Most legal corpora are graveyards that still legislate: rules nobody would re-pass, enforced because repeal requires the same heroic effort as passage and nobody's paid to bother. This design inverts the default. Every policy carries its own review schedule from birth, every review asks the public one question — does this remain? — and the economics of chapter 7 guarantee that someone is always paid to argue the affirmative. Institutional lifecycle management, made concrete: what is born must keep earning its place, and what stops earning it ends on schedule rather than by revolution.
Two images from the research set the stakes — the first about what an institution silently spends, the second about what it can silently become. The coordination substrate — the accumulated trust and shared capacity that lets a people act as one — is spendable, and its balance is invisible until drawn on: Rome after Cannae, its worst defeat in the war with Hannibal, lost a fifth of its adult male citizens and raised new armies; Rome six centuries later, incomparably richer, could not field a legion against Alaric, the Goth who sacked the city in 410. The resources were there — the substrate had been spent. Golden ages can be manic fire sales of accumulated civilizational capital, and a system without scheduled renewal only discovers the account is empty on the day it draws.
In chemistry the same molecules can crystallize into more than one form, and a drug in a different polymorph is a different drug. In 1998 a more stable crystal form of Ritonavir, an HIV drug, appeared spontaneously in the production line; the existing capsules stopped dissolving properly and the formulation had to be rebuilt. Nothing about the molecule had changed — only how it packed.
Institutions do this too. Same people, same charter, different packing: an organization can recrystallize into its extractive form without a single rule being rewritten. Which is why reviews exist to test the crystal, not the charter.
After its maturation window — the protected stretch after passage when a new policy cannot be challenged — a policy faces public reconfirmation at the broadest scale it binds, on a decreasing tail — roughly years one, three, seven, fifteen. Young laws are reviewed often, because that is when the surprises arrive; laws that keep passing review are reviewed rarely, because a policy that has survived four confirmations has demonstrated fit. The Lindy intuition — that what has already lasted is the best bet to last — mechanized. The arithmetic is what makes review of everything affordable: the burden per law is logarithmic, not linear, so a full legal corpus stays reviewable without drowning the electorate — ballot fatigue is a design failure, not a law of nature. Between reviews, policies retain by default; a policy that fails or is substantially modified resets to the fast end of the cadence and starts proving itself again.
The deeper move is the inversion of the burden of proof. Today an institution persists unless someone assembles the heroic coalition needed to kill it; here, persistence requires active maintenance. The question shifts from "why should this end?" to "why should this continue?" — asked on a schedule, answered by the people living under the answer, with silence counting as abstention rather than consent.
Backoff assumes the world holds still between reviews; it doesn't. So the schedule is a floor, not a cage. Once past maturation, any policy is challengeable at any time — citizens spending points, or delegates spending weight, past a modest threshold triggers an off-schedule review, the same structure as the Swiss facultative referendum, where enough signatures force a passed law back to a public vote. Leaders can likewise spend proposal budget to put repeal on the docket. "Settled" here means no scheduled poll; it never means immutable. That distinction is what prevents entrenchment-by-calendar — and its inverse, a two-week media storm, is absorbed by the window mechanics rather than toppling policy on a news cycle.
The same lifecycle discipline binds the people. A leader's tenure rests on continuously maintained public legitimacy, checked at scheduled gates with two distinct trip-wires: fall below the first threshold and you may finish your term but cannot stand again — the graceful exit for the merely mediocre, removing zombie leadership without the instability of a mid-term decapitation. Fall below the second, deeper threshold and removal is immediate — the genuine loss-of-mandate case, without impeachment's friction or its partisanship. Note what this replaces: arbitrary term limits. Term limits fire your best performers and your worst on the same clock; legitimacy-based tenure fires precisely the ones the public has stopped consenting to, and only those. And the bar is office-specific: a leader gated out of leadership for poor performance remains eligible as a delegate. Either way, the end of service is a return, not an exile — a departing leader rejoins the delegate pool, and whatever weight they carry next is re-earned, never carried over. Poor leadership costs you the office; only your delegators can cost you the room.
One principle governs all of these ending-mechanisms, and it is easy to state and hard to honor: whoever defines "the public has withdrawn consent" is the true sovereign. So the levers must be maximally dumb and procedural — simple thresholds, bright lines, nothing a clever lawyer or a captured court can reinterpret. If people are very unhappy, you cannot run again has the correct shape: no judgment calls, no standing committee of interpretation, nothing to capture. This is the promise of populism, made coherent — throw the bums out is the one thing populist movements have always been right about, and here it is a standing bright-line procedure rather than a strongman claiming to embody the people's will. The energy that otherwise elects demagogues gets a legitimate instrument instead. And the failure asymmetry is deliberate: drift is recoverable, takeover usually is not. Where this system must err, it errs toward friction — while instrumenting the friction well enough that drift can be told apart from health.