Designing a board that cannot be captured

A generation ago we solved the problem of the human heart wearing out. Now the harder engineering problem is a committee that will not.

The four-page charter the Charter Court occupiers ratified by show of hands asks for something that sounds modest and is not: an independent board empowered to review tenure in posts where anti-aging therapy has frozen turnover. The mechanics are borrowed, deliberate, and — I say this admiringly — paranoid. Nomination by the Assembly of Signatories. Confirmation not by vote but by lot, drawn from a standing pool. Terms of six years, staggered, with no renewal. Appointments split across bodies that cordially distrust one another, so no single institution owns a majority. You could mistake it for overengineering until you remember what it is trying to outlive.

"You are not designing a board," said Teodora Miksa, a governance designer the occupation has consulted, over tea in a Commons annex three floors below the sit-in. "You are designing against a failure mode. The failure mode has a name and it is capture." She has the flat affect of someone who has watched clean ideas die in committee. "Any panel with the power to end an incumbent's tenure becomes the single most valuable seat in the building. The people it judges will spend everything they have to sit on it, staff it, or fund it. So you build the thing assuming they will try, because they will."

The worry is not theoretical. The existing renewal-value board is exhibit A: seven of its eleven members have already received the tier-one therapy their own metric is meant to ration. The chart tells a different story than the press release. It usually does. A body meant to decide who ages first was staffed, quietly, by people who arranged not to.

The reactor-commons trick

The most interesting borrowing in the draft comes from an unglamorous place: the safety panels that oversee the reactor commons. There, a member who serves on a review panel is barred from the operational roles they oversaw for a decade after their term ends. You cannot cash in your oversight for a job at the thing you were watching. The cooling-off period outlasts most careers.

"It works in fusion because the panel has no upside," said Emil Sarraf, who spent years on one of those commons panels before rotating out. "You cannot be rewarded by the operator, so you stop trying to please the operator. Apply that to judges and the logic holds. A renewal-board seat should be a dead end, professionally. If it's a stepping stone, it's already compromised."

The lottery does similar work from the other direction. Miksa walked me through it with a surgeon's patience for a nervous patient. Nomination lets the Assembly signal competence and legitimacy; a large qualified pool passes that filter. Then the lot decides, which means no faction can guarantee its pick a seat, only improve its odds. "You cannot bribe a coin," she said. "You can pack a pool, so the pool is where the next fight goes. We know that. We are not promising a machine that cannot be gamed. We are promising one that is expensive to game and cheap to audit."

The question Miksa would not wave away is whether a board insulated from the majority that appoints it can stay insulated across a human — now inhuman — lifespan. Six years, no renewal, and out. The incumbents it would judge have served rather longer. Justice Aurelio Vance has sat one hundred and thirty years on his bench and declines to invoke any retirement standard, existing or invented. The median sitting justice is nineteen years past the retirement norm that prevailed before the therapies arrived. A body that never replaces its old cells gets cancer. Nobody has grown the therapy for an institution with the same habit, and looking at this bench, I am not sure we want to.

Ingrid Aaberg, the retired Accord arbiter the occupation elected as its negotiator on the forty-seventh day, declined tier-one extension herself — a detail her opposite numbers have noticed. When I asked her whether a chair won by attrition renews anything or merely moves the incumbency around, she did not soften it.

"A board is not a cure," she said. "It is a scheduled replacement of parts before they fail. If we build it right, nobody notices it working, the way nobody notices a grown heart. If we build it wrong, we have handed the people who never leave one more room they never have to leave."

Halvard Sunde has filed sixty days' notice, citing a metric the Assembly never enacted. Two hundred occupiers are on the ground floor for a thirty-fifth night, drafting the machine that would have made the notice unnecessary.