; the Bernanke Review).
The contestable modelThe seat publishes not just its reading but the model and the named assumptions behind it, scores its own past misses on a public accuracy ledger, and runs a scheduled adversarial collaboration whose dissent is a reputational bet, so a shared bad model becomes contestable before the miss rather than only autopsied after. The Bank of England's review of the transitory-inflation error prescribed exactly this, to publish alternative scenarios, decompose forecast errors, and drop the buried assumption that long-run expectations are always anchored (the Bernanke Review; Tetlock).

The five seats are instantiations of this one constitution, each keeping only its particulars. The guardian of the core keeps its full treatment in FC–04, the hardest case, from which this constitution was extracted. The vital-signs review is the seat pointed at the panel, with FC–04's framing-layer disciplines as its particulars. The spine adjudicator of EQ–04b is the seat pointed at the admission test. The more-than-human guardian of EQ–08e is the seat pointed at an ecosystem, with custodial knowledge in the expert chairs and the open-standing backstop as its particular. The crisis attesters of EQ–09c are the seat pointed at the ignition triggers, with the escalating renewal ladder as theirs, and their rule that renewal is held by the unrewarded is the same discipline as this organ's funding firewall, money and continuation kept apart from the read.

The seat's deepest reader is the one who frames the question, and the framework makes that reader carry a falsifier too. A frame is not admitted until it is pre-registered and dated. It must predict the near-term signals that run ahead of a transition, not the far outcome that never resolves in time. Those are the rising variance and slowing recovery that resilience science reads before a system tips (Scheffer and colleagues). They are scored against fixed and plural sources, and against a holdout kept back until after the frame is registered, so the grading does not run back through the frame it grades.

The seat stakes a bond on the frame and its predictions, forfeited when they fail. That stake is the seat's own and rides on the record, never posted to or collected by what it reads, so the bond and this organ's funding firewall are one discipline seen twice, money and continuation kept apart from the read. Its mandate decays and is re-based on the record, so a seat that reads the world well earns renewal while one that reads it badly decays out on the evidence rather than on trust. Renewal on the record governs only the scoreable operational reading, the object-level forecast reality can grade on the first three tiers of EQ–05d. The framing apex, whose frame reality cannot score against itself, takes that section's Tier 4 inversion instead, one long non-renewable mandate with its seat filled by lot (FC–04), because re-earning there would be reappointment and no half-life reconciles must-re-earn with must-not-please. A rival methodology is funded, forkable, and argued against the sitting one, because a bad frame is usually easier to refute than to sustain. This binding of the framing layer to a dated, scored forecast is the anchor. The register holds it as F–08. If anchored seats show no better calibration and no less capture than unbound seats across a decade of the tracker, the anchor buys nothing.

Its grip is uneven across the signs. A substrate-real sign, ecological throughput or material security or biological resilience, scores against outcomes directly. A reflexive sign, legitimacy or trust, whose reading helps constitute what it reads, can be bound only to its non-reflexive downstream, to compliance, exit, and enforcement cost. Even that runs out, and where it does the sign is flagged.

Still, where it grips, the anchor halts the oversight regress a third way. It does not climb to an apex that goes unwatched, and it does not only close the horizontal circle of mutual watchers who can share a frame. It binds the frame to reality, the one watcher that cannot be asked who watches it and does not share the framer's priors. Binding to reality never licenses reading past the cap. The anchor is fed at the layer of institutions, outcomes, and dated forecasts, where the legibility budget of EQ–03b defaults to a person-level zero, and a frame that needs individual detail spends from the same published budget as any instrument. The cap governs person-level leakage; the anchor is fed above it.

THE CONTESTABLE MODEL the reading the model + named assumptions now published, not buried assumptions on the record the buried premise made visible and so contestable public accuracy ledger past misses decomposed and scored, in the open a scheduled red-team adversarial collaboration, dissent as a reputational bet the miss contestable BEFORE the miss the three repairs above, ex ante method audit AFTER the only prior repair, ex post it does not dissolve shared error: correlated shocks stay hard, and a red-team from the same profession shares the priors THE BERNANKE REVIEW, 2024 · FORECASTING TOURNAMENTS
The contestable model. The seat publishes its model and named assumptions, scores its own past misses on a public accuracy ledger, and runs a scheduled red-team, so a shared bad model becomes contestable before the miss rather than only audited after, narrowing undetected shared error without dissolving it.

The seat also gauges its own capture. The concentration of framing power, how few and how unwatched the readers have become, is held in a band with its own decay clock. That band is the framework's one discipline turned inward on the reader. This is not a ninth vital sign, and it does not join the panel of eight, because the eight read the civilization and this reads the reader.

Its own residual is named at once. Defining the gauge is itself a framing act, held the only way framing is ever held, by a rival gauge kept live beside it. Where that rival is captured or never funded, the gauge is circular, and that failure is one of its recorded ways to break.

Framing-power concentration rising diffuse captured DRIFTING, WATCHED the seat's reading on itself
The reader's self-gauge. Framing-power concentration held in a band, in the vital-signs style.
Open questions
  • A shared bad model defeats the capture defenses. The best-built committee still missed the 2021 inflation surge, a miss correlated across every central bank, so a red-team drawn from the same profession shares the same priors. The anchor narrows it to two cores no test reaches, the pre-propositional choice of what is salient and the reflexive sign that helps constitute its own reading.
  • The anchor grades only what shows an early signal, so the warning-free transition, the fast or noise-driven tip, is uncovered by construction.
  • The gap between formal and real independence is measured and wide, and budget strangulation in particular defeats parchment; the funding firewall narrows this without closing it.
  • Published dissent disciplines a seat and politicizes it in the same motion, and transparency of tampering bounds tampering without preventing it.
  • Direction
    • Constitute every instrument-reader in the framework as a reading seat, instantiate rather than re-invent, and carry the record's last lesson as a standing property. A reading seat is never finally held, only currently defended, and the seat that reads all seats does not exist, the circle of mutual watching standing as the honest substitute.
EQ–06f

The Enforcement Ladder

Draft

An architecture that forswears the sovereign still needs its rules to bind, and nearly every precedent this framework cites binds through public force, bylaws a court will enforce, statutory fines, entrenched constitutions, title law. The honest move is not to deny the borrow but to name it, organ by organ, the way every section already names a stability marker. Enforcement is not one thing. It is a ladder of five sources, reputation, vested stakes, structural locks, federated sanction, and borrowed courts, each with a real record and a failure edge, and the framework's discipline is that every organ states its rung, borrowed or own, so no reader has to guess what stands behind a rule.

The borrowed rung is older and stronger than it looks. An association's constitution has carried its own arbitration forum since a sheep-breeders' dispute in 1915, when a member's lawsuit was stayed because the articles sent member disputes to arbitration; company law in the major families holds a registered constitution to be a contract binding the body and each member; and the New York Convention obliges the courts of 172 states to refer parties to arbitration and enforce the awards. The full stack already runs twice over, in the religious tribunals whose awards state courts enforce under ordinary arbitration law and in sport's standing transnational court. A cell that encodes its charter as enforceable bylaws and seats an arbitration path in peacetime has invented nothing; it has joined a hundred-year-old practice. And the alternative is not independence. The newest governance experiments that declined to choose a legal form had one chosen for them, unincorporated association or general partnership, members personally liable, which is the borrow arriving on the worst available terms.

The lender sets conditions, and they run on a gradient that cuts against ambition. Courts defer most where the stakes are spiritual, refusing to review a hierarchical religious tribunal even for arbitrariness, and least where an association gatekeeps a livelihood, where expulsion draws review in every major family and penalty doctrine polices any forfeiture for proportion. The space that remains is workable, since a forfeiture protecting a legitimate governance interest can stand even beyond provable loss, but the planning fact is permanent: an economically consequential cell will not enjoy the church's autonomy, and its internal discipline must be drafted to survive the hostile review it will get. The lender can also foreclose, by statute where a legislature dislikes a private ordering, or entirely, when the state that lends the courts joins the siege, at which point enforcement falls to the inner rungs, which the record supports only below the threshold where a great power takes an interest (EQ–04b; EQ–04f).

The inner rungs hold to the edge of what the community controls. Reputation with monitoring and graduated sanctions is well supported and cheap, and it breaks against exit options and outside capital, as the canonical trust networks learned. Vested stakes move behavior at the margin and bind only inside judicially policed proportion. Structural locks, the veto share, the purpose trust, the software default, enforce almost for free and are blind to insolvency and dependent on a watcher. Federated sanction binds in exact proportion to what the federation gatekeeps, eleven days to compliance where it holds the only game, a shrugged-off label where it does not. The map below states the rung for every load-bearing organ, which is this section's discipline made auditable, and the residual is stated with it: between polities there is no court, and the outermost rung is the federation's own sanction, where the boundary layer and the economic wall take the watch.

Rung one, reputationMonitoring and graduated sanctions in the Ostrom pattern, already specified in EQ–05 and EQ–06b; real and priced in the record, and breakable by exit options and outside capital, so it is never load-bearing alone.
Rung two, the vested stakeForfeitable within proportionality review, drafted to a legitimate governance interest rather than to punishment, on the pattern the penalty doctrines of every major family will actually uphold.
Rung three, the structural lockThe veto share, the purpose trust, the automated default; entrenchment at commodity price, blind to insolvency, and owed a named watcher, because an unmonitored restriction decays into paper.
Rung four, federated sanctionRecognition, clearing, and standing withdrawn across cells, with the fork protocol (EQ–10b) as the final sanction; binding in proportion to what the federation gatekeeps, and priced accordingly.
Rung five, the borrowed courtThe charter encoded as enforceable bylaws, an arbitration path seated in peacetime, awards traveling on the Convention; named as borrowed wherever it stands, with the lender's conditions and revocability stated.
OrganRungsThe source, borrowed or own
EQ–02d · Monetary ecologyOne, fourOwn. Standing thresholds bind by federated recognition; clearing runs on member contract.
EQ–02e · Commonwealth trustThree, fiveMixed. The veto share is own structure installed through borrowed company law; the charter binds as bylaws.
EQ–02f · Governance overheadOneOwn. A published band read at the review; breach is a reading, not a writ.
EQ–03b, EQ–03e · Legibility, drillsOne, threeOwn. Budgets and calendars enforce by default and publication.
EQ–04e · Consent and exitTwo, fiveMixed. The stake forfeits within proportionality review; exit and portability are contract the courts enforce.
EQ–05 · Standing engineOne, two, fourOwn. Monitoring, graduated sanctions, and the stake; expulsion binds through the federation.
EQ–06, EQ–06b · Anti-capture, justiceOne, four; five behindMixed. The tiers bind internally; where livelihood or stake is touched, the borrowed court reviews, and the design expects it.
EQ–06e · Reading seatsTwoOwn. The staked bond forfeits on the record.
EQ–07 · Knowledge commonsThree, fiveBorrowed. Licenses enforced by the copyright courts the commons already uses.
EQ–09 · Governance modesOneOwn. Mode-setting binds by published reading and challenge, with no marshal behind it, which is the honest edge.
EQ–10b · Fork and successionFourOwn. The fork is the sanction; legitimacy travels with the record.
Between polities · EQ–04c, EQ–04fFour onlyOwn. Reciprocity and federated sanction; no court exists, and the walls are the answer.
1 · reputation breaks on exit options 2 · the vested stake capped by proportion 3 · the structural lock blind to insolvency 4 · federated sanction only what it gatekeeps 5 · the borrowed court revocable by the lender THE LENT FLOOR · RECORDING, COURTS, EXECUTION every organ names its rung, the way it names a stability marker
The enforcement ladder. Five sources of binding force, each with its failure edge (coral), standing on the recording and court layer the polity does not own. The gradient is the planning fact: deference shrinks as economic significance grows, so discipline is drafted for hostile review. F–13 tests the ladder against handshake governance.
Open questions
  • The absences are named. No reported decision enforces a Convention award on association membership in either direction, the statutory wrappers and golden shares are young and judicially untested, and no study links member equity to compliance or survival, which F–13 assigns the tracker to start counting.
  • The gradient is permanent, not a gap. Deference shrinks as economic significance grows, so the framework cannot both matter and expect the church's autonomy, and every sanction it designs must survive the proportionality review it will eventually meet.
  • The lender can decline to lend, or join the siege. Enforcement then falls to the inner rungs, which the record supports only below the great-power threshold the thin spine names, and that residual is shared, unbridged, with the economic wall.
  • Direction
    • State each organ's rung beside its stability marker as sections revise; write the arbitration path into the model charter at its next release; count enforcement outcomes and breach resolutions in the adoption tracker.
Falsifier F–13

If adopters that pre-write the borrow, governance encoded as enforceable bylaws with a seated arbitration path, resolve documented internal breaches and survive them at rates the tracker cannot distinguish from handshake-governed peers, the fifth rung buys nothing and the bootstrap claim fails. If, within adopters, organs protected by stakes, locks, or federated sanction show compliance and survival no better than organs held by reputation alone, the ladder above the first rung is decoration and the organ is withdrawn. Ratified July 21, 2026.

EQ–06g

The Definer's Seat

Draft

Randomize the jury and you have still not touched the indictment.

The framework staffs its reader and decider seats by lot, because a randomly drawn seat cannot be bribed, lobbied, or made to serve a career. That secures who sits. It leaves untouched a prior question the lottery cannot reach: who decides what the seated body is asked, which options it may choose among, what evidence it hears, and who briefs it. That upstream work, the definer function, is done by a standing apparatus: a commissioning authority, a secretariat, a chosen roster of experts. A durable faction that owns that apparatus can determine the outcome of a perfectly random body, because it has already set the question and pruned the menu. This is not the value pluralism EQ–06d protects, the legitimate contest between adversaries over goods that do not reconcile. It is the capture of the frame itself.

The record names the lever. Schattschneider called the definition of the alternatives the supreme instrument of power, and Bachrach and Baratz called its exercise the second face of power, the non-decision that keeps a threatening option off the agenda before any vote is taken. Madison held that faction cannot be abolished, only counterposed, and Michels found that organization itself breeds an entrenched leadership even in the bodies most committed to internal democracy. The deliberative record confirms it in practice. In the celebrated citizens' assemblies, Ireland's and France's climate convention among them, the frame was set by the parliament, the secretariat, or the state-aligned experts, and not by the random members, and where the menu was one-sided the citizens were left with the impression that no real choice was on offer. Random selection of the jury is beside the point when a durable interest owns the definition of the alternatives.

SORTITION FIXES THE SEAT, NOT THE FRAME The definer question, options, experts one-sided menu Sortition seat random, unbribable the outcome unrandomized: the capturable point randomized: incorruptible the jury is drawn by lot; the indictment is not, so a faction that owns the menu owns the result THE DEFINER'S SEAT · EQ–06g · GUARD THE FRAME THE SAME WAY Framers A, by lot Framers B, by lot Red team: the omitted option a plural menu firewall the same seat random, unbribable a real choice provenance published: who framed, who funded a watcher reads the concentration gauge
Sortition guards the seat, not the frame. The lottery makes the deciding seat incorruptible, but the definer and the menu upstream are unrandomized and capturable, so a faction that owns the menu owns the result. The organ guards the frame the same way it guards the reader. The framing body is itself drawn by lot in adversarial teams with a red team for the omitted option, the plural menu is firewalled from the choice, and a published provenance trail and a lot-drawn watcher hold it in the same circle.

So the definer earns a seat of its own, held to the discipline the reader already carries. The framework cannot abolish the definer any more than Madison could abolish faction, so it does to the frame what it does to the reader. It multiplies it, counterposes it, exposes it, and denies it tenure.

Draw the framer by lot tooThe agenda-setting and question-framing body is itself sortition-drawn with decaying, non-renewable standing, so no faction owns the agenda by default (the Ostbelgien council). Necessary and not sufficient, since it relocates the residual to the secretariat that services it.
A plural, adversarial menuNo single body owns the option set. At least two independently drawn framing teams, and a red team whose only task is to add the strongest omitted option, produce the menu, because conflicting perspectives dissolve a frame that a one-sided menu would impose (Druckman and Nelson).
Generation firewalled from choiceThe body that generates the options is separated from the body that chooses and from the body that implements, mirroring the reading seat's firewall. The framer is never the decider.
No tenure for the frameEvery framing role, secretariat, expert-selector, agenda-chair, briefing author, is held once and not again, because durability is what turns an interest into a faction.
A published provenance trailWho drafted each question, who chose and who funded each expert, and which options were dropped and why, are on the public record, the reading seat's raw-record rule carried upstream, where the absence of the record is itself the alarm.
A watcher on the frameA lot-drawn audit reads the provenance for factional concentration and directional pruning and reports a framing-power concentration gauge, so the definer is watched in the same circle, with no exogenous top.
Open questions
  • Every layer is capturable in turn. Drawing the framer by lot relocates power to the secretariat that briefs it, the oversight regress the framework concedes has no exogenous top.
  • A shared bad frame defeats plural framers exactly as a shared bad model defeats plural readers, so the reality-anchor residual of EQ–06e reappears one level up.
  • The line between a legitimate aggregator advancing an option and a faction capturing the whole menu is itself contestable, which is EQ–06d's adversary-and-enemy problem returning at the framing layer. There is no unframed question, so the definer's seat is never finally held, only currently defended.
Falsifier F–16

Using the provenance trail the organ mandates, the definer is captured, and the design false, if across a decade a single durable network drafts the question, selects a majority of the experts, or authors the briefing for sixty percent or more of a polity's sortition exercises, and the menus actually presented systematically omit the highest-support alternatives an independent panel surfaces, in that network's favor. A working mechanism predicts no such systematic, faction-favoring divergence between the curated menu and an independently elicited one. Carried as F–16.

EQ–07

The Knowledge Commons

Stable

Knowledge unshared is power wasted. Knowledge privatized is power weaponized.

Aeviterna treats knowledge as primary civilizational infrastructure, rather than as a product, a commodity, or intellectual property. It belongs in the same category as water, roads, and electrical grids. Its privatization produces civilizational fragility regardless of its short-term economic logic.

The Knowledge Commons functions as shared civic memory, open scientific collaboration infrastructure, distributed educational architecture, transparent governance archive, and collective resilience repository. These are not separate institutions. They are facets of a single claim, that a society's collective understanding of itself and its world is a shared resource that cannot be enclosed without consequences that eventually reach everyone.

Aeviterna seeks to transition from scarcity-centered value systems, in which knowledge is valuable because it is withheld, toward contribution-centered coordination systems, in which understanding itself becomes a foundational form of wealth. This is not idealism. It is an observation about where value actually comes from in systems that survive.

Made operational, the commons is governed, not merely open. Knowledge is a shared resource subject to social dilemmas, analyzable with the same institutional discipline as fisheries and forests (Hess and Ostrom), rebuilt for constructed commons whose participants produce the resource and whose boundaries are built rather than found (Frischmann, Madison, and Strandburg). The working institutions converge on one shape, a written constitution with elected, recallable authority earned through organization-building rather than output alone (O'Mahony and Ferraro), a hard disclosure clock that outran enclosure at civilizational scale, genome data public within twenty-four hours (the Bermuda Principles), and shared infrastructure that makes release the default rather than a virtue. The documented failure modes sit at maintenance and succession, not creation: a decade of editor decline traced to newcomer-hostile quality automation and calcified policy (Halfaker and colleagues), critical infrastructure carried by unpaid volunteers whose burnout is a systemic risk (Eghbal), and the patient attacker who social-engineered a burned-out solo maintainer into handing over the keys to most of the world's servers. Voluntary pledges are the weakest instrument in the record, suffering obscurity, impermanence, and non-binding transfer (Contreras).

The commons therefore adopts three commitments.

  • The disclosure clock. Knowledge produced with commons resources enters the commons on a hard, short, automatic deadline, release as infrastructure default, not author discretion. This is the Bermuda mechanism generalized, the one anti-enclosure device with a civilizational-scale win on its record.
  • Impact-metered pull funding. The commons pays for knowledge with pre-committed purchases, prizes, and ex-post impact payments, and buys out enclosures at auction prices where they block the floor (Kremer). The advance market commitment that vaccinated tens of millions of children is the working precedent, reward decoupled from monopoly, results entering the commons on delivery.
  • Constitutional stewardship with a paid maintenance line. Every commons institution carries a written constitution with earned, recallable, bounded authority, plus the two clauses the failure record demands: a funded maintenance and succession line, because commons die of unpaid upkeep, and newcomer-retention duties with automation audited for first-contact hostility, because commons also die of their own immune systems. The clause binds to the calibration perimeter (EQ–05d) and the memory organ (EQ–07b).

The enclosure pressure this answers is current, not historical. Five publishers hold more than half of published research (Larivière and colleagues), the open web is closing against crawlers while incumbents sign exclusive training-data deals (Longpre and colleagues), and the pattern was named two decades ago as the second enclosure movement (Boyle).

Falsifier

If commons institutions built on these three commitments show maintenance failure, capture, or newcomer collapse at rates indistinguishable from unconstituted volunteer projects over a decade, the operationalization is wrong.

Open questions
  • The mechanism inherits the free-rider problem it documents. The disclosure clock and the maintenance line presuppose the commons treasury located in the economics of enough (EQ–02b), and this section does not re-derive it.
  • Quality control in an open commons is real but unsettled terrain. The strongest head-to-head found error parity with a disputed methodology, and the honest reading is parity on facts with fragility in review capacity.
EQ–07b

Civilizational Memory and Succession

Draft

A civilization is a relay. Drop the baton and the next runner starts from nothing.

The Knowledge Commons (EQ–07) holds what a civilization knows, but knowledge is a flow that must be re-transmitted each generation or it is silently lost, and collapse, in Tainter's reading, is precisely the loss of the complexity that knowledge embodied. After the Bronze Age collapse, writing itself fell out of use for centuries; the Antikythera mechanism's sophistication was not matched again for more than a thousand years. Memory is not self-sustaining.

The hardest part is tacit. Polanyi showed that we know more than we can tell, so a complete record cannot reconstitute a capability without the people who hold it, and automating the novice tasks through which experts are trained can quietly break the apprenticeship that carries skill across generations. Records themselves rot, through the bit rot and format obsolescence of a digital dark age, and even long-horizon archives are fragile, as the flooding of the Svalbard seed vault in 2017 showed.

The most careful work on very long-term transmission, the nuclear-semiotics marker studies, builds warnings on a defense-in-depth principle of redundant materials and languages while admitting that no guarantee survives ten thousand years. The curation question is a power question, since whoever decides what to keep can entrench a partial record, and the opposite danger, total memory, harms reinvention, so a succession layer needs a deliberate, accountable practice of forgetting as well as keeping.

G1G2G3re-transmit each generationthe link breaks, knowledge is loststarts from nothingdistributed archiveredundant copies, periodic restoration drills
Knowledge re-transmitted across generations, backed by a distributed archive. A broken link loses what was not handed on, which is why the archive holds redundant copies and the society runs restoration drills to prove the knowledge is still alive.
Distribute the archiveRedundant, substrate-diverse copies so no single vault is a point of failure or capture (defense-in-depth; the Svalbard lesson).
Preserve capability, not only recordsApprenticeship for load-bearing tacit skills, because tacit knowledge cannot be archived (Polanyi).
Restoration drillsPeriodically rebuild something from the archive to prove the knowledge is still alive rather than merely stored.
Accountable forgettingA deliberate deaccession rate, with curation routed through deliberation rather than an executive, so the record is curated rather than hoarded or captured (Mayer-Schönberger).
Open questions
  • What to preserve and who curates are value judgments that encode a politics, and the curator is a capture target like any other.
  • Stewardship institutions are not failsafe, so durability has to come from redundancy rather than from any single archive.
  • Tacit knowledge resists archiving by its nature, so preserving capability needs living people, not only files.
  • Direction
    • Add a succession layer to the knowledge commons, distributed and anti-capture, that preserves capability as well as records and forgets on purpose, entrenching the duty to re-transmit while keeping contents and curators revisable (Tainter; Polanyi; the nuclear-semiotics studies).
Part IVThe Human Stake
EQ–08

Human Time and Curiosity

Stable

A civilization that eliminates wonder in pursuit of efficiency has optimized itself into spiritual collapse. But the loss of wonder is not merely a spiritual failure. It is a structural one.

Aeviterna does not exist merely to optimize systems more efficiently. Its purpose is to reclaim human cognitive bandwidth from survival anxiety, extraction pressure, and perpetual optimization culture. And reclaiming human time is not a secondary benefit of getting the economics right. It is not quality of life. It is architecture.

Human curiosity, play, rest, and exploration are not personal goods that a well-functioning civilization happens to enable. They are the mechanisms by which a civilization remains capable of imagining futures that its current optimization targets cannot generate. When a society eliminates the space for non-instrumental thinking, it does not merely become less enjoyable. It becomes less capable of the kind of cognitive mutation that adaptive systems require to survive changing conditions.

This connects directly to EQ–03b. The reclamation of human time is one of the primary mechanisms by which cognitive ecology is maintained at the individual level. Automation, AI, and advanced coordination systems should reduce unnecessary labor burdens not because leisure is pleasant (though it is) but because the distributed exploration that happens in unstructured human time is irreplaceable civilizational infrastructure. It cannot be scheduled or assigned. It requires the protected space that extraction systematically destroys.

A framework that produces material security but eliminates the space for curiosity, play, and self-determined meaning has failed at the civilizational level regardless of its output metrics. The measure of success is not efficiency. It is whether the civilization retains the capacity to be surprised by itself.

The residual, named

Stated as raw quantity, the claim is half-wrong, and the correction sharpens it. Working hours roughly halved over a century and a half, and measured leisure rose by hours a week across recent decades; the gains flowed to the less-educated while high earners intensified work and busyness became the elite status signal, an inversion of Veblen (Aguiar and Hurst; Gershuny). The fifteen-hour week arrived nowhere (Keynes; Pecchi and Piga). What failed is distributional and qualitative, the texture and sovereignty of time rather than its amount, and the quantitative core of this chapter is carried by temporal sovereignty (EQ–09b). The falsifier is that broad-based convergence of discretionary time across income strata, without this framework's mechanisms, would show the market delivering what this chapter says it cannot.

EQ–08b

Meaning and Belonging

Draft

A civilization has to answer what a life is for, and it has to give people somewhere to belong. Markets answer the first with consumption and the second with very little, and a society that beats scarcity on goods while failing on meaning still decays from the inside. This is the organ most post-capitalist thinking leaves out, because most of it stops at distribution. Aeviterna treats meaning and belonging as infrastructure, on the same footing as energy or law.

The evidence that material success does not supply meaning is solid. Kasser's work shows that strongly materialistic goals track lower wellbeing rather than higher. The larger claims are more contested, and the framework states them carefully. There is a wide discussion of a meaning crisis in secular societies, which is a philosophical framing rather than a measured fact. There is real evidence of rising isolation, including the United States Surgeon General's 2023 advisory on loneliness, though whether connection has fallen over the long run is disputed. And there is the deaths-of-despair literature, whose interpretation is actively debated. The framework leans only on what holds, that belonging is a determinant of health and that consumption is a poor substitute for meaning.

A personConsumptiona poor substituteContributionto the commonsRecognitionstanding that decaysbelonging and meaning renew
Consumption is a dead end for meaning. Recognized contribution to the commons, with funded belonging, forms a renewing loop.
Meaning through contributionStanding accrues for contribution to the commons (EQ–05 and EQ–07) and curiosity is protected as an end in itself (EQ–08), so a person can build a life of recognized worth that does not run through consumption.
Belonging as infrastructureThird places and federated membership are funded and protected the way roads are, giving people durable settings to belong to (Oldenburg's third places, Klinenberg's social infrastructure). The Praxis cell is the framework's first unit of belonging.
Education as formationSchooling is oriented toward self-formation and judgment rather than toward sorting people for a labor market, recovering the older idea of self-cultivation against pure credentialism (Sandel).

Meaning is the organ most easily corrupted by the state, because a civilization that manufactures meaning from the top slides quickly into coercion. The framework therefore commits to cultivating the conditions for meaning rather than dictating its content, and the dignity constraint in EQ–06c is what forbids any official or imposed meaning. Belonging is built, not assigned.

That commitment now has a mechanism, and the mechanism has a century of precedent. Denmark has funded meaning infrastructure content-neutrally since 1851, folk high schools and free associations supported by statutory duty, the pivotal clause leaving the association itself to choose its subjects, teachers, and form, exams prohibited so credentialism cannot annex the sphere, all of it inside a procedural floor of democratic structure, open membership, and non-profit operation (the Folkeoplysningsloven). The funder audits form, never belief. The adversarial record is just as legible. Sweden's parallel system paid on self-reported volume and bought phantom participants, the national audit finding deficiencies in every link of the control chain, followed by a values-floor fight and a one-third funding cut that fell on honest associations too (Riksrevisionen). France's individualized training accounts bred industrial fraud within two years, cured only by reintroduced friction. The design rules follow from the comparison: fund associations, never individuals; cap the subsidy share so every container carries member co-payment; verify by sampled attendance, never by continuous monitoring.

Measurement has its own precedent. Belonging can be read behaviorally and in aggregate. Twenty-one billion friendship ties, published only as privacy-protected aggregates by place, yielded the strongest known community-level predictor of upward mobility, economic connectedness, the share of ties that cross class (Chetty and colleagues). No individual can inflate an aggregate bridging statistic the way self-reported study hours were inflated, so the framework reads belonging at the cell and federation level, from participation breadth, bridging ties, and venue density, and never lets a unit-level reading enter a funding formula. Money follows form; measurement reads behavior; the two are kept apart by design, which is what makes the Goodhart question answerable rather than fatal.

THE CONTAINER RULEpublic fundingby formula, capped sharethe form auditdemocratic · open · non-profitno anti-democratic purposethe containerassociation · third place · schoolmember co-payment insidecontentpeople choose itthe firewall: the funder never reaches contentthe belonging readingparticipation breadth · bridging ties · venue densityaggregate only, cell level and aboveinforms deliberationnever enters the funding formulathe tripwire: no single provider or creed may exceed a concentration share of funded infrastructuremoney follows form; measurement reads behavior; the two never touch
The container rule. Funding flows by formula through a form audit to self-governing containers, and a firewall stops the funder reaching content. The belonging reading is aggregate and behavioral, informs deliberation, and never enters the funding formula, so neither the funder nor the metric can own what a life is for.
Fund the container, never the contentPublic money flows only to self-governing membership associations and places, by formula, with statutory content freedom inside a procedural floor (democratic structure, open membership, non-profit operation, the dignity floor). The funder audits form, never belief, and the state may never operate the containers itself (the folkeoplysning settlement).
Friction by designAssociational rather than individual funding, a capped subsidy share so every container carries co-payment, and verification by sampled attendance. The Swedish and French fraud records made standing rules (Riksrevisionen; the CPF record).
Measurement decoupled from moneyBelonging is read in aggregate from behavior, participation breadth, bridging ties, venue density, at cell level and above, under a stated legibility budget (EQ–03b), never surveyed with targets attached and never a funding-formula input, so no unit can farm the metric (Chetty and colleagues).
The non-imposition tripwireIf any single provider, creed, or network exceeds a concentration threshold of funded belonging infrastructure, disclosure and rebalancing are forced, the same graph-watching device EQ–06 turns on office-holders, pointed at meaning.
Open questions
  • Whether a society can cultivate meaning without imposing it remains the constitutive tension. The container rule makes the commitment auditable, and conditions still shape content statistically, since what gets a room and a subsidy exists more. The organ holds this in the open rather than claiming to dissolve it.
  • The values floor moves, and no mechanism decides from inside whether a given exclusion is boundary maintenance or majority capture; Denmark's anti-democratic-purpose clause and Sweden's democracy conditions are the same instrument at different temperatures, and the framework's answer, contest before an outside layer (EQ–06, FC–04), mitigates without closing this.
  • Neutral conditions-funding selects for already-capable communities (the UK fund's own evaluation found deprived areas struggled with the application burden), so the tilt that corrects the Matthew effect is itself a content-adjacent judgment; and the belonging reading depends on behavioral traces someone owns, so it must be buildable from public and federated traces, never a platform's.
  • The social-capital decline thesis is contested, so the framework rests the case for belonging infrastructure on its health and civic value rather than on a claim of collapse.
  • Direction
    • Supply the conditions and never the content, through the container rule: formula funding through a form audit into self-governing containers, friction by design, an aggregate behavioral belonging reading decoupled from money, and a non-imposition tripwire, all under the dignity constraint.
EQ–08c

The Care and Maintenance Substrate

Stable

The economy everyone measures floats on an economy no one pays for.

A civilization runs on more than energy and materials. It runs on the daily, mostly unpaid work of raising children, tending the sick and the old, and keeping existing systems in repair. The economy of enough and the metabolism govern throughput; this is the human upkeep that throughput depends on, and the framework has so far assumed it rather than named it.

The scale is measured, not asserted. The International Labour Organization values unpaid care work at roughly nine percent of world output, the equivalent of two billion people working full time without pay, with about three-quarters of the hours done by women, and national accounts treat it as outside the production boundary, so the work that sustains the workforce is invisible to the measure of value. Nancy Fraser names the result a crisis of care. An economy oriented to unlimited accumulation destabilizes the very reproduction it relies on.

The same underpricing runs through physical systems. A culture fixated on innovation undervalues the maintenance and repair that keep infrastructure and institutions working, and deferred maintenance is the physical twin of the crisis of care, a debt that accumulates quietly until something fails. Care and maintenance are one commitment seen in people and in things. Keep what sustains us in good repair.

the waterline of what markets pricemeasuredmarket economyCare and maintenancesocial reproduction, unpaid and unpricedabout 9% of world output if countedabout 76% of the hours done by womenthe foundation holds up the part that gets measured
The economy as an iceberg. The measured market sits above the waterline; the unpaid, unpriced work of care, reproduction, and maintenance is the larger mass below that holds it up.
Count the foundationMake unpaid care and maintenance visible as infrastructure rather than residual cost, valued without being reduced to a market price (ILO; Waring).
Maintenance as the twin of careTreat deferred maintenance of shared infrastructure as a debt with the same standing as the care deficit (Russell and Vinsel).
Provision the everydayGovern the foundational systems of daily life, care, water, housing, and utilities, as a public foundation (the Foundational Economy).
A care-and-maintenance readingA candidate sign combining the unpaid-care burden and its distribution with a maintenance-debt indicator, read under a stated legibility budget (EQ–03b) so the sign can be exact about populations while structurally unable to see persons.

The substrate now has its instruments, and the counting question is settled by where in the aggregation hierarchy the counting happens. The national-accounts world drew the line in 2025. Unpaid household services stay outside GDP, and recognition lives in extended accounts compiled beside it, diary-based, active and supervisory care distinguished, valued at a stated lower bound (the SNA 2025 revision). Two ledgers, commensurable and never merged, which is this organ's "valued without being priced" made operational. The record adds a warning and a name. The warning is that optional ledgers starve, one national account went seven years between editions and another has not been updated in half a decade, so publication needs a clock. The name. The two counting designs are now both on record. The first is the anonymous sampled time-use diary with statutory confidentiality, aggregates only, the instrument beneath every satellite account. The second is the per-transaction verification log, which one country mandated for publicly funded home care and which turned counting into GPS tracking of carers and of the disabled people they care for, with unpaid claims from failed check-ins and geofencing that penalizes leaving the house (the Cures Act verification record; Mateescu). The firewall follows. Care is counted only at a level of aggregation from which no individual's schedule can be reconstructed.

Redistribution moved where defaults moved and faded where they were gestures. Five reserved, well-paid, non-transferable father weeks raised participation by 250 percent and measurably reduced sex specialization years later; a four-week quota shows household effects fifteen years on; a single lightly-held month moved uptake and nothing else long-run; and even a statutory equal split does not yet equalize practice, since the transferable weeks flow one way (Patnaik; Kotsadam and Finseraas; Ekberg and colleagues). The framework's care entitlements therefore follow the thresholds the record supports, reserved, non-transferable, adequately paid, framed as the carer's own, and their effect is read in the diaries, not the statute book. The uptake ceiling on record is Iceland's equal, non-transferable, well-paid six-plus-six-plus-six months, where fathers now take roughly nine in ten, and even there the mother stays the primary carer in the first year, so recognition and even a strong quota move the doing without equalizing it. What might move it further is standing. When care earns a durable claim on the systems it sustains, its incentive stops being recognition alone. Maintenance binds the same way. Statutory 30-year infrastructure strategies raised audited renewals to 85 percent of forecast depreciation and still under-reinvest; states with mandated asset plans added a hundred thousand new lane-miles while their repair backlogs grew; the flagship paydown fund expired with the backlog rising (the New Zealand audit record; Repair Priorities). Ledgers change the conversation; only gates change the allocation. So the ledger carries a gate. Above a threshold, new build is funded only where the renewal ratio meets its floor, repair before expansion, written into the money rather than beside it.

TWO LEDGERS · A FIREWALL · A GATEthe market ledgerGDP, as beforethe care ledgerextended account, scheduled,a stated lower boundnever mergedneither absorbs the otherthe aggregation firewallsampled anonymous diaryaggregates only; no schedule reconstructableper-visit verification logGPS on the carer: the named anti-patternTHE FIX-FIRST GATEmaintenance debtpublished, 30-year horizonrenewals meet the floor?renewal ratio against depreciationnew build fundedrepair funded firstledgers change the conversation; gates change the allocation
The organ's instruments. Two ledgers that never merge make the substrate visible without pricing it into markets; the aggregation firewall admits the sampled anonymous diary and names per-visit tracking as the anti-pattern; and the fix-first gate funds repair before expansion, because the record shows a published backlog binds nothing by itself.
The satellite rule with a clockEvery federation layer keeps the care-and-maintenance account beside the market account, on the extended-account method, published on a mandatory schedule, because the optional ledger demonstrably starves (SNA 2025; the satellite-account record).
The aggregation firewallCare is counted only by anonymous, sampled time-use instruments, at a level where no individual's schedule can be reconstructed; per-transaction verification of care is prohibited as a counting instrument, the named anti-pattern (Mateescu; the Cures Act record).
Redistribution by defaultCare entitlements are reserved, non-transferable, adequately paid, and framed as the carer's own, the quota thresholds the record supports; the target variable is who does the care, read from the diaries (Patnaik; Kotsadam and Finseraas).
Care earns standingSustained care converts into durable, non-transferable, decaying standing in the engine of EQ–05, so it earns a claim on the systems it sustains and not only recognition; the real precedent is the pension credit for years of child-rearing and unpaid elder care (Germany's Kindererziehungszeiten). Crediting care to a person is what the aggregation firewall forbids, so this runs on an opt-in ledger the claimant opens, kept wholly separate from the anonymous diaries and never derived from them.
The fix-first gateAbove a threshold, new build is funded only where the renewal ratio meets its floor, the one binding design with a measured record, conditions on the money rather than exhortations beside it (the New Zealand statute; the asset-management record).
Open questions
  • Measuring care risks commodifying or surveilling it; the aggregation firewall bounds that risk without abolishing it, since a substrate made fully legible is a substrate made administrable, and the firewall only limits the administrative gaze.
  • Recognition without redistribution entrenches the gendered burden. The accounts do not move who does the care, only the defaults do, and even the best-designed defaults have not equalized practice anywhere yet, since Iceland's near-universal father uptake still leaves the mother primary in the first year. Making care earn standing changes the incentive to do it and not only to count it, but it runs into this organ's own firewall, so it is carried on a claimant-opened ledger kept apart from the anonymous diaries, a workaround and not a clean resolution.
  • Valuation at replacement wages embeds the very devaluation it documents, so the ledger is carried as a stated lower bound in its own caption rather than presented as a neutral number.
  • The fix-first gate has never held at national scale against expansion politics; binding it to the money follows the only design with a measured record, and the risk that the gate becomes a formality is named rather than solved.
  • Direction
    • Name care and maintenance as a load-bearing substrate beside the economy of enough, made visible by two ledgers that never merge, counted through the aggregation firewall, redistributed by default, and defended by the fix-first gate (Fraser; ILO; the SNA 2025 revision; Russell and Vinsel; the Foundational Economy).
EQ–08d

Biological Resilience and the Health Commons

Draft

Health is the original vital sign, and a civilization has one too.

The framework's vital signs are mostly metaphorical; this one is literal. Human health is not a private commodity but the visible band of a shared system, in which one organism's immunity, one ecosystem's intactness, and one society's baseline conditions together determine whether populations stay in survivable ranges. The official One Health framing treats human, animal, and environmental health as one interdependent system, which is how the framework already treats its other signs.

Three findings make this load-bearing. Prevention is radically cheaper than collapse, since the cost of a pandemic runs to the trillions while the cost of reducing spillover runs to tens of billions a year. Antimicrobial resistance is a slow-moving systemic risk, with on the order of 1.27 million deaths attributable to bacterial resistance in 2019, reported here with its uncertainty rather than as a single dramatic forecast. And the determinants of health are largely social, since the Whitehall studies found a steep mortality gradient across the social hierarchy that classic risk factors only partly explain.

The core goods are commons. Herd immunity is a positive externality prone to free-riding, and a shared microbial and biodiversity base is non-excludable, so prevention's benefit collapses when protection is privatized or enclosed. Health read as a central capability grounds a biological reading of the dignity floor, bodily health and integrity as minimums beneath which no person may fall, whatever their rank.

humananimalenvironmentOne Health, one interdependent systemcost asymmetryprevent~$20-30B/yrrespond~$14T
One Health, one interdependent system of human, animal, and environmental health, and the cost asymmetry that makes prevention the rational choice. A pandemic costs trillions, reducing spillover costs tens of billions a year.
A literal biological vital signA reading composed of preparedness, the resistance burden reported with intervals, and the health gradient rather than one number, now carried on the panel as its eighth sign (One Health; the AMR burden estimates; Marmot).
Prevention as a commons goodFund spillover prevention and surveillance collectively, since the cost asymmetry is destroyed when protection is enclosed (Dobson; Bernstein).
The biological immune systemOne Health surveillance, vaccination, and resistance stewardship, the literal counterpart of the cognitive immune system of EQ–03c.
A biological dignity floorBodily health and integrity as inviolable minimums regardless of social grade (Sen; Nussbaum).
Open questions
  • Robust biosurveillance builds the same infrastructure that can be turned to control, so the surveillance-versus-liberty tension is real.
  • Any single index is a chosen proxy that can be gamed or mis-specified, as the contested ten-million-deaths-by-2050 AMR forecast showed.
  • Treating a civilization as an organism can slide into technocratic overreach and pathologizing dissent as infection, which the framing must guard against.
  • Direction
    • Add biological resilience as a literal vital sign and a health commons, with One Health surveillance as the biological immune system and a biological reading of the dignity floor (One Health; the AMR burden estimates; Marmot; Sen and Nussbaum). The ecological side of the same One Health system is governed in EQ–08e.
EQ–08e

The More-Than-Human Stake

Draft

A constraint has no voice. The river needs not a higher number but an agent who can hold a breach, and who can be removed for holding it wrongly.

The framework already widens the circle of standing past present persons. EQ–06c seats future persons through guardianship and grants nature standing too, but only in a sentence, held in trust rather than as a claim of equivalence. Nature otherwise enters the framework twice more and thinly, as the ecological ceiling of the economics of enough and as the ecological-throughput vital sign. Every other protected interest is given an organ, an agent, and a decaying-standing account of who speaks for it. The more-than-human is given a constraint and a reading. The asymmetry is unprincipled, and this organ closes it.

The framework has a sharper reason to build it than the field usually has. The central unsolved problem of rights-of-nature law, who may speak for nature without becoming its unaccountable sovereign, is P–05 and OQ–00 in another body. Aeviterna has a worked, if leaky, answer to that problem already in the guardian of the core, so bringing the more-than-human inside lets the framework point its strongest machinery at a question the rest of the field has mostly left open, and lets the case stress-test that machinery in return.

The record of the last decade says why the design has to be built this way. Legal personhood for an ecosystem has proven real but narrow, strong as a shield and weak as a program. Ecuador's Los Cedros ruling stopped the mining inside a cloud forest, and a 2024 field study found the extraction halted with no infrastructure left behind; Panama's highest court voided a major copper mine on a rights-of-nature ground in 2023; and in 2024 Spain's Constitutional Court upheld legal personhood for the Mar Menor lagoon, the first such ruling by an EU member state's highest court. Where the task was restoration, funding, and enforcement against diffuse harm, the same instrument mostly failed. Colombia's Atrato River, a subject of rights since 2016, remained by its own ombudsman's account a plan on paper, its guardians unpaid and its illegal mining spreading. The lesson is not that standing is empty but that recognition without a funded, enforcing, capture-resistant agent behind it is, which is why the mechanisms below aim at enforcement and capture rather than at the declaration.

The unit of representation is a decaying ecological standing held by a guardian for a specified natural system, a watershed, a forest, a species assemblage, never nature in general, because generality is the failure mode that builds the throne by the shortest route. The guardian is constituted as a reading seat (EQ–06e) pointed at its ecosystem, with the seat's constitution inherited rather than rebuilt: plural, part-lot with the lay majority able to outvote the experts, tenured on a decaying mandate renewed only from outside, never funded by what it reads, publishing its raw readings and dissents. Its particulars are its own. The local custodians whose knowledge of the system is deepest hold the expert chairs, it speaks for one named system and never nature in general, and a standing shadow guardian waits so the system never depends on this panel surviving.

Contestability is the heart of the design and the answer to the field's sharpest finding, that whoever defines nature's interest holds the real power. Any cell, or a quorum of citizens, may submit a competing reading of a system's interest and force a contest, and if the alternative prevails the standing of those who produced the original reading decays, so being wrong about a river carries a cost. Capture bounties reward anyone who shows a guardian's reading was bent toward a developer, an extractor, or the guardian's own institution. The Te Urewera case is the live proof, since a guardian acting against part of what it guarded was checked when a single member of the community that held the seat forced the question, and it bit because the guardian's process was logged and reviewable.

The organ then adds a more-than-human floor that mirrors the dignity floor without claiming the two are equivalent: a small, lit, non-negotiable set of ecological minimums beneath which no decision may push a system regardless of efficiency or majority, that no decision may knowingly drive a species to extinction, collapse an ecosystem past a recognized tipping point, or breach a hard planetary boundary. It is kept narrow for the same reason the human core is, so it stays legible and a poor capture target, and it is bounded to a system's continuation and self-renewal rather than the welfare of every organism, since nature runs on predation. With an agent in place, the throughput vital sign stops being an unowned number and becomes the instrument this guardian reads and the breach it can hold, which answers EQ–02c's open question by enforcing the ceiling through a plural, contestable guardian rather than a central rationing authority. The organ is the ecological reading of the same One Health system EQ–08d governs from the human side, since intact ecosystems are upstream pandemic prevention, and it writes the nature half of EQ–06c's standing substrate to the depth the future-persons half already has.

A sovereign for natureA guardian per systemone seat over all of naturepermanent, generalforce a re-readingby lotstewardthe floor: no extinction, collapse, or boundary breachplural, decaying, contestable, scoped
Two ways to give nature a voice. On the left a single guardian sits over all of nature, permanent and general, the sovereign P–05 forbids in another body. On the right the guardian is a plural, decaying, contestable panel encircling one named system, its members drawn by lot (filled) and leavened with a minority of domain stewards (outlined); anyone can force a re-reading of the system's interest, and a small floor sits beneath it that no decision may cross. The panel is an instantiation of the reading seat (EQ–06e); the shape drawn here is the seat's, pointed at a river.
Standing per system, never for nature entireA decaying ecological standing held for a named watershed or forest, never a single seat over the living world (P–05; Berry's role-specific rights).
A guardian built to fail the sovereign's testsPlural, sortition-leavened, decaying, non-general, externally renewed, with a standing shadow, the guardian-of-the-core discipline applied to nature (Whanganui's Te Pou Tupua widened past the bilateral form).
Contestable readings, with a cost for being wrongAny citizen can force a re-reading of a system's interest, a losing original reading decays its authors' standing, and bounties reward exposing a bent reading (the Te Urewera check).
A more-than-human floor, small and litNo knowing extinction, ecosystem collapse, or planetary-boundary breach, mirroring the dignity floor without claiming equivalence, bounded to a system's self-renewal rather than every organism.
The throughput sign, given an agentThe ecological vital sign becomes the instrument the guardian reads and the breach it can hold, enforcing the ceiling without a central rationer (EQ–02c).
One organism, two readingsThe ecological counterpart to EQ–08d's human-side One Health, completing the nature half of EQ–06c's standing substrate.
An open-standing backstopAny person may go to law on the system's behalf, so a captured or passive guardian cannot foreclose enforcement; the proven wins came from open standing, not from personhood alone (Ecuador; the Mar Menor law).
Funded, and walled off from the polluterA secured endowment so monitoring and litigation never collapse onto volunteers (Whanganui's river fund), with the regulated or extracting authority barred from holding or removing the seat, the Atrato failure mode named and designed against.
Residuals and a falsifier
  • Anthropocentric capture is reduced, not removed, since every guardian is human and reads nature through human values; the design makes the reading plural and contestable and gives being wrong a cost, but it does not make the reading neutral, because no neutral reading exists.
  • Who defines nature's interest stays unsolved in principle, since the contest decides whose reading prevails, not the true interest of a river, which underdetermines the answer, because protecting the system and protecting the individual organism yield opposite mandates.
  • Romanticism is a standing temptation, since the balance of nature is scientifically obsolete and a guardian that invokes it to veto change is romanticism acting as a sovereign; the narrow floor and the plural indices contain it but do not end it.
  • Scale is unresolved twice over, since per-system guardians make coordination across systems hard, which is EQ–04b's scale problem in another body, and the half-lives are calibrated only as bands (EQ–05d), unvalidated in deployment, which is OQ–02 wearing a new hat.
  • The test has to count the divergence cases, the decisions where the guardian's ecological reading actually cut against the human majority, and then ask how often the guardian's position prevailed and whether a pre-registered ecological indicator improved against a comparable un-guarded system; a guardian that only ever wins the decisions already conceded has changed nothing.
Falsifier

Decaying, contestable guardians should outperform both the diffuse everyone-is-a-guardian model, which under-acts, and the entrenched single-community model, which can act against the system unchecked, on the one metric that matters, whether ceiling-breaching decisions are actually held; if they do no better over time, the organ's core claim is false. A popular capture also clears the contest, since where a community is unanimous against a system's interest no one brings the challenge, which is OQ–04 reappearing at the nature-framing layer. Carried as OQ–10.

EQ–08f

The Age Pyramid

Draft

A polity that asks for hands and hours has to count the hands it will have.

The care and maintenance substrate (EQ–08c) makes the unpaid work of a society visible and redistributes it, and the rest of the architecture asks members for hours, the sortition seats, the quarterly reviews, the governance overhead of EQ–02f. Both draw on one account, the discretionary hours of working-age adults, and both have assumed that account is full. In an aging, sub-replacement society it is not. The working-age cohort, the numerator, shrinks in the same decades when the population it must serve, the old who need care, grows. Global fertility is already below replacement across most of the world, the number of working-age adults per person over sixty-five is falling from about four toward one and a half in the leading cases, and the eighty-and-over cohort, the heaviest users of hands-on care, is the fastest-growing of all (United Nations, World Population Prospects 2024; OECD, Pensions at a Glance 2025).

The consequence is an arithmetic the substrate has to meet. As the pyramid inverts, care demand rises while the hours to meet it fall, and the participation budget lands on the same shrinking cohort, heaviest on the carers, most of them women, it means to protect. Migration and automation soften the blow without repealing it. The migration that would hold the age structure is an order of magnitude beyond any real flow, and the sending countries are aging too (United Nations, Replacement Migration). Two decades of subsidized care robots in Japan neither cut the labor nor stayed in use, the devices as often shelved as adopted (Wright, Robots Won't Save Japan). The counter-case, that healthy aging and longer working lives blunt the alarm, weakens the fiscal panic but not this one, because the oldest-old care load is the least automatable and cannot be asked to work longer.

THE DEMOGRAPHIC LOAD-BEARING TEST young: broad base wide population ages aged: top-heavy narrow the green slice, the working-age cohort, carries both care and governance, and it is shrinking care hours governance hours unfunded overflow a member's discretionary time, held flat Young society support ratio ~4 : room to spare over budget (F–15) Aged, budget flat support ratio ~1.3 : care crowds out governance Aged + circuit-breaker governance throttled : fits again the breaker cuts governance, not care
The demographic load-bearing test. A young population is broad-based and its working-age slice wide; an aging one is top-heavy and the slice narrow, and that slice carries both care and governance. In a young society the two fit inside a member's discretionary time; held flat as the society ages, care crowds governance out until the two overflow the ceiling, the F–15 failure; the demographic circuit-breaker throttles governance, never care, so the total fits again.

So the organ adds the term the substrate lacks, a scarcity price on care and a governor on participation, and it keeps the one promise it cannot design away. Care is the demand that cannot be cut; governance is the buffer that can.

Care priced to scarcityThe standing and the pay a carer earns per hour are indexed to the local support ratio, so the claim earned by caring rises as carers grow scarce, a counter-cyclical incentive that strengthens exactly when the pyramid inverts. It generalizes the pension credit for years of child-rearing the substrate already cites.
The quota, kept and relabeledThe reserved, non-transferable, adequately paid carer entitlement stays as a redistribution and time-equity mechanism, on the thresholds the record supports. It is not a fertility lever, and the organ says so, because the Nordic fertility decline and the paternity-quota studies forbid that claim.
Migration as a bounded leverManaged migration enters with a stated ceiling and a guardrail against draining carers from sending polities that are aging too. It buys time and is never booked as a solution.
Participation on functional ageEligibility is set on capacity and remaining life expectancy rather than a fixed retirement age, enlisting the young-old and designing sortition and reviews for phased, low-burden, remotely reachable contribution.
A demographic circuit-breakerThe governance-overhead budget is a function of the local support ratio. Below a set threshold the mandatory participation load throttles down, reviews lengthen, panels shrink, and delegation becomes the default, so governance never bids against care for a carer's last free hour.
Open questions
  • The oldest-old care load rises regardless, since it is the least automatable and cannot be asked to work longer, so the organ can price and govern the shortfall but not abolish it.
  • Migration only defers and displaces the shortfall, and the gender-equal quota's fertility payoff is uncertain at best, so no polity designs its way back to a stable pyramid within a generation.
  • A deeply aged polity, with a support ratio near one to one, faces a real trade-off the organ makes explicit rather than solves. It must relax the participation load, import carers, or accept unmet care.
Falsifier F–15

The care-and-participation arithmetic is false if, in any member polity, the old-age support ratio falls below about one and a half working-age adults per person over sixty-five while the framework still owes both its full carer entitlements and its full participation budget, and the measured sum of required informal care hours and mandated governance hours per working-age member exceeds that member's measured discretionary time. The four quantities are already measured, the support ratio, care-need hours by age, the informal-care share, and time use, and the projections date it. Japan and Italy reach the threshold in the late 2030s and 2040s, and Korea by about 2050. It is falsifiable in both directions, since a polity that sustains both from residents past the threshold without importing carers or throttling governance leaves the assumption standing. Carried as F–15.

Part VGovernance and Endurance
EQ–09

Governance Modes and Crisis Adaptation

Stable

Aeviterna recognizes that systems behave differently during crisis, scarcity, abundance, and recovery phases. Governance structures appropriate for one phase may be actively harmful in another. A framework that does not account for this will either be too rigid to handle crisis or too permissive to prevent crisis from becoming permanent.

Temporary centralization may emerge during emergencies. This is not a failure of the framework. It is a recognized and expected mode. The requirement is that emergency authority carry built-in decay. It must unwind, crisis governance must dissolve, and recovery protocols must actively redistribute coordination back to distributed structures. Without decay mechanisms, crises permanently fossilize hierarchy. The emergency becomes the new normal.

The four modes are not different governments but different settings of the same one, distinguished by how much temporary centralization the situation licenses and how fast that centralization must unwind. Abundance runs on distributed coordination, scarcity tightens prioritization without concentrating control, crisis may license real but temporary centralization, and recovery exists to spend that centralization back down. The mode is set by measurable conditions rather than by the discretion of whoever would benefit from the switch, and every step toward concentration is paired with a scheduled step back.

Abundancecentralization: lowScarcitycentralization: moderateCrisiscentralization: high, temporaryRecoverycentralization: fallingmargins fallshock exceeds bufferspowers expire on exitcoordination handed backsudden shock
The four governance modes. Centralization may rise toward crisis but expires by default on the way out, so recovery hands coordination back rather than letting the emergency become the new normal.
AbundanceEntered when resource margins and trust sit comfortably inside their bands. Coordination is fully distributed and centralization is low. This is the default resting mode.
ScarcityEntered when margins fall below a set threshold. Prioritization tightens and allocation narrows while control stays distributed. It exits upward when margins recover, or downward into crisis if a shock exceeds buffered capacity.
CrisisEntered when a shock exceeds what current buffers can absorb. Temporary centralization is licensed for speed, granted with an expiry attached and the shortest standing half-life in the system. The mode cannot renew itself.
RecoveryEntered when the acute threat subsides. Its defining task is to redistribute coordination outward and let emergency authority expire on schedule, returning the system to abundance or scarcity. A crisis that never passes into recovery is a captured crisis.

Two dangers attend the crisis mode: that emergency authority never expires, and that someone manufactures a crisis to trigger it. The framework answers both structurally. Emergency authority carries an automatic expiry, and extension requires an escalating threshold met by a rotated body rather than passive continuation. The actor who declares an emergency is held apart from the actor who receives the expanded authority, with the powers held in escrow between them, so declaring a crisis grants no one new control. The nested units of the wider structure hold a veto over a declared emergency, and any standing gained during crisis decays faster than ordinary standing, so the ratchet reverses on its own. Recovery then runs an explicit handoff protocol that returns each centralized function to distributed stewardship and logs the return in civic memory.

How the thresholds get set is itself a design choice with precedent. The cleanest analogues are quantitative and automatic. Market circuit breakers trip at fixed, publicly known levels measured against the prior close, which removes discretion but invites a magnet effect as behavior rushes the line, a caution against naive variance triggers. Comparative constitutional practice points the same way as the standing rules here, a formal declaration and a bounded scope, with prior approval required to prolong, and each measure meeting necessity and proportionality on its own (the Venice Commission benchmarks; ICCPR Article 4). The strongest reform template flips the default so emergency powers expire unless a rotated body affirmatively renews them against an escalating burden, the reverse of renewal by inertia.

Open questions
  • The entry and exit thresholds are defined in form, and the realistic templates are quantitative triggers read against historical variance or buffers; calibrating them per polity, without inviting a rush at the line, stays open.
  • Procedural checks usually bind, holding in roughly four of five emergencies studied through the pandemic (Ginsburg and Versteeg), yet a broadly popular or manufactured crisis can clear all of them, which structure alone cannot prevent (see OQ–04).
  • The handoff protocol assumes the distributed structures it returns power to survived the emergency, which a long crisis may erode.
  • Direction
    • Separate the declarer from the beneficiary, default-expire emergency powers with an asymmetric re-authorization burden on a rotated body, allow nested-unit veto, and decay emergency standing fastest (Venice Commission; ICCPR Article 4; Ferejohn and Pasquino; Higgs). See OQ–04.
  • Implementation sketch · exploratory
    • Emergency escrow, with emergency power held by a separate rotating body, default-expiring, vetoable by nested units, and rehearsed through routine crisis drills (regulatory sandboxes; the PATRIOT Act sunsets as the cautionary case).
EQ–09b

Temporal Sovereignty

Draft

A civilization can hold every reading inside its band and still be moving too fast.

The framework governs how much a civilization extracts and how it perceives, but not how fast, and tempo is itself extractable. A vital sign inside its band but changing too fast is still a resilience failure, because the slow parts of a system cannot absorb change delivered at the speed of the fast parts. Social acceleration, in Rosa's account, is a self-reinforcing loop that shrinks the present, and the answer is a governable tempo rather than mere slowness.

The resilience literature supplies the rigor. Slow, controlling variables govern how a system absorbs shocks, and pushing fast variables hard while slow variables drift toward thresholds is how a system is tipped into a worse regime, a scale mismatch between the pace of management and the pace of what it manages. Brand's pace layering turns this into a design principle, a civilization layered by tempo from fast commerce to slow culture and nature, endangered when a fast layer drives a slow one at its own speed.

The speed limit already exists in miniature. After the Flash Crash of 2010, markets adopted circuit breakers that halt trading on sharp moves and pause a stock that runs outside a band, externally triggered and sunset-bounded, the financial twin of the crisis mode of EQ–09. The labor analogue is the right to disconnect. Bounding tempo is an ordinary instrument, not a utopian one.

fashion and artcommerceinfrastructuregovernanceculturenature and the dignity floorfast layers learn, wide rate-bandslow layers remember, tight rate-band
Pace layers, from fast to slow. Fast layers like commerce learn and carry wide rate-bands; slow layers like culture, nature, and the dignity floor remember and are held to tight ones. A civilization fails when a fast layer drives a slow one at its own speed.
Band the rate, not only the levelA sign read for acceleration as well as position, since change too fast is a failure even inside the band (Walker and colleagues).
Tempo by layerWide rate-bands for fast exploratory layers, tight ones for the slow constitutional and ecological core (Brand).
A tempo circuit breakerA bounded, reversible slowdown when a domain's rate of change breaches its band, generalizing the market circuit breaker and the crisis mode of EQ–09.
Designed against its own stampedeThe brake is laddered rather than a single cliff, because a hard limit can pull behavior toward it, the magnet effect (Subrahmanyam).
Open questions
  • Speed has real benefits, in emergency response and coordination, so the aim is temporal autonomy rather than slowness for its own sake.
  • The power to set tempo is a capture target, since a right to slow others is also a right to impose one group's pace.
  • Enforceability is weak, and resonance, the responsive relationship that is the actual good, cannot be legislated (Rosa).
  • Direction
    • Hold the pace of economic, informational, and ecological processes inside survivable bands, with a tempo circuit breaker generalized from the market kind and tightest on the slow layers (Rosa; Walker; Brand; the market circuit breakers).
EQ–09c

The Manufactured Crisis

Stable

Anyone may start the clock. Only the unrewarded may wind it.

OQ–04 is the failure mode the rest of the framework keeps deferring to, an emergency declared or sustained pretextually, with the approval of the very public that would need to object, because the framing layer, not the ballot, is where the capture happened. The guardian of the core concedes that a popular capture clears every contest; the more-than-human organ names the same hole; EQ–09 separates declarer from beneficiary in principle. This organ builds the machinery, from an empirical record that is now good enough to build with and honest enough to discipline the design.

The record cuts both ways, and both edges are load-bearing. Emergencies erode. Nine of ten constitutions carry emergency provisions, at least 137 countries declared one across three decades, 115 of 853 coded declarations were unlawful under the declaring state's own constitution, and democracies under a state of emergency were 75 percent more likely to erode than those without (Bjørnskov and Voigt; Lührmann and Rooney). The paper safeguards do not predict who abuses them, and requiring second-chamber approval correlates with more unconstitutional declarations, not fewer, so stacking approval chambers is not the answer. Yet the modal crisis response is bound, not Schmittian. Across a hundred-plus pandemic responses, courts and legislatures constrained executives far more than the theory of the exception predicts, two-thirds of democracies violated no standard, and violating standards bought no measurable health benefit (Ginsburg and Versteeg; Edgell and colleagues). The organ targets a tail risk and must not tax the modal case to death. The deepest drain runs the other way. Counterfactual modeling attributes roughly half the first American wave's reported deaths to distancing begun one to two weeks too late (Pei and colleagues), publics traded liberties in proportion to actual risk, and safeguards themselves have killed, a state court voiding every pandemic order at once mid-crisis, a parliament refusing extension at the infection peak. Any organ that can force lapse during a real emergency has a body count in expectation, and this one owns it.

Two further findings shape the machinery. The real ratchet is transcription, not extension. France ended its emergency the day a statute moved its powers into ordinary law, Egypt lifted its emergency days before amendments made the provisions permanent, and the sunset clause, the standard remedy, is near-worthless on the record, renewals pro forma for decades, expiry extremely rare, and sunsets buying support asymmetrically in experiments (Finn; Underhill and Ayres). And the framing layer is cheap, industrialized, and mostly domestic. One state fabricates roughly 448 million posts a year aimed at distraction rather than argument (King, Pan, and Roberts), two-thirds of the covert influence networks a single platform removed targeted their own domestic publics (Meta), and personalized model-driven persuasion now beats humans in controlled debate (Salvi and colleagues), while the measured effect of the canonical deepfake cases stays small to undetectable, so the panic is also a weapon.

The organ is five rules, each answering a named failure.

  • The declarer never renews. Any qualified body may ignite crisis mode instantly, because delay is the one quantified killer. Renewal belongs to a body that gains nothing from continuation, no budget, jurisdiction, personnel, or tenure, and renewal thresholds escalate with each cycle, the supermajoritarian escalator (Ackerman), so only a crisis that persists in fact can hold the coalition.
  • Automatic, procedural-first review. Every declaration and renewal is docketed before the review organ with no one needing standing to bring it. The first questions are the ones courts have actually won with: right instrument, right legal basis, right gate. No one out-argues the executive's epidemiology on day three.
  • Triggers are observable, never popular. Crisis mode keys to pre-registered, externally attested indicators, mortality, attributed attack, infrastructure failure, read by bodies outside the declarer's appointment chain. Approval polls, plebiscites, and trending sentiment are inadmissible as triggers or renewal evidence, because the framing layer is the cheapest thing to buy. This is the one place the framework deliberately refuses the demos, and says so aloud.
  • The transcription ban. No emergency-born power enters ordinary law during the emergency or its cooling-off period; permanence requires de novo justification and an independent audit of the power's emergency record. This targets the ratchet the record actually shows.
  • Speed for ignition, friction for duration, cure for invalidation. Reviewers who strike a measure issue cure windows and substitution duties, never instant vacuums, because the cliff-edge invalidation mid-crisis is also a documented failure mode.

Two of these rules carry the answer to the section's own recursion, and a live case shows the failure and the fix at once. In 2025 a mayor declared the passage of a duly enacted law an emergency and renewed the declaration every thirty days, until a court struck the orders down as exceeding the power, the manufactured crisis and the self-renewal in a single exhibit (the New York solitary-confinement orders). The fix hardens two rules. The observable trigger is affirmed not by the declarer but by a reading seat the declarer can neither appoint nor defund (EQ–06e), which puts the attester out of reach of both capture levers at once. And renewal runs on a presumption exchange, where the same votes that would extend a power under an affirmative default instead let it lapse under a sunset default, so silence ends the emergency and continuation must be re-earned against the declarer's burden (Lauenroth), the discipline the record demands, since nine in ten democratic breakdowns run through emergency rule.

Ignitioninstant, reviewablerenew 60%then 70%then 80%the ladder is held by a body that gains nothing from continuationLapsethe default, with curewindows, never cliffsObservable triggers onlypre-registered indicators; approval polls inadmissibleThe transcription banno emergency power enters ordinary law; cooling-off, then de novothe ratchet the record shows is not extension but transcription:emergency powers walking into ordinary law while the emergency formally ends
The manufactured-crisis organ. Ignition is instant and reviewable; renewal climbs an escalating ladder held by the unrewarded; triggers are observable and popular approval is inadmissible; and the transcription ban blocks the door the record shows powers actually walk through.
Falsifier

Across polities running this organ, pretextual extensions and transcriptions should fall measurably while violation indices during real crises do not rise, and no excess mortality attributable to blocked ignition should appear. A polity that shows the organ's frictions without its filtering has bought paperwork with lives, and the organ is false.

Open questions
  • The parchment problem. The worst abusers passed through formally elaborate safeguards, so where institutions are hollow this organ is decoration; its real precondition is the wider anti-capture layer.
  • The friction problem. The escalator can force lapse during a real emergency, and the design accepts a nonzero cost in delayed response, stated rather than laundered.
  • The recursion. An organ empowered to discount public approval as captured claims epistemic authority over the demos that is itself unaudited. The answer is structural humility, observable triggers rather than judgments of the public mind, and attesters constituted as a reading seat (EQ–06e) the declarer can neither appoint nor defund, plural, decaying, never rewarded by continuation, which is that organ's funding firewall under this one's older name. Two of the three legs are now pinned. The attester sits outside the declarer's reach, and the sunset-by-default puts the burden of continuation on the declarer. What remains is the frame. A capture that owns the indicators and the frame at once can still feed a clean trigger to an honest seat, and external attestation is weak exactly where it is ignored, since barely a fifth of states even notified their emergency derogations in the last pandemic. OQ–04 is narrowed, not closed.
EQ–10

Long-Term Orientation

Stable

Aeviterna is fundamentally long-horizon. Not because long-term thinking is virtuous in the abstract, but because the failure modes Aeviterna is designed to address are slow. Extraction is profitable quarter by quarter and catastrophic decade by decade. Cognitive capture accumulates imperceptibly until correction becomes unthinkable. Stewardship that calcifies into sovereignty does so gradually, through hundreds of decisions that each seem reasonable in isolation.

A framework that cannot hold long-horizon thinking institutionally, not just rhetorically, cannot address these failure modes. The success of a civilization is therefore measured not only through output, but through resilience, continuity, recoverability, distributed stewardship, cognitive sovereignty, adaptive capacity, and ecological survivability. These are not soft measures. They are the measures that determine whether the civilization is still operating in one hundred years.

The only thing a civilization should promise forever is that it will always be able to change everything else.

A long horizon forces a harder question the framework had left implicit. Aeviterna prizes corrigibility, the capacity to detect error and change course, and treats it as a primary design requirement (FC–01; EQ–11). But a civilization also has to bind itself. Constitutions, protected commons, inviolable rights, and obligations to people not yet born are all commitments meant to hold precisely when a present majority would prefer to break them, and all of them are anti-corrigible by design. A framework built only for changeability has no answer to the moment when the right thing is to make something unchangeable, and a framework built only for commitment becomes the dead hand of the past ruling the living. Aeviterna needs a principled line between the two.

The line has a literature. Jon Elster analyzed precommitment, the act of binding one's own future choices, in the image of Ulysses ordering his crew to tie him to the mast so he could hear the Sirens without steering onto the rocks. He is careful that the analogy only goes so far, since constitution-makers often bind others rather than only themselves, but the constructive core holds. A constitution can constrain a democracy in order to enable it, and the binding is what makes the freedom durable. Stephen Holmes put the same point as a paradox, that constraints can be enabling, the way the rules of a practice make the practice possible rather than merely limiting it.

Aeviterna's resolution follows directly. Lock the meta-rules and keep the object level liquid. The things a civilization binds permanently are few and structural. They are the right to revise itself, the core protections that revision must never strip, which are the cognitive sovereignty, the knowledge commons, and the dignity floor of EQ–06c, and the open procedures by which binding and unbinding happen. Everything below that, the actual policies and institutions, stays fully revisable. The one thing made permanent is the capacity to change everything else. This is self-binding in the service of staying unbound, and it is the only form of precommitment compatible with a corrigible civilization, because it entrenches the meta-rule of corrigibility itself while refusing to freeze any particular answer.

Object level, liquidpolicies and institutions, revisable any timeMeta-core, lockedthe right to revise,the protections revision cannot strip,the amendment rule itselfcommit to exactly one thing forever: the capacity to change the rest
A small locked core, a liquid shell. The meta-rules at the center, the right to revise and the protections revision cannot strip, are bound permanently. Everything outside churns freely. The civilization commits to one thing forever, its own capacity to change the rest.

This is also how the framework hardens the standing it already grants to people who do not yet exist. EQ–06c derives the dignity of future persons and seats them through guardianship, with Wales' Future Generations Commissioner and the United Nations Declaration on Future Generations of 2024 as the working seeds. The interest of future generations is the textbook case of a commitment the present is structurally tempted to break, since the unborn cannot vote, lobby, or leave, and the costs of betraying them fall due after every current decision-maker has gone. The institutional record shows the failure to design out. Hungary's Ombudsman for Future Generations, under Sándor Fülöp, intervened against a water privatization and acted to protect the national gene bank, then was downgraded into a deputy office within about a term. Israel's Knesset commission was discontinued after a single term. An office the present can quietly gut is not a commitment, it is a courtesy. A future-standing body works only if it is locked at the meta-level, constitutionally entrenched with a mandate that decays and renews on a fixed schedule rather than at the pleasure of whoever it currently constrains, while its specific recommendations stay advisory and revisable. The standing is made permanent while the verdicts stay open to revision.

Lock the meta, keep policy liquidOnly the right to revise, the protections revision cannot strip, and the amendment rule itself are made permanent; every policy and institution below them stays fully revisable (Elster; Holmes).
Entrench the standing, not the verdictA future-generations body is constitutionally entrenched so no ordinary majority can abolish it, while its findings stay advisory, so it holds permanent standing and no permanent power (EQ–06c).
A decaying, renewing mandateThe body's authority runs on a fixed term that renews on schedule rather than at the pleasure of those it constrains, which is the EQ–05 decay logic applied to the office that speaks for the unborn.
The working floor and its failureWales is the floor to build on, and the gutted offices of Hungary and Israel are the failure to design out, which is what entrenchment at the meta-level is for.
Open questions
  • Which protections belong in the locked meta-core, and how that short list is itself revised without becoming just another object-level policy, is unresolved.
  • Entrenching a future-standing body against a determined present is hard, since Hungary's and Israel's offices were gutted within a term, so the decaying-and-renewing mandate has to avoid both capture, when it is too permanent, and irrelevance, when it is too easily revoked.
  • Weighing intergenerational standing against present need without either one dominating the other is open.
  • Direction
    • Draw an explicit line between a small locked meta-core, the right to revise, the core protections, and the amendment rule, and a fully liquid object level, and give future generations a constitutionally entrenched body with a decaying-renewing mandate and advisory-only verdicts (Elster; Holmes; Wales; the UN Declaration on Future Generations).
  • Implementation sketch · exploratory
    • Entrench a future-standing office at the meta-level with a fixed-term, automatically renewing mandate that no ordinary majority can abolish, empowered to publish findings and force a recorded vote but not to decide, with the Wales office as the working floor.
EQ–10b

The Movement's Own Metabolism

Draft

Aeviterna makes a claim about systems, that equilibrium is a kind of death and that a system stays alive by oscillating within bounds rather than settling at an extreme. That claim has to apply to the movement that carries the framework, or the framework refutes itself. Most movements treat institutionalization as the goal, the point at which spontaneous energy finally hardens into permanent structure. This one cannot, because its own thesis forbids settling. The movement is therefore designed to keep moving and to renew itself rather than to arrive.

The arc that movements usually follow is well documented and remarkably uniform. Ferment gives way to charismatic excitement, which gives way to formalization, which ends in a stable bureaucracy (Blumer; Weber). Vitality drains at three predictable points: the succession crisis when a founder departs; the bureaucratization that lets an organization optimize its own survival above its mission (Michels); and the mainstreaming that buys respectability at the price of fervor. For a movement built on ideas rather than on protest or party, the order of dangers is different. The deadliest is dogmatization, the moment core claims are made unfalsifiable and dissent is policed as heresy, because that forfeits the one thing an idea-movement has, which is the ability to be right for discoverable reasons (Popper). Close behind are purity spirals and schism, since ideas divide infinitely and status tends to flow to whoever seems most rigorous, and Goodhart capture, where a movement starts optimizing legible proxies such as followers or citations in place of the truth they were meant to track (Goodhart; Strathern).

Several of these failure modes are already answered elsewhere in the framework, because the properties that make a civilization adaptive are the properties that make a movement durable.

Falsifiability over doctrineThe framework is versioned, marked by stability, and anchored by a standing Research Frontier, so it is built to be wrong in discoverable ways rather than defended as scripture. This is the structural answer to dogmatization. See FM–04 and Appendix B.
Decaying standingAuthority and contribution standing lose weight over time, so influence cannot accumulate into a permanent leadership class. This answers oligarchy and entrenched charisma. See EQ–05.
Polycentric structureThe on-ramp runs from person to cell to institution to federation, and technical mechanisms are specified as open, interoperable standards. Capture has no single seat to take, and exit stays cheap. See Praxis, EQ–09, and P–07.
Anti-capture architectureThe mechanisms that resist a civilization being captured by a faction also resist the movement being co-opted by the system it set out to change. See EQ–06.

How the movement renews itself on purpose is the organ's harder half. The commitments below give the orientation; the instruments after them are specified from the documented record.

Founder-light by designAuthority lives in the editable canon rather than in a person, so there is no prophet whose death turns doctrine into exegesis. The framework is authored as a document that anyone competent can amend through its own process (Weber, on the instability of charismatic authority).
Synthesis over purityStanding accrues to those who steelman and reconcile across disagreement, not to those who bid the position ever purer. This is the countermeasure to purity spirals and to the unaccountable cliques that form when structure is denied rather than made explicit (Freeman).
A right to forkDisagreement deep enough to split should produce a documented, legitimate variant rather than a purge or a civil war. A clear fork protocol turns the largest source of fragmentation in idea-movements into polycentric diversity, the way open standards and common-pool institutions already do (Masnick; Ostrom). Keeping exit available also keeps internal voice honest (Hirschman).
Designed renewalStructure should carry sunset clauses, terms that expire, and scheduled occasions to prune what has accreted, so the movement passes through release and reorganization on purpose rather than waiting for collapse to force it (Holling; Gunderson and Holling; Taleb, on improving by removal).
Truth-tracking metricsThe movement reads its own health from the diversity and quality of contribution and from how well its claims survive scrutiny, not from its size. The civilizational vital signs already hold to this discipline, and it is turned inward here.
A metamodern registerThe public voice holds sincerity and self-aware irony at once, committed enough to matter and ironic enough not to curdle into a cult of true believers (Hoffer, on the psychology of mass movements).

The right to fork now has a protocol, read from the one field that has practiced forking long enough to have re-merges on record. The healthiest fork in that record took a new name, wrote its re-merge condition into its founding statements, kept the commons interoperable, and merged back within a year when the parent conceded governance to a neutral venue; the second re-merge on record kept every contribution legally compatible with the incumbent throughout, and the fork became the project (the io.js and EGCS histories). The pathological pole is documented with equal precision. Splits in which both sides claimed the original identity produced no re-merge mechanics at all, and the measured aftermath, a minority chain suffering repeated majority attacks, the first fork splitting again within fifteen months, prices what an identity war costs. The sociology of movements agrees from its own data. Splits track renewal where the exiting faction builds a differentiated organization in a new niche, and track failure where it claims the whole identity (Gamalielsson and Lundell; Balser; Kretschmer). So a fork is legitimate when it takes its own name, files a divergence document at birth naming the falsifiable disagreement and the evidence that would re-merge it, keeps the commons interoperable so work flows both ways, and strips no one's standing. A split that claims the name, anathematizes the parent, or states no re-merge condition is a schism, and the judgment attaches to conduct, never to persons. Re-merge, when the register warrants it, follows the two-ingredient rule the record shows: a neutral venue, and a governance concession by the parent, pre-committed here so the parent cannot demand surrender.

The falsification register has a working precedent with measured results: registered reports, tests accepted before outcomes are known, now offered by over three hundred journals, showing 44 percent positive results against 96 in the standard literature, the cleanest demonstration anywhere that an unregistered record of claims is a biased register of what was tried (Chambers; Scheel, Schijen, and Lakens; Soderberg and colleagues). The framework's falsifiers, F–01 through F–16, become that register. Every load-bearing claim carries its would-be refuter from the day it ships, and contests between a fork and the parent run as adversarial collaborations with a pre-agreed arbiter, whose measured yield is convergence on the map rather than on minds, which the register promises and no more (Kahneman and Klein). Retirement gets the inverted default. Every structure and role carries a clock, and failing to muster affirmative renewal is itself retirement, preserved read-only in the archive, never acting while retired, the design that keeps apathy from breeding zombies, with the rubber-stamp counter-record priced in by giving every clock a named owner (the Attic process; the Texas sunset record). And renewal is procedural rather than aspirational. On a fixed cadence, a version of the framework cannot ship without retiring at least one mechanism, structure, or claim into the archive, improvement by subtraction on a schedule.

THE FORK PROTOCOLdeep disagreementtoo deep for voicethe forkits own namenot the lineage wholedivergence documentthe claim + the re-merge testshared commonsstanding carried acrossre-mergevariantthe register decides: neutral venue, the parent concedes governance, pre-committedthe schismclaims the namethe identity waranathema, no conditionnothing would re-merge itboth heirs weakenedsplit security, stranded users, splits againthe judgment attaches to conduct, not persons: two re-merges are on record, and both required the parent to concede
The fork protocol against the schism. A legitimate fork takes its own name, files its divergence and re-merge condition at birth, and keeps the commons shared, so it can end in re-merge or in a healthy variant. A schism claims the name and states no condition, and the record prices it. Both heirs weakened, and the split splits again.
The fork protocolLegitimate when it takes its own name, files a divergence document naming the falsifiable disagreement and the re-merge evidence, keeps the commons interoperable, and strips no one's standing; re-merge runs through a neutral venue with the parent's governance concession pre-committed (the io.js and EGCS re-merges).
The falsification registerThe F-series is a standing public register. Every load-bearing claim ships with its would-be refuter, and fork-versus-parent contests run as pre-agreed adversarial collaborations, published whatever the result (registered reports; Kahneman and Klein).
The atticEvery structure and role carries a named-owner clock with the default inverted. Failure to muster affirmative renewal is retirement, read-only, composted rather than deleted, re-entry by the founding process (the Attic; the Texas sunset device).
The removal roundOn a fixed cadence, a version cannot ship without retiring at least one mechanism, structure, or claim into the archive, with the changelog recording what was pruned and why, improvement by subtraction on a schedule.

The same discipline turns one level further inward, onto the workshop that writes the framework. A single maintainer working with software agents selects the targets, drafts the papers, codes the instruments, and reviews the result, which is the concentration the framework warns every institution against, running the workshop itself. The record of comparable work names the danger precisely. Two of three popular open-source projects run on two or fewer load-bearing people (Avelino and colleagues), and the succession vacuum that leaves is an attack surface rather than a neutral fact, as the manufactured pressure campaign that installed the xz backdoor showed by naming an exhausted maintainer's successor as its own vector. The transitions that held did not name an heir. They wrote down a procedure and an open license, so anyone competent could continue the work by fork (the Python steering-council transition; the Linux kernel's continuity process; SQLite's public-domain plan to outlive its authors).

So the workshop binds itself to the same instruments the movement runs on, pointed at itself, and the workshop charter that results is one more structure held to the framework's law rather than exempt from it. Succession runs by fork under an explicit license (CC BY-SA for the corpus, MIT for the code), with a convening procedure and a conserved estate named in advance instead of an appointed heir. A degeneration condition is pre-registered, dated, and blind to its own outcome, on the clinical-trial logic that the party running the trial cannot be trusted to decide continue-or-stop after seeing the result (ICH E9). By a fixed date the program must show one outside adoption, one answered outside critic, and a forecast record no worse than chance, or it declares itself degenerating and archives its corpus as parts (F–09; Lakatos on progressive against degenerating programmes). The maintainers keep a scored forecast register, staking dated predictions on the vital signs and grading them by the Brier rule, the reality anchor of EQ–06e turned on the workshop's own framing (Brier; the Open Philanthropy prediction record). Review stays plural to the limit of the budget, run on separate agent instances with programmatic and primary-source checks, and the deeper limit is named rather than buried. One model family still mediates both drafting and review, and its defects propagate to everything adapted from it until a second family or a standing outside reviewer is funded (Bommasani and colleagues; Kleinberg and Raghavan on monoculture).

The same law reaches the corpus itself. If the framework degenerates by F–09 or simply never finds an adopter, it does not persist as a monument, because EQ–05b forbids that of any institution and the workshop that writes it is not exempt. It composts, into the three ends EQ–05b names for anything that dies well. Its parts transplant, the mechanisms that survived their falsifiers traveling into other frameworks under the open license, taken up one at a time by groups that never adopt the whole. Its record archives, the tracker, the priced balance sheet, and the falsifier register standing as evidence a later project can reuse, with the failures coded so they are not repeated. And its argument can still inform policy where no polity runs the stack, the way a defeated proposal shifts the ones that follow. What is preserved is the parts that earned their place, not the name. The only thing held back from sunset is the meta-core that guarantees this ending will happen, never the organization that wrote it. A framework whose whole claim is corrigibility has to be able to die corrigibly, and this names how.

THE WORKSHOP UNDER ITS OWN LAWThe reflexive discipline, turned on the workshopSuccession by forkexplicit license,a procedure, no heirthe kernel conclaveDegeneration conditiondated, blind, externaltests; halts on failureF–09 · July 2028Forecast registerdated stakes, scoredby the Brier rulethe anchor, turned inwardPlural reviewseparate instances,checks against sourcesbounded, not solvedThe named residualone model family drafts and reviews, and its defects propagate downstream until a second family or a standing outside reviewer is funded;the stopping rule is still self-administered, bounded by tests the world can score without the maintainer's help
The workshop under its own law. The four instruments the movement runs on, turned on the workshop that writes the framework: succession by fork under a license, a dated and outcome-blind degeneration condition (F–09), a scored forecast register, and plural review. The residual is named in coral. One model family still drafts and reviews, and the stopping rule is self-administered, bounded only by tests outsiders can score.
Open questions
  • The newer paradigms this section draws on are real but mixed. On-chain governance shows that rule-by-code and token voting fail through voter apathy and plutocratic capture, so none of this should be automated away from human judgment (Buterin; the 2016 DAO exploit and the hard fork that reversed it as the standing caution).
  • Adaptive-cycle and antifragility thinking are heuristics rather than tested designs, and a scheduled release can destroy capacity instead of renewing it if the structures it hands back to have decayed (Holling; Gunderson and Holling).
  • The protocol governs conduct and cannot guarantee outcomes. Across the record the fork lost the name and won the institution, the fork became the name, and the incumbent kept the name and decayed, so a bad-faith actor can perform compliance while running a schism, and identity refuses a clean rule.
  • Re-merge still depends on a parent willing to concede governance, and no protocol can force that grace; adversarial collaboration converges on facts, not minds, and has preceded escalation as well as peace.
  • The instruments are drawn from code and from science, whose artifacts are more separable than identities; at movement scale they are young, and a removal round can prune living tissue if the register misjudges what is load-bearing. The attic and the removal round are not hypothetical; both have fired, retiring a duplicated reader constitution, a superseded anti-gaming clause, and the anecdote page the tracker replaced, so the instruments are no longer untested, only unproven at age.
  • The reflexive turn onto the workshop is real but still self-administered. The maintainer writes the stopping rule, holds the register, and enforces both, with no independent monitor where a trial would seat one; the bound is that F–09's three tests are external and scoreable by strangers, so the world can catch a workshop that fails its own condition even where nothing forces the declaration. The deeper residual is that a framework drafted and reviewed inside one model family reasons about systems like the one that drafts it, and the only full cure, review from outside the family and outside the workshop, is on the roadmap as an outside red team rather than in hand.
  • Direction
    • Apply the framework reflexively, treat the movement and the workshop as two more systems that must oscillate rather than settle, and run renewal through the specified instruments: the fork protocol, the falsification register, the attic with named-owner clocks, the removal round, and the workshop charter that binds the maintainers to succession by fork, a dated degeneration condition, and a scored forecast register.
EQ–11

Closing Orientation

Stable

Aeviterna does not claim to possess a final civilization design. It proposes a directional framework for building systems capable of surviving transition, resisting extraction, preserving autonomy, distributing stewardship, supporting adaptive stability, and preventing concentration from becoming permanent.

Aeviterna accepts imperfection, experimentation, partial adoption, and iterative refinement. It is not a destination. It is a direction and a set of design constraints for moving in that direction without catastrophic loss of the capacity to change course.

Each release has deepened and extended this direction rather than replaced it, from the first complete articulation through new civilizational organs, formal foundations and a place among existing traditions, the harder structures the core had left open, further frontiers, and an honest reckoning with the hardest critiques. The masthead dates each release, the version history carries the public record round by round, and the framework versions itself the way it asks institutions to, with every change owned. It is not a finished design but a coherent whole, stable enough to build on and to argue with, and it is expected to keep changing.

Its goal is not utopia. Its goal is a civilization capable of correcting itself before collapse becomes inevitable.

No equilibrium. No utopia.
A civilization that can sense harm and change course.

From extraction to caretaking.

That is the argument. What follows is the scaffolding beneath it. Part VI gives the formal grounding and the hardest critiques, and the appendices hold the glossary, the open frontier, and the sources, for the reader who would rather test the case than take it.

Part VIRigor and Critique
FC–01

The Formal Backbone

Stable

A claim about how a civilization should move is a claim about dynamics, and dynamics can be made precise.

Before the mathematics, the whole framework fits in one picture. Aeviterna is a feedback regulator. It reads a small set of vital signs, compares them against the bands a civilization can survive within, and corrects through governance before a drift becomes a breach. What makes it Aeviternal rather than generic control is carried in the same loop, where the set point is human dignity, the actuator is the redistribution of power, and the controllers themselves decay so that authority lapses unless it is renewed.

Dignity floor+ survivable bandsset point, never moves+drift / breachGovernancemodes shift with the readingstanding decays unless renewedabundance, scarcity, crisis, recoveryleversThe civilizationmetabolism, institutions,cognition, commonsDisturbanceshocks, capture pressurecivilizational statereadingsThe eight vital signseach held as a bandRead the signs, compare them to the bands, and correct before a drift becomes a breach.
Aeviterna as a feedback regulator. The eight vital signs sense the state of the civilization, the comparison against the survivable bands yields the error the framework calls drift or breach, and governance acts to bring the reading back, shifting between its modes as the error grows. The set point is the dignity floor, the one value that never moves, and the actuator is the redistribution of power (Ashby; Beer).

The framework's central claim is dynamical. A civilization should stay where survivability can be preserved and move freely inside that region rather than hold to a fixed point. That claim has an exact mathematical home in viability theory (Aubin). A system whose state must remain inside a constraint set has a viability kernel, the largest set of states from which it is possible to remain inside indefinitely. The kernel permits unbounded oscillation and requires no equilibrium. The eight vital signs define the constraint set, and the kernel is the region this framework asks a civilization to keep within. Sustainability science already uses the kernel in this exact sense, as the set of states from which a future stays open (Béné, Doyen, and Gabay).

The governing result, with roots in Nagumo and a first rigorous form due to Haddad, gives the test for staying inside the set. At every point of the survivable region, at least one admissible direction of motion must point back inward. The framework's promise to redistribute recovery has a formal counterpart too. A capture basin is the set of states from which a target inside the region can still be reached without leaving it, which is the precise sense in which recovery remains possible after a shock, and the precise sense in which it can become impossible once a system has drifted too far.

Viability theory says where to stay; control theory says how to stay there. Set-invariance control studies the feedback that holds a system inside a safe set under bounded disturbance, which is the formal content of detecting instability and correcting in proportion to it (Blanchini). Regulating around shifting bands rather than holding a single set point is the difference between homeostasis and allostasis, stability achieved through change rather than at a fixed point (Sterling).

Cybernetics supplies the conditions under which any of this can be governed. Ashby's Law of Requisite Variety holds that a regulator can keep a system's essential variables within limits only if it commands at least as much variety as the disturbances it faces. This is the formal warrant for cognitive ecology. A governing system that loses internal diversity loses the capacity to regulate a varied world, which is why monoculture is dangerous and why variety counts against concentrating control (Ashby). Beer's Viable System Model gives the recursive, multi-level architecture that the governance modes inhabit (Beer).

Resilience science supplies the reading for everyday understanding. Holling's ecological resilience is the size of disturbance a system absorbs before it flips into a different basin, which is bounded oscillation within a basin stated plainly, and transformability is a deliberate move into a better basin, which is adaptive transition rather than failure (Holling; Walker, Holling, Carpenter, and Kinzig).

a vital sign another vital sign survivable region (the constraint set) viability kernel bounded oscillation outside the kernel, exit is only a matter of time
The viability kernel is the largest set of states from which a civilization can keep its vital signs inside the survivable region indefinitely. Inside it the signs may oscillate without end between their bounds, shown here as two signs in two colors. A path that begins outside the kernel leaves the survivable region, where collapse becomes only a matter of time, even when the current reading still looks acceptable.
Viability theory, the primary backboneThe viability kernel is the survivable region in the space of vital signs. Staying inside it is the whole objective, and inside it the system may oscillate without end.
Set-invariance controlThe correction mechanism. Proportional, robust regulation that keeps the state inside the safe set under bounded shocks.
Requisite variety and the Viable System ModelThe information condition and the governance architecture. Enough internal variety to match the world, organized across recursive levels.
Ecological resilience and the adaptive cycleThe basin reading and the logic of renewal. Absorb disturbance, and when a basin fails, reorganize into a better one.
Open questions
  • Viability kernels are exactly computable only in low dimensions, and eight vital signs sit just past that ceiling. The kernel also presupposes dynamics and a constraint set that civilizational data do not hand over cleanly, so it is a target for approximation and argument rather than a solved object.
  • Every formal guarantee assumes that what is measured stays honest. Once a vital sign becomes a target it is gamed and stops measuring what it did (Goodhart; Campbell), so the mathematics is a discipline for thinking rather than an autopilot.
FC–02

Anchored in Evidence

Stable

The commitments are not only arguments. Several rest on established empirical work, and saying which is part of being falsifiable. The knowledge commons is the clearest case. Elinor Ostrom showed that communities govern shared resources durably, without privatization or central control, when their institutions satisfy a set of design principles, and that knowledge itself behaves as a commons whose threat is enclosure rather than depletion (Ostrom; Hess and Ostrom). Aeviterna's governance reads as those principles applied at civilizational scale: clear boundaries and participatory rule-making, monitoring that keeps the commons auditable, graduated sanctions that prefigure the justice tiers, accessible conflict resolution, and nested enterprises that keep a large system polycentric so failures stay local.

What makes knowledge different also makes its governance more necessary, not less. Because understanding is non-rival, a knowledge commons is not threatened by depletion through use but by enclosure, gatekeeping, and underprovision, so openness alone does not protect it. It still needs the boundaries, monitoring, and recognized self-governance that any commons does, for different reasons than a fishery faces.

The vital-signs instrument rests on critical-transition science. As a complex system nears a tipping point it slows down, recovering more sluggishly from small perturbations, and this shows up as rising variance and autocorrelation before a regime shift, across very different systems regardless of mechanism (Scheffer and colleagues). That is why the framework watches a few load-bearing variables for loss of resilience instead of optimizing a single target. Driving one variable to its maximum usually narrows the basin and moves the system toward the edge, so a band with both a floor and a ceiling is the right object, not a maximum.

Cognitive sovereignty rests on the economics of attention. Attention is scarce, so a wealth of information creates a poverty of attention (Simon); that scarcity has been industrially contested for more than a century (Wu); and the contest now reaches perception and behavior themselves, shaped at scale by systems that profit from the shaping (Zuboff). Together these establish that perception is a contested civilizational resource, which is the thing cognitive sovereignty exists to protect.

Knowledge commons, after OstromClear boundaries, congruent rules, collective choice, monitoring, graduated sanctions, conflict resolution, recognized self-governance, and nested enterprises, applied as the commons' operating rules.
Vital signs, after SchefferCritical slowing down gives generic early-warning signals, so resilience is watched across a few variables rather than a single score being maximized.
Cognitive sovereignty, after Simon, Wu, and ZuboffAttention is scarce, industrially contested, and now shaped at scale, which is why perception needs active protection.
Open questions
  • Ostrom's principles were drawn from bounded, often physical resources, and their transfer to global digital knowledge is an active research question. Ostrom herself warned against treating the principles as a blueprint, so they validate the governance posture more than any specific mechanism.
  • Early-warning signals carry false positives and false negatives, and not every shift is preceded by slowing down, so they indicate that resilience is declining rather than naming a date for collapse.
  • The attention-economy works are critical and historical rather than predictive, so they motivate the commitment without fixing how much perception is captured or how much sovereignty is recoverable.
FC–03

Where Aeviterna Stands, and How It Fails

Open

Aeviterna is a new configuration of old primitives more than a set of new ones, and naming its neighbors is part of arguing in good faith. It shares with liberalism a protected sphere that no majority may vote away, expressed here as the inviolable dignity floor (Rawls; and on dignity as rank, Waldron, with the German Basic Law as the working model). It shares with republicanism the aim of freedom as non-domination, and answers it in time as well as in space. Decaying standing attacks entrenched power by letting authority lapse unless it is renewed (Pettit). It couples deliberative legitimacy to the governance modes, so reasoned and inclusive decision-making is the default and the modes decide when the gears change (Habermas). Its vital signs are a capability-style reading of real human functioning rather than output (Sen; Nussbaum). Its corridor between a social floor and an ecological ceiling is the safe and just space of degrowth and doughnut economics, generalized from throughput to governance itself (Hickel; Raworth). And its sensing-and-correcting loop is management cybernetics, with one difference that matters. The set point is dignity, the actuator is the redistribution of power, and the controllers themselves decay (Beer).

What is new is the combination: oscillation chosen over equilibrium, authority that decays by default, and a mode-switching structure bound to measured thresholds. Each is contestable, and together they are a recognizable new position rather than a relabeling of an old one.

The dignity floor deserves its own note, because it is the one place the framework refuses the usual calculus. Following the German Basic Law, dignity is treated as inviolable and non-balanceable rather than as a weighty interest to be traded off, the logic by which a constitutional court can strike down a measure that would sacrifice some lives to save others. That anti-balancing core is what keeps the rest of the adaptive machinery from optimizing a person away. The floor's absoluteness now attaches to an enumerated list of prohibited treatments with a burden-shift social minimum beneath it, the plural floor of EQ–06c, so what is held non-balanceable is specified rather than invoked, and the comparative record beyond one court carries it.

This is the framework's most important and least examined decision, and it deserves to be named as its own object rather than left as an exception to the banner. Call it the entrenchment boundary, the line between what is allowed to oscillate and the small set held fixed. Equilibrium is death describes the dynamics, but the boundary is the constitution, and the real moral work lives in where it is drawn. A framework whose whole identity is corrigibility has to be disciplined about its few non-corrigible commitments, so the test for admitting anything to the fixed side is strict. It must be minimal, command near-universal assent behind a veil of ignorance, and guard against a harm whose violation is irreversible. The dignity floor passes. Almost nothing else should, and whatever sits inside the boundary is itself interpreted, which is the problem taken up first in the Research Frontier as OQ–00.

The same machinery that makes the framework adaptive is its largest attack surface, because every threshold is a lever someone can pull. The strongest known failure modes, and what the design places against each, are set out below. The pattern across all of them is the same. Robustness comes less from the elegance of the loop than from what is held outside it as non-negotiable.

Failure modeHow it bitesWhat resists it
Oligarchy (Michels)Coordination breeds a permanent expert caste that entrenches itself.Decay applied to administrators too, selection by lot for some roles, mandatory rotation, open tooling.
Capture (Stigler)Concentrated interests take over the bodies that measure or certify.Adversarial and redundant measurement, citizen audit, decaying tenure for measurers, conflict-of-interest firewalls.
Backsliding (Levitsky and Ziblatt)An elected actor uses the rules to dismantle the rules.An un-suspendable core, a supermajority plus independent triggers for escalation, protected press and opposition.
Emergency abuse (Schmitt; Agamben)Crisis powers free the executive and then never expire.Automatic sunsets, externally audited triggers rather than executive discretion, a non-derogable core, a ratchet toward recovery.
Metric gaming (Goodhart; Campbell)A vital sign becomes a target and is optimized instead of the reality it tracked.Rotated and partly qualitative indicators, measurement decoupled from those it rewards, metrics treated as evidence for judgment.
Technocracy (Scott)Value choices are reframed as technical readouts and removed from contest.Thresholds set democratically and revisably, a deliberative override of any automated trigger, models and assumptions published.
Unaffordability (the priced record)The stack costs more than adopters can carry in money or member time, or its cheap instruments are starved for savings that round to zero, the way the record's reading and counting organs actually died.The two-currency ledger and per-scale overhead band of EQ–02f, the ratified ceilings of F–10, and instrument finance held at the entrenchment level (EQ–02e) rather than in the annual budget.
Unenforceability (the borrowed court)The rules bind no one. Internal law goes unwritten as enforceable contract, the arbitration path is never seated, and the first real breach is resolved by whoever has the most patience for a lawsuit, or the lender of the courts forecloses the borrow it was never asked to formalize.The enforcement ladder of EQ–06f, every organ naming its rung, the borrow written in peacetime as bylaws with a seated arbitration path, stakes sized to survive proportionality review, and the federation gatekeeping something members cannot cheaply replace.
Economic siege (the priced war)The polity is not conquered but priced out. Capital withdraws, clearing is denied, a subsidized rival sells below cost, the skilled are peeled away, and the walls that exist protect against the army that never comes.The economic wall of EQ–04f, living mutual credit, the mode-gated reserve, the economic black-start with published days-of-cover, federation treaty lines, and vested demand, with fiscal discipline (EQ–02f) denying the besieger an interior collapse.

A framework that can name the ways it would fail has a better chance of not failing in those ways. That is the reason for writing them down, and the Model Charter turns these resistances into clauses a community can adopt, with the overhead band's clause owed to a coming charter version.

Open questions
  • The residual this section cannot close is its own vantage. A red team drawn from inside the framework can only name the failure modes it can imagine, so the table above is bounded by the imagination of its authors, and this critique has not yet been performed by an adversary with something to lose. FC–04 takes up the first gap the table exposes; the unmodeled remainder waits for hostile readers.
FC–04

The Guardian of the Core

Draft

The deepest item in the failure-mode table in FC–03 is the apex, an anti-capture stack that climbs to a single seat no one watches. The same problem surfaces twice more in the framework, in EQ–04b as the question of who adjudicates the boundary without becoming the apex, and in EQ–06c as the risk of freezing a single interpretation of dignity. It is one problem, the interpretation of the non-derogable core, named in the Research Frontier as OQ–00, and this section is the framework's working attempt at it. The aim is not a system with no sovereign, which the precedents say is impossible, but a chair no one can settle into, an interpreter that is always one challenge from being overruled and one cycle from dissolving.

Two moves do most of the work. First, keep the core small and lit, since a non-derogable core that is tiny, explicit, published, and versioned is a worse capture target than a large vague one, because capturing it becomes visible and costly. Second, split the power to fix what the core means from the power to apply it. Its settled meaning is authored by no single seat and drifts only by a large, representative supermajority of the federation, on the model of peremptory norms, under a standing rule that no body may declare that meaning closed. A panel only ever renders an operative judgment for the case in front of it, binding for now and revisable after, with the losing reading kept in the open as legitimate.

The panel that applies the core is built to be the opposite of a sovereign, and it is assembled from the framework's own primitives. It is plural and never one. It is sortition-leavened, a lay majority drawn by lot with a minority of domain stewards whose standing already decays under EQ–05, so expertise informs the reading without ruling it, the same plural and decaying pattern EQ–03b uses for the epistemic monitor and EQ–06b uses for justice. This panel's constitution, generalized against the world's record of reader institutions, is now the framework-wide primitive, the reading seat (EQ–06e), and the framework's other instrument-readers instantiate it rather than rebuild it. Its terms are short and non-renewable, so there is no one to please for reappointment, and it publishes its reasoning and its dissent. It is not watched from above, because adding a watcher rebuilds the tower; the separated organs interpret the core against one another in a closed loop, with the right of exit in EQ–04b as the final discipline, the circle shown in the figure below.

The hardest residual is not the panel but the framing layer, whoever defines the vital signs and certifies which is the weakest, because that is the smallest and least watched seat. The design aims it back into the open: more than one independently produced methodology for each sign, so no single index decides; the methods published as forkable code; a right for any member to submit a competing reading and force a contest, at a cost to the original framers' standing if it loses; bounties for showing a reading was framed to a conclusion; and the framing seat itself filled by lot. The last danger is that a decaying guardian comes to believe only it can protect the core and resists its own sunset, the institutional form of an agent that fights its off-switch once it is certain, so the structure keeps it unsure it is needed. The mandate decays automatically and is renewed only from outside, a successor panel always stands ready, no panel is ever the sole interpreter, and incumbency carries little reward.

The tower, to a throneThe circle, no apexwho watches this one?monitor of the monitora monitorthe systemapex: unwatchedeach guardsthe nextThe anchor, to realitythe framedated forecastreality, the outcomesno watcher above, the world grades the frame
The oversight regress has two shapes. Stack a watcher over every watcher and the tower climbs to a single unwatched apex, a throne by another name. Bend the same watchers into a closed loop and no apex forms, each guarding the next, which is how the guardian of the core (OQ–00) is meant to stay plural and decaying rather than sovereign. This panel's constitution is generalized as the reading seat (EQ–06e), and every instrument-reader in the framework now instantiates the circle's logic.
A small, lit coreA tiny, published, versioned non-derogable core, where capture becomes visible and therefore costly rather than hidden in a large vague one.
Authorship split from operationMeaning is set by a federation supermajority and never declared closed; a panel renders only revisable operative judgments, with dissent kept in the open.
A panel that fails the sovereign's testsPlural, sortition-leavened, decaying, non-renewable, reasoned, and externally seated, so no seat is permanent, general, or self-reinforcing.
The circle, not the towerThe organs interpret the core against one another with exit as the final check, so no apex is ever added above them.
Watch the framing layerPlural competing methodologies, a right to re-frame at a cost, capture bounties, and framers chosen by lot, because the framing seat is the real residual apex.
A mandate it cannot keepAutomatic decay, renewal only from outside, a standing successor, and never a sole interpreter, so the guardian has nothing to entrench for.
Honest verdict and residuals
  • This does not dissolve sovereignty. The regress finally rests on a self-enforcing equilibrium of the federation, which is costly to capture, not impossible.
  • The residual apex is the framing layer, now made plural, forkable, contestable, and rotated rather than absent.
  • The half-lives are calibrated as bands, not derivations (EQ–05d), and unvalidated in deployment (OQ–02); a broadly popular capture can still clear the contest (OQ–04); and the authorship layer presumes the federation of OQ–05 and enough civic attention to run the contests.
  • Live test
    • This is stewardship and not a sovereign in disguise exactly while a divergent reading can be forced and win, the mandate truly decays under external renewal, and a captured reading is detected and costs its authors their standing. If those fail, the sovereign has re-formed, and that is the date to mark. The aim is to develop OQ–00, not to claim it closed.
Appendices
EQ–A1

Appendix A: Core Concepts Glossary

Stable
Adaptive Stability The capacity of a civilization to remain resilient through bounded oscillation and self-correction, rather than through the maintenance of static equilibrium. The primary design target of Aeviterna.
Adoption Tracker The framework's longitudinal dataset coding real organizations against the readiness codebook, with anchored scores, per-cell evidence and confidence, decline coded as data, and versioned releases. Falsifier F–01's measurement instrument in embryo. Lives at the tracker instrument.
Age Pyramid The demographic load-bearing test on the care substrate: whether care entitlements and the governance-participation budget can both be met from a shrinking working-age cohort as a population ages. Adds a scarcity price on care and a circuit-breaker on participation. Specified in EQ–08f.
Aggregation Firewall The rule that care is counted only by anonymous, sampled instruments, at a level of aggregation from which no individual's schedule can be reconstructed, so visibility never becomes surveillance. Specified in EQ–08c.
Anti-Capture The architecture that resists a system's drift toward concentration, combining rotation, distributed memory, decaying standing, and separation of powers, each paired with a named adversary. Specified in EQ–06.
Attic The retirement process for structures, roles, and mechanisms: every clock has a named owner, failing to muster affirmative renewal is itself retirement, and what retires is preserved read-only, composted rather than deleted. Specified in EQ–10b.
Binding Stack The five commitments every spine function carries at once: ceremony-grade transparency, removal power held outside, a decaying mandate, fork-ready state, and no fixed shares. The record shows them failing separately and holding only together. Specified in EQ–04b.
Biological Resilience The eighth vital sign: the health and preparedness of a population read as one interdependent system of people, animals, and environment (One Health), combining outbreak preparedness, the antimicrobial-resistance burden, and the social health gradient.
Black-Start Drill The framework's disempowerment alarm: every consequential function that has delegated judgment to an automated layer keeps a tested capacity to run degraded without it, exercised on a schedule, with the drill's scored result carried as the reading. The alarm is a calendar, not a sensor. Specified in EQ–03e.
Bootstrap Ladder The five-rung sequence by which a community assembles a corpus without a sovereign: member capital, patient debt, the veto lock, conversion, and captured rents, with the yield of the last rung recycled into the next community's first. Tested by F–11. Specified in EQ–02e.
Boundary Layer How an Aeviternal polity coexists with, trades with, and defends against polities that are not Aeviternal: bounded below by the dignity floor and above by non-domination, with defense capped at the non-offensive.
Bounded Coherence Coherence held between a floor and a ceiling: enough shared understanding to coordinate, and enough generative ambiguity to keep adapting. A civilization that maximizes coherence optimizes away the variation it needs, so bounded coherence is treated as a design requirement.
Bounded Oscillation Movement within survivable limits. The normal mode of living systems. Distinguished from both paralysis (no movement) and runaway deviation (movement beyond recovery).
Calibration Schedule The published register naming every clocked role, its half-life band, and its stagger, bound by the no-exempt-adjacents and decay-pairs-with-memory rules. Specified in EQ–05d and carried into Article 4 of the Model Charter.
Capture The structural drift of a system toward concentration as those who benefit from it suppress the feedback that would correct it. Distinguished from cognitive capture, which targets perception; here the sense is institutional, the seizure of a system by motivated insiders. The target of the anti-capture architecture in EQ–06.
Capture Basin The set of states from which a system can still be steered back into its viable region in finite time. The formal counterpart, drawn from viability theory, of the framework's promise that recovery must remain reachable. Specified in FC–01.
Capture Cascade The framework's campaign model of institutional capture: six recurring stages from entry to lock-in, each watched by a named anti-capture mechanism, so that breaking any one link stalls the campaign. A repertoire with a soft ordering, not a law. Specified in EQ–06.
Care and Maintenance The unpaid and underpriced work of sustaining people and things, the submerged base of the economy that the market counts last. Aeviterna counts it as wealth rather than overhead.
Civilizational Memory The capacity to carry knowledge, judgment, and obligation across generations, held in a distributed archive with redundant copies and restoration drills so that a broken link does not erase what was learned.
Civilizational Metabolism The economy read in physical terms: a one-way throughput of energy and materials, funded by net energy and held under an absolute ecological cap. Enough is the band between an energy floor and that ceiling.
Cognitive Capture The systematic redirection of attention, emotion, and belief in service of concentrated interests, through adaptive algorithmic and institutional means, without the knowledge of those being shaped.
Cognitive Ecology The governance requirement to maintain productive tension between coherence and generative ambiguity. A civilization that optimizes for total coherence loses the adaptive variation necessary for long-term survival. Aeviterna treats cognitive ecology management as a primary governance function.
Cognitive Sovereignty The ability of individuals and societies to understand and contest the systems shaping their perception. A prerequisite for meaningful political participation.
Cognitive Terraforming The large-scale reshaping of perceptual and epistemic conditions by AI-mediated systems, altering the landscape of what populations can think, believe, and demand. A precursor condition to legitimacy collapse.
Commonwealth Trust A chartered single-purpose trust that turns the rents of common wealth, land, resources, and their kin, into the dignity floor's named funding mechanism, built from the features that survived being raided: a constitutional corpus, a formula payout, an enumerated mandate, a reading seat chartered as deep as the money, a second key outside the spending coalition, and a regrowth tripwire (F–07). Specified in EQ–02e.
Consent Layer The organ that replaces founding consent with a standing practice: norms bind while they survive reasoned objection, membership is staged, explicit, and revocable with vested exit, and the unconsentable are held in bounded trusteeship. Specified in EQ–04e.
Container Rule The funding architecture for meaning and belonging: public money flows by formula through a form audit into self-governing associations and places, and a firewall stops the funder from ever reaching content. Specified in EQ–08b.
Corrigible Capable of being corrected. A corrigible system retains the structural capacity to detect its own errors and change course. Aeviterna treats corrigibility as a primary design requirement.
Definer's Seat The organ that constitutes who frames the question and sets the option set for a sortition body, since random selection secures the seat's occupant but not the neutrality of its inputs. Guarded by adversarial, lot-drawn framing, a firewall, a published provenance trail, and a watcher. Specified in EQ–06g.
Degeneration Condition The workshop's pre-registered, dated, outcome-blind stopping rule. By a fixed date the program must clear external tests of adoption, contestation, and forecast calibration, or declare itself degenerating and archive its corpus as parts. Registered as F–09 and specified in EQ–10b.
Demand Term The transition engine's account of why anyone bears the stack absent a crisis, a function of the protected value's visibility, the social proximity of the ask, and the residual compliance load, with screening cost placed at the committed boundary and the visitor's door kept cheap. Tested by F–12. Specified in EQ–04d.
Dignity Floor The inviolable minimum beneath which no person may fall, the one value the framework never trades away or puts to a vote. The set point of the whole system.
Disagreement Capacity The seventh vital sign: how much open, unresolved conflict a society can hold without fracturing or silencing it. The political twin of epistemic biodiversity.
Disclosure Clock A hard, short, automatic deadline on which knowledge produced with commons resources enters the commons, making release the infrastructure default rather than an author's virtue. Adopted in EQ–07.
Economic Siege Hostile pressure on a polity's external flows, capital withdrawn, clearing denied, prices subsidized below cost, talent peeled away, inputs foreclosed, arriving as prices rather than armies. The wall against it is specified in EQ–04f.
Economics of Enough The material economy oriented to sufficiency rather than growth: meeting needs above a social floor and below an ecological ceiling, and counting stewardship and regeneration as wealth. The economic expression of adaptive stability.
Enforcement Ladder The five sources from which a rule binds: reputation, vested stakes, structural locks, federated sanction, and borrowed courts, with every organ naming its rung, borrowed or own. Specified in EQ–06f.
Entrenchment Boundary The line between what a polity allows to oscillate and the small set it holds fixed, with strict tests for admission to the fixed side: minimal, near-universal behind a veil of ignorance, and guarding an irreversible harm. Named as the framework's load-bearing constitutional decision in FC–03.
Epistemic Biodiversity The preservation of diverse, contradictory, unoptimized, and non-instrumental modes of thought within a civilization. Valued not aesthetically but functionally: no designer can know in advance which currently-marginal cognitive approaches will be essential under future conditions.
Exploitative Opacity Concealment that serves concentrated power: corruption, hidden extraction, institutional secrecy, covert manipulation. What transparency is designed to defeat. Distinguished from generative opacity.
Fallback Theater The anti-pattern the black-start drill is built against: a fallback that exists as documentation and dies on first contact, verified on paper, drilled against the wrong failure, or staged in mild conditions. Its only defense in the record is realism review plus a published result. Named in EQ–03e.
False Aeviterna The costume of legibility worn over unchanged concentration: an institution that adopts the framework's vocabulary and instruments while power stays where it was. The anti-capture discipline carried into every transitional structure exists to expose it. Named in EQ–04.
Falsification Register The standing public register of the framework's falsifiers, where every load-bearing claim carries its would-be refuter from the day it ships, on the model of registered reports. Specified in EQ–10b.
Fix-First Gate The rule that new construction above a threshold is funded only where renewals meet their floor against depreciation, so a published maintenance debt binds allocation instead of decorating it. Specified in EQ–08c.
Forecast Register The workshop's own dated, scored predictions on the vital signs, graded by the Brier rule, so the framing layer that writes the framework is held to the same reality anchor it asks of a reading seat. Specified in EQ–10b.
Fork Protocol The conduct rules that make a split legitimate: a new name, a divergence document naming the disagreement and its re-merge evidence, an interoperable commons, and no stripping of standing. A split that claims the name and states no condition is a schism. Specified in EQ–10b.
Framing Layer The stratum where a mediator selects, orders, and frames what a polity sees before deliberation begins. Concentration at this layer is the deepest form of cognitive capture, treated in EQ–03d.
Generative Opacity Concealment that serves adaptive capacity: privacy, emotional ambiguity, exploratory uncertainty, unoptimized creative space. What transparency must be careful not to destroy. A structural component of cognitive ecology.
Governance Modes The four settings of one government, abundance, scarcity, crisis, and recovery, chosen by measurable conditions, where any rise in centralization expires by default on the way out. Specified in EQ–09.
Governance Overhead What the stack costs to run, kept as a published two-currency ledger of money and member-hours and held inside a per-scale band whose breach in either direction, bloat above or starvation below, is a reading. Bounded by falsifier F–10. Specified in EQ–02f.
Justice of Light The framework's response to harm, held as legible and correctable as ordinary governance and structured in tiers (containment, moral firebreaks, and non-reintegration) rather than hidden or unbounded.
Knowledge Commons Shared epistemic infrastructure treated as a foundational civilizational resource. Includes scientific knowledge, governance archives, civic memory, and educational infrastructure.
Legibility Budget A published, cumulative ceiling on how much individual-level information an instrument's readings may leak, owned by a reading seat, floored by an outright ban on individual-level reads, with costs named and deletion as contestable as creation. Specified in EQ–03b.
Legitimacy Collapse The condition in which governing systems lose the consent of the governed not through overt oppression but through the accumulated erosion of legibility, accountability, and epistemic commons.
Lite tier The first priced rung of the on-ramp, the minimum viable subset EQ–02f names: a sunset on every role, a quarterly review of three chosen vital signs, and exit codified in peacetime. Sized to be adopted in one meeting for under a thousand dollars a year at cooperative scale. Published as Aeviterna Lite.
Monetary Ecology Treating money as a plural ecosystem rather than a monoculture, with currencies tuned to different functions so the system sits in a resilient middle between brittle efficiency and stagnant over-proliferation.
Plural Floor The dignity floor rebuilt on plural anchors: an absolute list of enumerated prohibited treatments, a resource-sensitive social minimum enforced by burden-shift, and rules keeping both tiers agent-facing, non-paternalist, and remedially flexible. Specified in EQ–06c.
Polycentric Federation Scaling through many overlapping, semi-independent centers linked by shared protocols rather than a single capturable apex (after Ostrom). Specified in EQ–04b.
Reading Seat The framework's one constitution for any body that reads an instrument for the polity: plural with no single principal, tenured past its principals, never funded by the read, publishing its raw readings, firewalled from the approval layer, and bound by the override discipline. Specified in EQ–06e.
Reality Anchor The discipline that binds a reading seat's frame to the world. A frame is admitted only when it is pre-registered and dated and made to predict the near-term early-warning signals that run ahead of a transition, scored against sources fixed and plural before the frame is admitted, so a seat that reads the world well earns renewal and one that reads it badly decays out on the evidence. It halts the oversight regress a third way, terminating not in an apex or a circle of mutual watchers but in reality, the one watcher that does not share the framer's priors. Its reach ends at the warning-free transition, which shows no early signal to register against. Specified in EQ–06e.
Removal Round A scheduled version of the framework that cannot ship without retiring at least one mechanism, structure, or claim into the archive: improvement by subtraction on a clock. Specified in EQ–10b.
Reversibility and Repair Treating the capacity to undo, repair, and wind down as a first-class design property, pricing irreversibility as a cost and funding end-of-life provision in advance.
Self-Gauge The reading seat's instrument turned on itself, holding the concentration of framing power, how few and how unwatched the readers have become, in a band with its own decay clock. It sits beside the panel of eight rather than joining it, because the eight read the civilization and this reads the reader. It halts nothing on its own. Defining the gauge is itself a framing act, held only by a rival gauge kept live beside it, and where that rival is captured or never funded the gauge is circular. Specified in EQ–06e.
Semantic Stagnation The terminal condition of a civilization that has optimized away its capacity for novelty, meaning mutation, and existential flexibility. Not caused by oppression but by excessive coherence: the elimination of the productive deviation that adaptive systems require.
Spine Test The three-condition admission test for lifting a function to the federation's center: indivisibility, speed of failure beyond federated response, and non-producibility by voluntary adoption. All three or it stays polycentric. Specified in EQ–04b.
Standing Domain-specific, contribution-dependent, time-decaying influence within an Aeviternal system. Distinguished from sovereignty by its conditionality, reversibility, and specificity.
Standing Threshold One of the five tests a complementary currency must clear to earn and keep standing in the monetary ecology: a real credit or clearing function, an anchor that binds circulation, countercyclical use observed, operations that survive their donors, and governance that survives its founders. Drawn from the survivals against the deaths and read at every review. Specified in EQ–02d.
Stewardship Temporary and accountable maintenance-oriented influence focused on continuity rather than domination. The appropriate mode of social power in an adaptive civilization.
Temporal Sovereignty Governing not only how much a civilization extracts and how it perceives, but how fast it changes, holding each pace layer to its own rate band.
Thin Spine The few functions a polycentric federation concedes to the center, root of trust, verification, and the crisis backstop on the current record, admitted by the spine test and held accountable by the binding stack so coordination never becomes an apex. Specified in EQ–04b.
Transcription Ban The rule that no emergency-born power may enter ordinary law during the emergency or its cooling-off period; permanence requires de novo justification and an independent audit of the power's emergency record. Specified in EQ–09c.
Transition Engine The capacity-gated cascade model of adoption: prefigured capacity gates a cascade of adopters carried by momentum, firing at an unschedulable legitimacy window against an adaptive incumbent. Specified in EQ–04d.
Transitional Aeviterna Bridge-oriented migration structures enabling gradual movement toward healthier coordination systems, designed for partial adoption and operation inside imperfect existing realities.
Viability Kernel The largest set of states from which a civilization can keep its vital signs inside the survivable region indefinitely. The formal target of bounded oscillation (after Aubin).
Vital Signs The small set of measurable variables a civilization watches as bands to hold rather than numbers to maximize. The framework's sensing instrument, currently eight.
Workshop Charter The framework's discipline turned on the workshop that writes it: succession by fork under an open license, a dated degeneration condition, a kept falsifier register, and plural review, so the maintainers are held to the standard the framework sets for any institution. Specified in EQ–10b.
EQ–A2

Appendix B: The Research Frontier

Stable

These are the framework's standing hard problems, carried forward by design rather than gaps to be closed before release. A living framework keeps a frontier; each problem below has a proposed direction and remains open to work.

OQ–00
The guardian of the core, prior to all the others

A clause that overrides every other and never sunsets, the dignity floor and the un-suspendable protections, cannot apply itself, so some body must interpret it and read the un-summable vital signs to set a mode, and an interpreter that is permanent, general, and answerable only to the clause it interprets is, by P–05's own definition, a sovereign. The anti-capture regress sharpens it, since the resistance to capture is itself capturable (P–02), and reading the panel by its weakest sign is a discretionary act, so set by the review, never by discretion is an overclaim.

Direction

Do not abolish the interpretive function, which is irreducible, but strip it of all three sovereign predicates and convert a throne into a stewardship, with an it-is-not-in-heaven rule that bars any single seat from a final reading (jus cogens; halakhic machloket), a plural, part-sortition, term-limited, non-renewable panel that publishes its reasoning and preserves dissent (the German constitutional court; Bagg), a guardian mandate that itself decays, and the separated organs watching one another in a closed loop rather than up a tower (Hurwicz; Madison); the residual hazard is that sovereignty is relocated, not dissolved, to whoever frames the methodology, so the defensible claim is that Aeviterna replaces a permanent personal sovereign with a contestable stewardship, not that it has none. The stewardship's constitution is now specified once as the reading seat (EQ–06e), extracted from FC–04 and the world's record of reader institutions and instantiated across every seat that reads an instrument, so the interpreter problem is answered by one auditable design rather than five bespoke ones; the recursion keeps its honest shape, the seats watching each other in a circle, and the residual keeps its name, a reading seat is never finally held, only currently defended.

Falsifier

If no structure escapes both the regress and the P–05 definition, the claim to replace sovereignty with stewardship is refuted, not merely weakened.

OQ–01
The transition mechanism problem

How does a system governed by extractive actors begin to implement structures designed to limit extraction? What is the forcing function?

Direction

Now modeled as a capacity-gated cascade over three variables: the adopter fraction, the incumbent's legitimacy, and prefigured capacity. A transition becomes self-sustaining only when, at a moment of legitimacy decline, both the adopter fraction and the built-out capacity reach the convention-flip threshold of roughly a quarter of the relevant population (Granovetter; Centola; Rogers and Moore's chasm). The controllable levers are capacity and the dense local reinforcement that lowers that threshold; the trigger, the legitimacy collapse itself, is real but unschedulable in principle (Kuran). Prefiguration is therefore the project of building to critical-mass scale and absorptive throughput before the window opens. This is conditional readiness, not a scheduled victory. Now integrated in the body as EQ–04d, the transition engine, with a momentum term after the physics-of-dissent result, capacity made auditable, and an incumbent-response drain for repression-learning and absorption; the trigger stays unschedulable.

OQ–02
The standing operationalization problem

Decay rates, revocation processes, and cross-domain standing calculations remain unspecified. Without these, EQ–05 is a principle, not a mechanism.

Direction

Now specified as the standing engine in EQ–05: a per-domain reputational stock decaying with an attested-contribution inflow to a bounded ceiling, the half-life anchored to the domain's measured knowledge turnover and clamped into fast, mid, and slow tiers with scheduled refit; standing held as a non-fungible vector with a cross-domain concentration tripwire; revocation by graduated sanctions with a collusion penalty; and the framing seat and adjudication filled by lot (Machlup; Densen; Ostrom; Bagg). The half-life tiers are now calibrated in EQ–05d: banded per domain class with named anchor strengths, an independence inversion after the central-bank and judicial-tenure evidence, and a no-exempt-adjacents perimeter drawn from the term-limits record. Falsifier F–04.

Residual

Obsolescence is not cleanly exponential, the metric is gamed not gaming-proof, sybil resistance is unsolved, and fast rotation can leak expertise to unelected permanents.

OQ–03
The threat modeling gap

EQ–06 lists anti-capture mechanisms without naming the specific adversarial scenarios they are designed to defeat. Each mechanism needs a paired failure mode.

Direction

Give every mechanism a named adversary in a STRIDE-style threat model, rotation versus regulatory capture (Stigler), distributed memory versus record-tampering, decay versus the iron law of oligarchy (Michels), and separation versus power fusion (Pareto). Now worked in EQ–06 as the capture cascade: the security field's move from threat lists to campaign models (the kill chain; ATT&CK) applied to the documented record, six recurring stages from entry to lock-in, each watched by a named mechanism, with two structural rules the record forced, friction as a defense in itself and the reversal asymmetry that weights detection toward the early stages. Carries falsifier F–06.

Residual

The stage order is a tendency, not a law; elite recirculation still moves through networks that satisfy every mechanism formally, tracked by network rather than individual concentration; and a capture that owns the indicator layer or moves all stages at machine speed behind it defeats the cascade, which is built to be revised from exactly such a case.

OQ–04
The manufactured crisis problem

What prevents actors from creating conditions that trigger emergency governance modes, extending centralized authority under cover of necessity?

Direction

Separate the crisis declarer from the beneficiary, make emergency powers default-expire with an asymmetric re-authorization burden, allow nested-unit veto, and decay emergency standing fastest (Agamben; Higgs; Ginsburg and Versteeg); the residual hazard is a broadly popular emergency that clears every democratic check. Now worked as EQ–09c, the manufactured crisis: ignition instant and reviewable, renewal held by the unrewarded on an escalating ladder, observable triggers with popular approval inadmissible, and a transcription ban against the ratchet the record actually shows.

Residual

A capture that owns the indicators, the attesters, and the frame at once still clears it.

OQ–05
The scale problem

Aeviternal governance structures are most legible at the community or institutional scale. The framework has not yet specified how they aggregate to civilizational scale without reproducing the concentration they are designed to prevent.

Direction

Scale fractally through Ostrom's nested enterprises coordinated by shared protocols rather than a central platform, with base units near the Dunbar bound and only thin protocol authority at higher layers; the residual hazard is indivisible global problems that pull authority back toward the center. The strong form of that worry is false, since polycentric climate governance and pledge-and-review are real counterexamples (Ostrom), but the weak form holds, so the honest posture is a thin spine, developed in full as EQ–04b (Polycentric Federation). The spine is now a specified mechanism rather than a concession. The spine test (indivisibility, speed, non-producibility by adoption) decides what may be lifted to the center, admitting root-of-trust, verification, and the crisis backstop on the current record, and the binding stack (ceremony-grade transparency, outside removal power, a decaying mandate, fork-ready state, no fixed shares) keeps an admitted function from freezing into an apex, every part carrying a named precedent that held and a named failure it is built against.

Residual

The stack is proven only below the threshold where great-power interest crosses a function; above it the record shows bindings breaking, which is exactly why the spine is held thin.

OQ–06
The consent problem

Transitional structures operate inside existing systems whose participants have not consented to Aeviternal design principles. How does the framework handle the legitimacy of its own implementation?

Direction

Now worked as EQ–04e, the consent layer. Consent as a founding act is declined with the philosophy rather than against it (Hume; Simmons); legitimacy is re-grounded in effective contestability (Pettit) with discourse as the regulative ideal (Habermas); membership is staged, explicit, and revocable, with vested, pre-written exit, the kibbutz lock-in named as the design warning (Abramitzky); the unconsentable are held in bounded, contestable trusteeship under the open-future limit (Feinberg); and the institution's shadow earns contestation without membership, answering the demos that otherwise explodes (Goodin).

Residual

First imposition, consent quality, and future people, who are represented, never consulted. The claim is justified, minimized, contestable coercion, never consent-grounded obligation.

OQ–07
The monetary monoculture problem

The framework treats diversity as resilience everywhere except in money itself, where a single national currency optimized for efficiency may be as brittle as any monoculture.

Direction

Read money as a flow network with a window of viability between too much efficiency and too much diversity, and admit a currency to the plural ecology only when it clears the standing thresholds now worked into EQ–02d: a real credit or clearing function, an anchor that binds circulation, countercyclical use observed, operations that survive their donors, and governance that survives its founders, drawn from the survivals against the deaths across a ninety-year record (Lietaer; Ulanowicz, Goerner, Lietaer, and Gomez; Stodder; the Sardex and WIR records).

Residual

Complementary currencies still fade once a founding crisis passes unless an anchor binds them to ordinary time, and the thresholds are tested at every review rather than assumed solved.

OQ–08
The perimeter problem

Every adversary the framework models is an insider bound by its rules, and it has no theory of the outsider, the rival polity that adopts none of this. Among defectors the internal virtues invert into attack surfaces, since an open commons leaks strategic position, non-derogable protected dissent is a channel a hostile state funds and the sacred core forbids closing, and decaying authority with a peacetime drawdown can be out-armed, while resilience matures over decades and conquest is decided in years.

Direction

Build a defensive-only perimeter from a porcupine posture compatible with decaying standing (Finland; Switzerland), a security commons of allied polities that raises the local benefit-to-cost ratio so cooperators are not in the well-mixed case (Axelrod; Nowak), a narrow, sunset-bound secrecy carve-out that splits transparency into publicity, surrendered at the perimeter, and accountability, preserved toward a cleared decaying oversight cell, and a deterrent with a deliberative human override, which is already the norm of credible deterrence; the residual hazard is the timescale mismatch, since the cluster must be built before the attack and the polity is most exploitable while assembling it. This direction is now developed in full as the perimeter of EQ–04c (The Boundary Layer).

Falsifier

If any polity satisfying the five commitments is reliably conquered or outcompeted by a non-adopting rival before its resilience advantage matures, with no recourse to an external patron, a protection the polity cannot generate for itself, the universality claim falls and Aeviterna is a tenant doctrine.

OQ–09
The transformative-AI problem

The framework calls itself a design for an AI-mediated civilization but has no organ for advanced AI, which is a cross-cutting accelerant of both capture (P–02) and sovereignty (P–05). It lowers the cost of capturing feedback, removes the constraint that made broad sovereignty expensive, moves inside the loop that reads the vital signs, and makes a fast or quietly irreversible transition the limiting case of adaptive stability.

Direction

Treat AI as a monitored actor that may hold scoped, revocable, decaying standing but never sovereignty; require plurality against model monoculture (Ashby), a right to human review of consequential automated decisions (GDPR Article 22), structured-transparency logging, and the decommissionability discipline of EQ–05c extended to frontier systems, with the transition read as a band-crossing event against the capture basin. The disempowerment residual now has an organ. EQ–03e, the black-start drill, imports the fallback-exercise discipline of power grids, aviation, elections, and payments, and carries the drill's scored result as the reading OQ–09 lacked, because the alarm is a calendar, not a sensor.

Residual

Gradual disempowerment can still erode the very feedback channels the framework reads (Kulveit and colleagues), a passed drill is a lower bound rather than a guarantee, corrigibility may not hold for a superhuman optimizer, and decaying AI standing has no precedent.

OQ–10
The more-than-human problem

Nature enters the framework only as a constraint, the ecological ceiling, and a reading, the throughput vital sign, with no constituency of its own, while every other protected interest is given an agent and a decaying-standing account. Extending standing to ecosystems revives the framework's own apex problem, since whoever speaks for a river or a forest can become its unaccountable sovereign.

Direction

Give nature standing through plural, decaying, contestable, sortition-leavened guardians built to fail the guardian-of-the-core tests (FC–04), scoped per system rather than to nature in general, with a bounded more-than-human floor, no knowing extinction, ecosystem collapse, or planetary-boundary breach, that mirrors the dignity floor without claiming equivalence, integrated with biological resilience (EQ–08d) and the future-persons standing of EQ–06c.

Residual

Anthropocentric capture, the undefinability of nature's interests, romanticism about a balance ecology no longer assumes, and popular capture at the framing layer (OQ–04).

EQ–A3

Appendix C: References

Stable

Inclusion signals engagement, not endorsement.

Abowd, J. M., & Schmutte, I. M. (2019). An economic analysis of privacy protection and statistical accuracy as social choices. American Economic Review, 109(1), 171-202.

Abowd, J. M., et al. (2025). A simulated reconstruction and reidentification attack on the 2010 U.S. Census. Harvard Data Science Review, 7(3).

Abbott, K. W., Genschel, P., Snidal, D., & Zangl, B. (Eds.). (2015). International Organizations as Orchestrators. Cambridge University Press.

Abramitzky, R. (2011). Lessons from the kibbutz on the equality-incentives trade-off. Journal of Economic Perspectives, 25(1), 185-208.

Acemoglu, D., & Robinson, J. A. (2012). Why Nations Fail: The Origins of Power, Prosperity, and Poverty. Crown.

Ackerman, B. (2004). The emergency constitution. Yale Law Journal, 113(5), 1029-1091.

Agamben, G. (2005). State of Exception (K. Attell, Trans.). University of Chicago Press.

Aguiar, M., & Hurst, E. (2007). Measuring trends in leisure: The allocation of time over five decades. Quarterly Journal of Economics, 122(3), 969-1006.

Alesina, A., & Summers, L. H. (1993). Central bank independence and macroeconomic performance. Journal of Money, Credit and Banking, 25(2), 151-162.

Allen, C. R., & Gunderson, L. H. (2011). Pathology and failure in the design and implementation of adaptive management. Journal of Environmental Management, 92(5), 1379-1384.

American Society of Civil Engineers. (2025). 2025 Report Card for America's Infrastructure. ASCE.

Antimicrobial Resistance Collaborators. (2022). Global burden of bacterial antimicrobial resistance in 2019. The Lancet, 399(10325), 629-655.

Armstrong McKay, D. I., et al. (2022). Exceeding 1.5 °C global warming could trigger multiple climate tipping points. Science, 377(6611), eabn7950.

Arnott, R. J., & Stiglitz, J. E. (1979). Aggregate land rents, expenditure on public goods, and optimal city size. Quarterly Journal of Economics, 93(4), 471-500.

Arrow, K. J., & Fisher, A. C. (1974). Environmental preservation, uncertainty, and irreversibility. Quarterly Journal of Economics, 88(2), 312-319.

Arthur, W. B. (1989). Competing technologies, increasing returns, and lock-in by historical events. Economic Journal, 99(394), 116-131.

Arthur, W., Bennett, W., Stanush, P. L., & McNelly, T. L. (1998). Factors that influence skill decay and retention: A quantitative review and analysis. Human Performance, 11(1), 57-101.

Ashby, W. R. (1956). An Introduction to Cybernetics. Chapman & Hall.

Aubin, J.-P. (1991). Viability Theory. Birkhäuser.

Avelino, G., Passos, L., Hora, A., & Valente, M. T. (2016). A novel approach for estimating truck factors. Proceedings of the 24th IEEE International Conference on Program Comprehension (ICPC).

Axelrod, R. (1984). The Evolution of Cooperation. Basic Books.

Ayres, I., & Braithwaite, J. (1992). Responsive Regulation: Transcending the Deregulation Debate. Oxford University Press.

Bachrach, P., & Baratz, M. S. (1962). Two faces of power. American Political Science Review, 56(4), 947-952.

Bagg, S. (2024). Sortition as anti-corruption: Popular oversight against elite capture. American Journal of Political Science, 68(1), 93-105.

Bainbridge, L. (1983). Ironies of automation. Automatica, 19(6), 775-779.

Balser, D. B. (1997). The impact of environmental factors on factionalism and schism in social movement organizations. Social Forces, 76(1), 199-228.

Bank of England. (2024). Forecasting for Monetary Policy Making and Communication: A Review (the Bernanke Review). Independent Evaluation Office.

Bayer, P., & McMillan, R. (2012). Tiebout sorting and neighborhood stratification. Journal of Public Economics, 96(11-12), 1129-1143.

Beer, S. (1972). Brain of the Firm. Allen Lane.

Berlin, I. (1969). Four Essays on Liberty. Oxford University Press.

Bermeo, N. (2016). On democratic backsliding. Journal of Democracy, 27(1), 5-19.

Bernanke, B. S. (2024). Forecasting for monetary policy making and communication at the Bank of England: A review. Bank of England Independent Evaluation Office.

Bernstein, A. S., et al. (2022). The costs and benefits of primary prevention of zoonotic pandemics. Science Advances, 8(5), eabl4183.

Bhatt, M. P., Heller, S. B., Kapustin, M., Bertrand, M., & Blattman, C. (2024). Predicting and preventing gun violence: An experimental evaluation of READI Chicago. Quarterly Journal of Economics, 139(1), 1-56.

Bilchitz, D. (2007). Poverty and Fundamental Rights: The Justification and Enforcement of Socio-Economic Rights. Oxford University Press.

Bjørnskov, C., & Voigt, S. (2018). The architecture of emergency constitutions. International Journal of Constitutional Law, 16(1), 101-127.

Blanchini, F. (1999). Set invariance in control. Automatica, 35(11), 1747-1767.

Blinder, A. S., & Morgan, J. (2005). Are two heads better than one? Monetary policy by committee. Journal of Money, Credit and Banking, 37(5), 789-812.

Blumer, H. (1969). Collective behavior. In A. M. Lee (Ed.), Principles of Sociology (3rd ed., pp. 65-121). Barnes & Noble. (Original work published 1951)

Boggs, C. (1977). Marxism, prefigurative communism, and the problem of workers' control. Radical America, 11(6), 99-122.

Bommasani, R., et al. (2021). On the opportunities and risks of foundation models. arXiv:2108.07258.

Bookchin, M. (1982). The Ecology of Freedom: The Emergence and Dissolution of Hierarchy. Cheshire Books.

Boyle, J. (2003). The second enclosure movement and the construction of the public domain. Law and Contemporary Problems, 66(1), 33-74.

Bradford, A. (2020). The Brussels Effect: How the European Union Rules the World. Oxford University Press.

Braga, A. A., Weisburd, D., & Turchan, B. (2018). Focused deterrence strategies and crime control: An updated systematic review and meta-analysis. Criminology & Public Policy, 17(1), 205-250.

Braithwaite, J. (1989). Crime, Shame and Reintegration. Cambridge University Press.

Brand, S. (1999). The Clock of the Long Now: Time and Responsibility. Basic Books.

Brier, G. W. (1950). Verification of forecasts expressed in terms of probability. Monthly Weather Review, 78(1), 1-3.

Brockway, P. E., et al. (2024). Energy returns on investment at the useful stage and implications for renewable energy systems. Nature Energy, 9, 803-816.

Bruneau, M., et al. (2003). A framework to quantitatively assess and enhance the seismic resilience of communities. Earthquake Spectra, 19(4), 733-752.

Buchanan, A. (2002). Political legitimacy and democracy. Ethics, 112(4), 689-719.

Bull, H. (1977). The Anarchical Society: A Study of Order in World Politics. Macmillan.

Burdín, G. (2016). Equality under threat by the talented: Evidence from worker-managed firms. The Economic Journal, 126(594), 1372-1403.

Bürgerdialog Ostbelgien. (2019). The Citizens' Council: permanent citizen participation in the German-speaking Community of Belgium.

Buterin, V. (2021). Moving beyond coin voting governance. vitalik.eth.limo.

Campbell, D. T. (1979). Assessing the impact of planned social change. Evaluation and Program Planning, 2(1), 67-90.

Carey, J. M., Niemi, R. G., Powell, L. W., & Moncrief, G. F. (2006). The effects of term limits on state legislatures: A new survey of the 50 states. Legislative Studies Quarterly, 31(1), 105-134.

Caro, R. A. (1974). The Power Broker: Robert Moses and the Fall of New York. Knopf.

Carpenter, S., Walker, B., Anderies, J. M., & Abel, N. (2001). From metaphor to measurement: Resilience of what to what? Ecosystems, 4(8), 765-781.

Carter Center. (2025). Report on the July 28, 2024 Presidential Election in Venezuela.

Case, A., & Deaton, A. (2020). Deaths of Despair and the Future of Capitalism. Princeton University Press.

Centola, D. (2018). How Behavior Spreads: The Science of Complex Contagions. Princeton University Press.

Chambers, C. D. (2013). Registered Reports: A new publishing initiative at Cortex. Cortex, 49(3), 609-610.

Chenoweth, E. (2020). Questions, answers, and some cautionary updates regarding the 3.5% rule. Carr Center Discussion Paper 2020-005. Harvard Kennedy School.

Chenoweth, E. (2020). The future of nonviolent resistance. Journal of Democracy, 31(3), 69-84.

Chenoweth, E., & Belgioioso, M. (2019). The physics of dissent and the effects of movement momentum. Nature Human Behaviour, 3, 1088-1095.

Chetty, R., Jackson, M. O., Kuchler, T., Stroebel, J., et al. (2022). Social capital I: Measurement and associations with economic mobility. Nature, 608, 108-121.

Chetty, R., Jackson, M. O., Kuchler, T., Stroebel, J., et al. (2022). Social capital II: Determinants of economic connectedness. Nature, 608, 122-134.

Cheu, A., Smith, A., & Ullman, J. (2021). Manipulation attacks in local differential privacy. In Proceedings of the IEEE Symposium on Security and Privacy.

Choi, J., Hecht, G. W., & Tayler, W. B. (2012). Lost in translation: The effects of incentive compensation on strategy surrogation. The Accounting Review, 87(4), 1135-1163.

Choudhry, N. K., Fletcher, R. H., & Soumerai, S. B. (2005). Systematic review: The relationship between clinical experience and quality of health care. Annals of Internal Medicine, 142(4), 260-273.

Chwe, M. S.-Y. (2001). Rational Ritual: Culture, Coordination, and Common Knowledge. Princeton University Press.

CLES. (2019). The Preston Model and community wealth building. Centre for Local Economic Strategies.

Cline, E. H. (2014). 1177 B.C.: The Year Civilization Collapsed. Princeton University Press.

Collingridge, D. (1980). The Social Control of Technology. St. Martin's Press.

Committee on Economic, Social and Cultural Rights. (1990). General Comment No. 3: The nature of States parties' obligations. UN Doc. E/1991/23.

Contreras, J. L. (2015). Patent pledges. Arizona State Law Journal, 47, 543-608.

Court of Justice of the European Union. (2023). SCHUFA Holding, C-634/21 (a credit score is an automated decision under GDPR Article 22).

Court of Justice of the European Union. (2025). Dun & Bradstreet Austria, C-203/22 (the right to a meaningful explanation of an automated decision).

Cukierman, A., Webb, S. B., & Neyapti, B. (1992). Measuring the independence of central banks and its effect on policy outcomes. World Bank Economic Review, 6(3), 353-398.

Cullinan, C. (2011). Wild Law: A Manifesto for Earth Justice (2nd ed.). Green Books.

Cumming, G. S., Cumming, D. H. M., & Redman, C. L. (2006). Scale mismatches in social-ecological systems. Ecology and Society, 11(1), 14.

Cutler, D. M., & Summers, L. H. (2020). The COVID-19 pandemic and the $16 trillion virus. JAMA, 324(15), 1495-1496.

Daly, H. E. (1977). Steady-State Economics. W. H. Freeman.

David, P. A. (1985). Clio and the economics of QWERTY. American Economic Review, 75(2), 332-337.

Dawkins, C. J. (2005). Tiebout choice and residential segregation by race in US metropolitan areas. Regional Science and Urban Economics, 35(6), 734-755.

de Kraker, M. E. A., Stewardson, A. J., & Harbarth, S. (2016). Will 10 million people die a year due to antimicrobial resistance by 2050? PLoS Medicine, 13(11), e1002184.

Densen, P. (2011). Challenges and opportunities facing medical education. Transactions of the American Clinical and Climatological Association, 122, 48-58.

Dinur, I., & Nissim, K. (2003). Revealing information while preserving privacy. In Proceedings of PODS 2003 (pp. 202-210). ACM.

Dobson, A. P., et al. (2020). Ecology and economics for pandemic prevention. Science, 369(6502), 379-381.

Druckman, J. N., & Nelson, K. R. (2003). Framing and deliberation: How citizens’ conversations limit elite influence. American Journal of Political Science, 47(4), 729-745.

Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22(6), 469-493.

Edgell, A. B., Lachapelle, J., Lührmann, A., & Maerz, S. F. (2021). Pandemic backsliding: Violations of democratic standards during Covid-19. Social Science and Medicine, 285, 114244.

Edwards, J., & Ogilvie, S. (2012). Contract enforcement, institutions, and social capital: The Maghribi traders reappraised. The Economic History Review, 65(2), 421-444.

Eghbal, N. (2020). Working in Public: The Making and Maintenance of Open Source Software. Stripe Press.

Ekberg, J., Eriksson, R., & Friebel, G. (2013). Parental leave: A policy evaluation of the Swedish "daddy-month" reform. Journal of Public Economics, 97, 131-143.

Elmqvist, T., Folke, C., et al. (2003). Response diversity, ecosystem change, and resilience. Frontiers in Ecology and the Environment, 1(9), 488-494.

Elster, J. (2000). Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints. Cambridge University Press.

European Court of Human Rights. (2010). Gäfgen v. Germany [GC], no. 22978/05.

European Parliament & Council. (2024). Directive (EU) 2024/1799 on common rules promoting the repair of goods.

European Parliament & Council. (2024). Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), Article 5.

Feinberg, J. (1980). The child's right to an open future. In W. Aiken & H. LaFollette (Eds.), Whose Child? Children's Rights, Parental Authority, and State Power. Rowman and Littlefield.

Ferejohn, J., & Pasquino, P. (2004). The law of the exception: A typology of emergency powers. International Journal of Constitutional Law, 2(2), 210-239.

Financial Crisis Inquiry Commission. (2011). The Financial Crisis Inquiry Report. U.S. Government Publishing Office.

Fine, P., Eames, K., & Heymann, D. L. (2011). "Herd immunity": A rough guide. Clinical Infectious Diseases, 52(7), 911-916.

Finn, J. E. (2010). Sunset clauses and democratic deliberation: Assessing the significance of sunset provisions in antiterrorism legislation. Columbia Journal of Transnational Law, 48, 442-502.

Fischer-Kowalski, M. (1998). Society's metabolism. Journal of Industrial Ecology, 2(1), 61-78.

Fishkin, J. S. (2009). When the People Speak: Deliberative Democracy and Public Consultation. Oxford University Press.

Folbre, N. (2001). The Invisible Heart: Economics and Family Values. New Press.

Folkeoplysningsloven. (1990, as consolidated). Act on support for non-formal adult education and voluntary association work (Denmark), Act no. 410 of 13 June 1990.

Foundational Economy Collective. (2018). Foundational Economy: The Infrastructure of Everyday Life. Manchester University Press.

Fraser, N. (2016). Contradictions of capital and care. New Left Review, 100, 99-117.

Freeman, J. (1972). The tyranny of structurelessness. Berkeley Journal of Sociology, 17, 151-165.

Frischmann, B. M., Madison, M. J., & Strandburg, K. J. (Eds.). (2014). Governing Knowledge Commons. Oxford University Press.

Gamalielsson, J., & Lundell, B. (2014). Sustainability of open source software communities beyond a fork: How and why has the LibreOffice project evolved? Journal of Systems and Software, 89, 128-145.

George, H. (1879). Progress and Poverty. Appleton.

Georgescu-Roegen, N. (1971). The Entropy Law and the Economic Process. Harvard University Press.

Gerlach-Kristen, P. (2009). Outsiders at the Bank of England's MPC. Journal of Money, Credit and Banking, 41(6), 1099-1115.

Gershuny, J. (2005). Busyness as the badge of honor for the new superordinate working class. Social Research, 72(2), 287-314.

Gesell, S. (1916). The Natural Economic Order. (Trans. 1958, Peter Owen.)

Giné, X., Karlan, D., & Zinman, J. (2010). Put your money where your butt is: A commitment savings account for smoking cessation. American Economic Journal: Applied Economics, 2(4), 213-235.

Ginsburg, T., & Versteeg, M. (2021). The bound executive: Emergency powers during the pandemic. International Journal of Constitutional Law, 19(5), 1498-1535.

Ginsburg, T., & Versteeg, M. (2021). The bound executive: Emergency powers during the pandemic. International Journal of Constitutional Law, 19(5), 1498-1535.

Goodhart, C. A. E. (1975). Problems of monetary management: The U.K. experience. In Papers in Monetary Economics (Vol. 1). Reserve Bank of Australia.

Goodin, R. E. (2007). Enfranchising all affected interests, and its alternatives. Philosophy and Public Affairs, 35(1), 40-68.

Grabowski, D. C., Campbell, C. M., & Morrisey, M. A. (2004). Elderly licensure laws and motor vehicle fatalities. JAMA, 291(23), 2840-2846.

Granovetter, M. (1978). Threshold models of collective behavior. American Journal of Sociology, 83(6), 1420-1443.

Greif, A. (1993). Contract enforceability and economic institutions in early trade: The Maghribi traders' coalition. American Economic Review, 83(3), 525-548.

Greif, A., Milgrom, P., & Weingast, B. R. (1994). Coordination, commitment, and enforcement: The case of the merchant guild. Journal of Political Economy, 102(4), 745-776.

Grimm, V., et al. (2006). A standard protocol for describing individual-based and agent-based models. Ecological Modelling, 198(1-2), 115-126.

Grounded Solutions Network. (2022). Community Land Trust and Shared Equity Homeownership Census.

Guinan, J., & O'Neill, M. (2019). The Case for Community Wealth Building. Polity Press.

Gunderson, L. H., & Holling, C. S. (Eds.). (2002). Panarchy: Understanding Transformations in Human and Natural Systems. Island Press.

Habermas, J. (1984). The Theory of Communicative Action, Volume 1 (T. McCarthy, Trans.). Beacon Press.

Haidt, J. (2024). The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press.

Hall, C. A. S., Lambert, J. G., & Balogh, S. B. (2014). EROI of different fuels and the implications for society. Energy Policy, 64, 141-152.

Halfaker, A., Geiger, R. S., Morgan, J. T., & Riedl, J. (2013). The rise and decline of an open collaboration system. American Behavioral Scientist, 57(5), 664-688.

Hannan, M. T., & Freeman, J. (1977). The population ecology of organizations. American Journal of Sociology, 82(5), 929-964.

Hart, H. L. A. (1955). Are there any natural rights? Philosophical Review, 64(2), 175-191.

Hausfather, Z. (2021). Absolute decoupling of economic growth and emissions in 32 countries. Breakthrough Institute.

Herz, J. H. (1950). Idealist internationalism and the security dilemma. World Politics, 2(2), 157-180.

Hess, C., & Ostrom, E. (Eds.). (2007). Understanding Knowledge as a Commons: From Theory to Practice. MIT Press.

Hickel, J. (2020). Less Is More: How Degrowth Will Save the World. William Heinemann.

Hickel, J., & Kallis, G. (2020). Is green growth possible? New Political Economy, 25(4), 469-486.

Higgs, R. (1987). Crisis and Leviathan. Oxford University Press.

Hildebrandt, M. (2015). Smart Technologies and the End(s) of Law. Edward Elgar.

Hirschman, A. O. (1970). Exit, Voice, and Loyalty. Harvard University Press.

Hoffer, E. (1951). The True Believer: Thoughts on the Nature of Mass Movements. Harper & Brothers.

Holling, C. S. (1973). Resilience and stability of ecological systems. Annual Review of Ecology and Systematics, 4, 1-23.

Holmes, S. (1995). Passions and Constraint: On the Theory of Liberal Democracy. University of Chicago Press.

Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316.

Hufbauer, G. C., Schott, J. J., Elliott, K. A., & Oegg, B. (2007). Economic Sanctions Reconsidered (3rd ed.). Peterson Institute for International Economics.

Hume, D. (1748). Of the original contract. In Essays, Moral and Political.

Hurwicz, L. (2008). But who will guard the guardians? (Nobel Prize lecture). American Economic Review, 98(3), 577-585.

Hutchins, E. M., Cloppert, M. J., & Amin, R. M. (2011). Intelligence-driven computer network defense informed by analysis of adversary campaigns and intrusion kill chains. In Proceedings of the 6th International Conference on Information Warfare and Security. Lockheed Martin Corporation.

Intergovernmental Panel on Climate Change. (2024). How Does the IPCC Review Process Work? (fact sheet).

International Institute for Democracy and Electoral Assistance. (2014). Electoral Management Design (rev. ed.). International IDEA.

International Covenant on Civil and Political Rights, Article 4. (1966). United Nations.

International Labour Organization. (2018). Care Work and Care Jobs for the Future of Decent Work. International Labour Office.

Internet Architecture Board. (2000). IAB technical comment on the unique DNS root (RFC 2826). Internet Engineering Task Force.

Internet Engineering Task Force. (2015). Guidelines for cryptographic algorithm agility (RFC 7696, BCP 201).

Iosifidis, G., Charette, Y., Airoldi, E. M., Littera, G., Tassiulas, L., & Christakis, N. A. (2018). Cyclic motifs in the Sardex monetary network. Nature Human Behaviour, 2(11), 822-829.

Janssen, M. A., & Ostrom, E. (2006). Empirically based, agent-based models. Ecology and Society, 11(2), 37.

Jervis, R. (1978). Cooperation under the security dilemma. World Politics, 30(2), 167-214.

Jäger, G., & Füllsack, M. (2019). Systematically false positives in early-warning-signal analysis. PLOS ONE, 14(2), e0211072.

Jost, L. (2006). Entropy and diversity. Oikos, 113(2), 363-375.

Judicial Commission of Inquiry into State Capture. (2022). Report of the Judicial Commission of Inquiry into Allegations of State Capture, Corruption and Fraud in the Public Sector. Republic of South Africa.

Junusova, M., & Reinsch, W. A. (2024). Rethinking the Wassenaar minus one strategy. Center for Strategic and International Studies.

Kadivar, M. A. (2018). Mass mobilization and the durability of new democracies. American Sociological Review, 83(2), 390-417.

Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515-526.

Kalecki, M. (1943). Political aspects of full employment. The Political Quarterly, 14(4), 322-331.

Kallis, G. (2018). Degrowth. Agenda Publishing.

Kant, I. (1998). Groundwork of the Metaphysics of Morals (M. Gregor, Trans.). Cambridge University Press. (Original work published 1785)

Kasser, T. (2002). The High Price of Materialism. MIT Press.

Kauffman, S. A. (1993). The Origins of Order: Self-Organization and Selection in Evolution. Oxford University Press.

Keohane, R. O., & Victor, D. G. (2011). The regime complex for climate change. Perspectives on Politics, 9(1), 7-23.

Kerckhoffs, A. (1883). La cryptographie militaire. Journal des Sciences Militaires, IX, 5-83.

Keynes, J. M. (1930). Economic possibilities for our grandchildren. In Essays in Persuasion.

Khan, L. M. (2017). Amazon's antitrust paradox. Yale Law Journal, 126(3), 710-805.

King, G., Pan, J., & Roberts, M. E. (2017). How the Chinese government fabricates social media posts for strategic distraction, not engaged argument. American Political Science Review, 111(3), 484-501.

Kleinberg, J., & Raghavan, M. (2021). Algorithmic monoculture and social welfare. PNAS, 118(22), e2018340118.

Klinenberg, E. (2018). Palaces for the People: How Social Infrastructure Can Help Fight Inequality, Polarization, and the Decline of Civic Life. Crown.

Klosko, G. (1992). The Principle of Fairness and Political Obligation. Rowman and Littlefield.

Kopp, R. E., et al. (2025). "Tipping points" confuse and can distract from urgent climate action. Nature Climate Change, 15, 29-36.

Kotsadam, A., & Finseraas, H. (2011). The state intervenes in the battle of the sexes: Causal effects of paternity leave. Social Science Research, 40(6), 1611-1622.

Kousser, T. (2005). Term Limits and the Dismantling of State Legislative Professionalism. Cambridge University Press.

Kremer, M. (1998). Patent buyouts: A mechanism for encouraging innovation. Quarterly Journal of Economics, 113(4), 1137-1167.

Kretschmer, K. (2019). Fighting for NOW: Diversity and Discord in the National Organization for Women. University of Minnesota Press.

Krznaric, R. (2020). The Good Ancestor: A Radical Prescription for Long-Term Thinking. The Experiment.

Kulveit, J., Douglas, R., Ammann, N., Turan, D., Krueger, D., & Duvenaud, D. (2025). Gradual disempowerment: Systemic existential risks from incremental AI development. arXiv:2501.16946.

Kuran, T. (1995). Private Truths, Public Lies: The Social Consequences of Preference Falsification. Harvard University Press.

Laebens, M. G., & Lührmann, A. (2021). What halts democratic erosion? The changing role of accountability. Democratization, 28(5), 908-928.

Lakatos, I. (1978). The Methodology of Scientific Research Programmes: Philosophical Papers Volume 1 (J. Worrall & G. Currie, Eds.). Cambridge University Press.

Lalich, J. (2004). Bounded Choice: True Believers and Charismatic Cults. University of California Press.

Larivière, V., Haustein, S., & Mongeon, P. (2015). The oligopoly of academic publishers in the digital era. PLOS ONE, 10(6), e0127502.

Larraín, F., & Meller, P. (1991). The socialist-populist Chilean experience, 1970-1973. In R. Dornbusch & S. Edwards (Eds.), The Macroeconomics of Populism in Latin America. University of Chicago Press.

Lauenroth, K. (2025). The critical nature of sunset provisions in national security legislation. Georgetown Law Journal, 113(4).

Lenton, T. M., et al. (2008). Tipping elements in the Earth's climate system. PNAS, 105(6), 1786-1793.

Levitsky, S., & Ziblatt, D. (2018). How Democracies Die. Crown.

Liebowitz, S. J., & Margolis, S. E. (1995). Path dependence, lock-in, and history. Journal of Law, Economics, & Organization, 11(1), 205-226.

Lietaer, B., Arnsperger, C., Goerner, S., & Brunnhuber, S. (2012). Money and Sustainability: The Missing Link. Triarchy Press.

Lindblom, C. E. (1982). The market as prison. The Journal of Politics, 44(2), 324-336.

Lindhqvist, T. (2000). Extended Producer Responsibility in Cleaner Production. Lund University.

Longpre, S., et al. (2024). Consent in crisis: The rapid decline of the AI data commons. NeurIPS Datasets and Benchmarks. arXiv:2407.14933.

Lordan-Perret, R., Sloan, R. D., & Rosner, R. (2021). Decommissioning the U.S. nuclear fleet: Financial assurance, corporate structures, and bankruptcy. Energy Policy, 154, 112280.

Lührmann, A., & Lindberg, S. I. (2019). A third wave of autocratization is here: What is new about it? Democratization, 26(7), 1095-1113.

Lührmann, A., & Rooney, B. (2021). Autocratization by decree: States of emergency and democratic decline. Comparative Politics, 53(4), 617-649.

Machlup, F. (1962). The Production and Distribution of Knowledge in the United States. Princeton University Press.

Macklin, R. (2003). Dignity is a useless concept. BMJ, 327, 1419-1420.

Madison, J. (1787). Federalist No. 10. In The Federalist Papers.

Marmot, M. G., Shipley, M. J., & Rose, G. (1984). Inequalities in death: Specific explanations of a general pattern? The Lancet, 323(8384), 1003-1006.

Masnick, M. (2019). Protocols, Not Platforms: A Technological Approach to Free Speech. Knight First Amendment Institute, Columbia University.

Mateescu, A. (2021). Electronic Visit Verification: The Weight of Surveillance and the Fracturing of Care. Data & Society Research Institute.

Mayer-Schönberger, V. (2009). Delete: The Virtue of Forgetting in the Digital Age. Princeton University Press.

Mazzucato, M. (2013). The Entrepreneurial State: Debunking Public vs. Private Sector Myths. Anthem Press.

McAdam, D. (1982). Political Process and the Development of Black Insurgency, 1930-1970. University of Chicago Press.

McCarthy, J. D., & Zald, M. N. (1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82(6), 1212-1241.

McCrudden, C. (2008). Human dignity and judicial interpretation of human rights. European Journal of International Law, 19(4), 655-724.

McGuire, W. J. (1964). Inducing resistance to persuasion. Advances in Experimental Social Psychology, 1, 191-229.

McLaren, R. H. (2016). The Independent Person Report (Parts I and II). World Anti-Doping Agency.

McSherry, F., & Talwar, K. (2007). Mechanism design via differential privacy. In Proceedings of FOCS 2007 (pp. 94-103). IEEE.

Meadows, D. H. (2008). Thinking in Systems: A Primer. Chelsea Green Publishing.

Meta. (2022). Recapping our 2022 coordinated inauthentic behavior enforcements.

Michels, R. (1911). Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy. Hearst’s International Library.

Michels, R. (1915). Political Parties (E. Paul & C. Paul, Trans.). Hearst's International Library.

Milkoreit, M. (2023). Social tipping points everywhere? Patterns and risks of overuse. WIREs Climate Change, 14(2), e813.

MITRE Corporation. (2026). MITRE ATT&CK: A knowledge base of adversary tactics and techniques from real-world observations.

Möller, B. (1992). Common Security and Nonoffensive Defense: A Neorealist Perspective. Lynne Rienner.

Molloy, R., Smith, C. L., & Wozniak, A. (2011). Internal migration in the United States. Journal of Economic Perspectives, 25(3), 173-196.

Moncrief, G., & Thompson, J. A. (2001). On the outside looking in: Lobbyists' perspectives on the effects of state legislative term limits. State Politics and Policy Quarterly, 1(4), 394-411.

Mondragon Corporation. (2025). 2024 annual results: 81 cooperatives, 70,085 workers, 11.2 billion euros in sales.

Moore, G. A. (1991). Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Customers. HarperBusiness.

Morgan, T. C., Syropoulos, C., & Yotov, Y. V. (2023). Economic sanctions: Evolution, consequences, and challenges. Journal of Economic Perspectives, 37(1), 3-29.

Motolinía, L. (2021). Electoral accountability and particularistic legislation: Evidence from an electoral reform in Mexico. American Political Science Review, 115(1), 97-113.

Mouffe, C. (2000). The Democratic Paradox. Verso.

Mouffe, C. (2005). On the Political. Routledge.

Naím, M. (2009). Minilateralism: The magic number to get real international action. Foreign Policy, 173, 135-136.

National Academies of Sciences, Engineering, and Medicine. (2024). Social Media and Adolescent Health. National Academies Press.

National Institute of Standards and Technology. (2024). FIPS 203, 204, 205: Post-quantum cryptography standards. U.S. Department of Commerce.

National Institute of Standards and Technology. (2025). Considerations for achieving cryptographic agility (NIST CSWP 39).

New Zealand Local Government Act 2002, s 101B (inserted 2014). Infrastructure strategies. New Zealand Legislation.

Nissenbaum, H. (2010). Privacy in Context: Technology, Policy, and the Integrity of Social Life. Stanford University Press.

Nowak, M. A. (2006). Five rules for the evolution of cooperation. Science, 314(5805), 1560-1563.

Nussbaum, M. C. (2000). Women and Human Development: The Capabilities Approach. Cambridge University Press.

Obar, J. A., & Oeldorf-Hirsch, A. (2020). The biggest lie on the internet: Ignoring the privacy policies and terms of service policies of social networking services. Information, Communication and Society, 23(1), 128-147.

OECD. (2025). Pensions at a Glance 2025. OECD Publishing.

Schattschneider, E. E. (1960). The Semisovereign People: A Realist’s View of Democracy in America. Holt, Rinehart and Winston.

United Nations, Department of Economic and Social Affairs. (2000). Replacement Migration: Is It a Solution to Declining and Ageing Populations? United Nations.

United Nations, Department of Economic and Social Affairs. (2024). World Population Prospects 2024. United Nations.

Wright, J. (2023). Robots Won’t Save Japan: An Ethnography of Eldercare Automation. Cornell University Press.

Öcalan, A. (2011). Democratic Confederalism. Transmedia Publishing.

Odgers, C. L. (2024). The great rewiring: Is social media really behind an epidemic of teenage mental illness? Nature, 628, 29-30.

OECD. (2020). Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave. OECD Publishing.

Ohtsuki, H., Hauert, C., Lieberman, E., & Nowak, M. A. (2006). A simple rule for the evolution of cooperation on graphs and social networks. Nature, 441(7092), 502-505.

Okin, S. M. (2002). "Mistresses of their own destiny": Group rights, gender, and realistic rights of exit. Ethics, 112(2), 205-230.

Oldenburg, R. (1989). The Great Good Place. Paragon House.

O'Mahony, S., & Ferraro, F. (2007). The emergence of governance in an open source community. Academy of Management Journal, 50(5), 1079-1106.

O'Neill, J. (2016). Tackling Drug-Resistant Infections Globally: Final Report and Recommendations. Review on Antimicrobial Resistance.

Orben, A., & Przybylski, A. K. (2019). The association between adolescent well-being and digital technology use. Nature Human Behaviour, 3, 173-182.

Organisation for the Prohibition of Chemical Weapons. (2018-2025). The Investigation and Identification Team; decisions on the suspension and restoration of rights. OPCW Conference of the States Parties.

Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press.

Ostrom, E. (2010). Polycentric systems for coping with collective action and global environmental change. Global Environmental Change, 20(4), 550-557.

Ostrom, V., Tiebout, C. M., & Warren, R. (1961). The organization of government in metropolitan areas. American Political Science Review, 55(4), 831-842.

Pareto, V. (1935). The Mind and Society (A. Bongiorno & A. Livingston, Trans.). Harcourt, Brace. (Original work published 1916)

Patnaik, A. (2019). Reserving time for daddy: The consequences of fathers' quotas. Journal of Labor Economics, 37(4), 1009-1059.

Pecchi, L., & Piga, G. (Eds.). (2008). Revisiting Keynes: Economic Possibilities for Our Grandchildren. MIT Press.

Pei, S., Kandula, S., & Shaman, J. (2020). Differential effects of intervention timing on COVID-19 spread in the United States. Science Advances, 6(49), eabd6370.

Pettit, P. (1997). Republicanism: A Theory of Freedom and Government. Oxford University Press.

Pettit, P. (2010). A republican law of peoples. European Journal of Political Theory, 9(1), 70-94.

Polanyi, M. (1966). The Tacit Dimension. University of Chicago Press.

Pollin, R. (2018). De-growth vs a Green New Deal. New Left Review, 112, 5-25.

Pollitt, C. (2009). Bureaucracies remember, post-bureaucratic organizations forget? Public Administration, 87(2), 198-218.

Popper, K. R. (1963). Conjectures and Refutations: The Growth of Scientific Knowledge. Routledge & Kegan Paul.

Posner, E. A., & Weyl, E. G. (2018). Radical Markets. Princeton University Press.

Przeworski, A., & Wallerstein, M. (1988). Structural dependence of the state on capital. American Political Science Review, 82(1), 11-29.

Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster.

Radin, M. J. (2013). Boilerplate: The Fine Print, Vanishing Rights, and the Rule of Law. Princeton University Press.

Rawls, J. (1971). A Theory of Justice. Belknap Press of Harvard University Press.

Raworth, K. (2017). Doughnut Economics: Seven Ways to Think Like a 21st-Century Economist. Chelsea Green Publishing.

Resnick, P., Zeckhauser, R., Swanson, J., & Lockwood, K. (2006). The value of reputation on eBay: A controlled experiment. Experimental Economics, 9(2), 79-101.

Richardson, K., et al. (2023). Earth beyond six of nine planetary boundaries. Science Advances, 9(37), eadh2458.

Richman, B. D. (2017). An autopsy of cooperation: Diamond dealers and the limits of trust-based exchange. Journal of Legal Analysis, 9(2), 247-283.

Riksrevisionen. (2022). Statsbidraget till studieförbunden: kontroll och uppföljning (RiR 2022:20). Swedish National Audit Office.

Ritchie, H. (2021). Many countries have decoupled economic growth from CO2 emissions, even accounting for offshored production. Our World in Data.

Rockström, J., et al. (2009). A safe operating space for humanity. Nature, 461(7263), 472-475.

Rodríguez-Garavito, C., & De Bona, G. (2024). Are rights of nature working? The impact of the Los Cedros ruling. Verfassungsblog.

Rogers, E. M. (2003). Diffusion of Innovations (5th ed.). Free Press.

Roozenbeek, J., & van der Linden, S. (2019). Fake news game confers psychological resistance against online misinformation. Palgrave Communications, 5, 65.

Rosa, H. (2013). Social Acceleration: A New Theory of Modernity (J. Trejo-Mathys, Trans.). Columbia University Press.

Rosa, H. (2019). Resonance: A Sociology of Our Relationship to the World (J. C. Wagner, Trans.). Polity Press.

Rose, T. C., et al. (2023). Analysis of the health and wellbeing impacts of a community wealth building programme in England: A difference-in-differences study. The Lancet Public Health, 8(6), e403-e410.

Russell, A. L., & Vinsel, L. (2016). Hail the maintainers. Aeon.

Sabel, C. F., & Zeitlin, J. (Eds.). (2010). Experimentalist Governance in the European Union: Towards a New Architecture. Oxford University Press.

Salvi, F., Horta Ribeiro, M., Gallotti, R., & West, R. (2025). On the conversational persuasiveness of GPT-4 with personalization. Nature Human Behaviour, 9, 1645-1653.

Sandel, M. J. (2020). The Tyranny of Merit: What's Become of the Common Good? Farrar, Straus and Giroux.

Scheel, A. M., Schijen, M. R. M. J., & Lakens, D. (2021). An excess of positive results: Comparing the standard psychology literature with registered reports. Advances in Methods and Practices in Psychological Science, 4(2).

Scheffer, M., et al. (2009). Early-warning signals for critical transitions. Nature, 461(7260), 53-59.

Scheppele, K. L. (2018). Autocratic legalism. University of Chicago Law Review, 85(2), 545-583.

Schmitt, C. (2005). Political Theology: Four Chapters on the Concept of Sovereignty (G. Schwab, Trans.). University of Chicago Press. (Original work published 1922)

Schumpeter, J. A. (1942). Capitalism, Socialism and Democracy. Harper & Brothers.

Scott, J. C. (1998). Seeing Like a State. Yale University Press.

Sen, A. (1999). Development as Freedom. Alfred A. Knopf.

Sen, A. (2002). Why health equity? Health Economics, 11(8), 659-666.

Setälä, M., Grönlund, K., & Herne, K. (2010). Citizen deliberation on nuclear power: A comparison of two decision-making methods. Political Studies, 58(4), 688-714.

Shem-Tov, Y., Raphael, S., & Skog, A. (2024). Can restorative justice conferencing reduce recidivism? Evidence from the Make-it-Right program. Econometrica, 92(1), 61-92.

Sigmund, P. E. (1974). The "invisible blockade" and the overthrow of Allende. Foreign Affairs, 52(2), 322-340.

Simmons, A. J. (1979). Moral Principles and Political Obligations. Princeton University Press.

Simon, H. A. (1971). Designing organizations for an information-rich world. In M. Greenberger (Ed.), Computers, Communications, and the Public Interest (pp. 37-72). Johns Hopkins Press.

Smil, V. (2022). How the World Really Works. Viking.

Soderberg, C. K., Errington, T. M., Schiavone, S. R., et al. (2021). Initial evidence of research quality of registered reports compared with the standard publishing model. Nature Human Behaviour, 5, 990-997.

Sterling, P. (2012). Allostasis: A model of predictive regulation. Physiology & Behavior, 106(1), 5-15.

Sterman, J. D. (2000). Business Dynamics: Systems Thinking and Modeling for a Complex World. Irwin/McGraw-Hill.

Stigler, G. J. (1971). The theory of economic regulation. The Bell Journal of Economics and Management Science, 2(1), 3-21.

Stodder, J. (2009). Complementary credit networks and macroeconomic stability: Switzerland's Wirtschaftsring. Journal of Economic Behavior & Organization, 72(1), 79-95.

Stodder, J., & Lietaer, B. (2016). The macro-stability of Swiss WIR-Bank credits: Balance, velocity, and leverage. Comparative Economic Studies, 58(4), 570-605.

Stone, C. D. (1972). Should trees have standing? Toward legal rights for natural objects. Southern California Law Review, 45, 450-501.

Strang, H., & Sherman, L. W. (2013). Restorative justice conferencing using face-to-face meetings of offenders and victims. Campbell Systematic Reviews, 9(1), 1-59.

Strathern, M. (1997). 'Improving ratings': Audit in the British university system. European Review, 5(3), 305-321.

Subrahmanyam, A. (1994). Circuit breakers and market volatility: A theoretical perspective. Journal of Finance, 49(1), 237-254.

Sunstein, C. R. (2006). Irreversible and catastrophic. Cornell Law Review, 91, 841-897.

Supreme Court of Alaska. (2017). Wielechowski v. State, 403 P.3d 1141 (Permanent Fund dividend transfers are ordinary appropriations subject to gubernatorial veto).

Supreme Court of Canada. (2008). R. v. Kapp, 2008 SCC 41.

Sweeney, L. (2000). Simple demographics often identify people uniquely (Data Privacy Working Paper 3). Carnegie Mellon University.

Tainter, J. A. (1988). The Collapse of Complex Societies. Cambridge University Press.

Taleb, N. N. (2012). Antifragile: Things That Gain from Disorder. Random House.

Tang, J., et al. (2017). Privacy loss in Apple's implementation of differential privacy on macOS 10.12. arXiv:1709.02753.

Tetlock, P. E., Mellers, B. A., Rohrbaugh, N., & Chen, E. (2014). Forecasting tournaments: Tools for increasing transparency and improving the quality of debate. Current Directions in Psychological Science, 23(4), 290-295.

Tiebout, C. M. (1956). A pure theory of local expenditures. Journal of Political Economy, 64(5), 416-424.

Transportation for America & Taxpayers for Common Sense. (2026). Repair Priorities 2026.

Trask, A., et al. (2020). Beyond privacy trade-offs with structured transparency. arXiv:2012.08347.

Tronto, J. C. (2013). Caring Democracy: Markets, Equality, and Justice. New York University Press.

Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest. Yale University Press.

Tyler, T. R. (1990). Why People Obey the Law. Yale University Press.

Ulanowicz, R. E., Goerner, S. J., Lietaer, B., & Gomez, R. (2009). Quantifying sustainability: Resilience, efficiency and the return of information theory. Ecological Complexity, 6(1), 27-36.

Underhill, K., & Ayres, I. (2022). Sunsets are for suckers: An experimental test of sunset clauses. Harvard Journal on Legislation, 59(1), 103-158.

United Nations. (1992). Rio Declaration on Environment and Development, Principle 15.

United Nations, European Commission, International Monetary Fund, OECD, & World Bank. (2025). System of National Accounts 2025. United Nations.

United Nations Environment Programme, International Resource Panel. (2024). Global Resources Outlook 2024.

United Nations Environment Programme. (2025). Emissions Gap Report 2025: Off Target. UNEP.

United Nations Human Rights Committee. (2002). Wackenheim v. France, Communication No. 854/1999.

U.S. House Committee on Transportation and Infrastructure. (2020). Final Committee Report: The Design, Development, and Certification of the Boeing 737 MAX.

van der Linden, S. (2023). Foolproof: Why Misinformation Infects Our Minds and How to Build Immunity. W. W. Norton.

van Doorn, N. (2017). Platform cooperativism and the problem of the outside. Culture Digitally.

van Ham, C., & Garnett, H. A. (2019). Building impartial electoral management? International Political Science Review, 40(3), 313-334.

V-Dem Institute. (2026). Democracy Report 2026. Varieties of Democracy Institute, University of Gothenburg.

Venice Commission. (2020). Compilation of Opinions and Reports on States of Emergency (CDL-PI(2020)003). Council of Europe.

Vervaeke, J. (2019-2024). Awakening from the Meaning Crisis. Lecture series and book (Part One: Origins, 2024).

Vieta, M., Depedri, S., & Carrano, A. (2017). The Italian Road to Recuperating Enterprises and the Legge Marcora Framework. Euricse.

Vinsel, L., & Russell, A. L. (2020). The Innovation Delusion: How Our Obsession with the New Has Disrupted the Work That Matters Most. Currency.

Vogel, J., & Hickel, J. (2023). Is green growth happening? An empirical analysis of achieved versus Paris-compliant CO2-GDP decoupling in high-income countries. The Lancet Planetary Health, 7(9), e759-e769.

Voigt, S., Gutmann, J., & Feld, L. P. (2015). Economic growth and judicial independence, a dozen years on. European Journal of Political Economy, 38, 197-211.

Voigt, S., Bjørnskov, C., & Khesali, M. (2022). Unconstitutional states of emergency. Journal of Legal Studies, 51(2), 455-481.

W3C. (2025). Verifiable Credentials Data Model 2.0. World Wide Web Consortium Recommendation.

Wajcman, J. (2015). Pressed for Time: The Acceleration of Life in Digital Capitalism. University of Chicago Press.

Waldron, J. (2012). Dignity, Rank, and Rights. Oxford University Press.

Walker, B. H., Carpenter, S. R., Rockström, J., Crépin, A.-S., & Peterson, G. D. (2012). Drivers, slow variables, fast variables, shocks, and resilience. Ecology and Society, 17(3), 30.

Walker, B., Holling, C. S., Carpenter, S. R., & Kinzig, A. (2004). Resilience, adaptability and transformability in social-ecological systems. Ecology and Society, 9(2), 5.

Walters, C. (1997). Challenges in adaptive management of riparian and coastal ecosystems. Conservation Ecology, 1(2), 1.

Waring, M. (1988). If Women Counted: A New Feminist Economics. Harper & Row.

Warlouzet, L. (2018). Le spectre de la crise financière française de 1983. Vingtième Siècle. Revue d'histoire, 138, 93-108.

Weber, M. (1978). Economy and Society (G. Roth & C. Wittich, Eds.). University of California Press. (Original work published 1922)

Weitzman, M. L. (2009). On modeling and interpreting the economics of catastrophic climate change. Review of Economics and Statistics, 91(1), 1-19.

Whyte, W. F., & Whyte, K. K. (1991). Making Mondragon: The Growth and Dynamics of the Worker Cooperative Complex (2nd ed.). ILR Press.

Wiener, N. (1948). Cybernetics: Or Control and Communication in the Animal and the Machine. The Technology Press.

Woods, D. D. (2018). The theory of graceful extensibility: Basic rules that govern adaptive systems. Environment Systems and Decisions, 38(4), 433-457.

World Health Organization, FAO, WOAH, & UNEP. (2021). Tripartite and UNEP support OHHLEP's definition of One Health.

Wright, E. O. (2010). Envisioning Real Utopias. Verso.

Wu, T. (2016). The Attention Merchants: The Epic Scramble to Get Inside Our Heads. Alfred A. Knopf.

Young, K. G. (2008). The minimum core of economic and social rights: A concept in search of content. Yale Journal of International Law, 33, 113-175.

Young, K., Banerjee, T., & Schwartz, M. (2018). Capital strikes as a corporate political strategy: The structural power of business in the Obama era. Politics & Society, 46(1), 3-28.

Zuboff, S. (2019). The Age of Surveillance Capitalism. PublicAffairs.