The right to be wrong

There is a defensive reflex in institutions that treats every challenge as an accusation of error, and every admission of error as the loss of legitimacy. The reflex assumes that to be questioned is to be threatened, and that the institution's standing depends on never being shown to have been wrong. This is a misunderstanding of where legitimacy actually comes from, and it leads institutions to defend the wrong thing — to spend their energy concealing mistakes rather than answering for them, which is the surest way to forfeit the standing they were trying to protect.

Begin with what should be obvious and is somehow controversial: institutions are allowed to be wrong. A hospital can misdiagnose. A bank can misprice. An agency can misjudge a risk. None of these is, by itself, an injustice. They are the unavoidable residue of acting under uncertainty, which is the only condition under which any institution ever acts. A decision made with incomplete information, in good faith, by a competent process, can still turn out to be wrong, and when it does, the wrongness is a misfortune, not a wrong in the moral sense. We do not, and should not, demand that institutions be infallible. We could not have institutions at all on that condition.

What we are entitled to demand is something different and much more achievable. Not that the institution never err, but that when it errs, the error can be seen — traced to its source, examined for the reasoning that produced it, and corrected. The injustice is not the mistake. The injustice is the mistake that cannot be found. A wrong decision that leaves a legible trail is a misfortune the institution can answer for. A wrong decision that leaves no trail, that cannot be reconstructed or questioned or even located, is something else: it is a wrong that has been placed beyond the reach of correction, and placing it there is the actual offense.

The right an institution should defend is not the false right to never be wrong. It is the right to err and answer for it. The first cannot be defended. The second is the whole of legitimacy.

Two kinds of error

So there are two errors hiding under the single word, and an institution that conflates them defends the indefensible while neglecting the thing it could actually have. The first error is the wrong decision — the misdiagnosis, the misjudgment, the call that went the wrong way. This error is forgivable, because it is inherent in deciding under uncertainty, and an honest accounting of it can even strengthen an institution, by showing that its process is sound even when its outcomes are not. The second error is the unaccountable decision — the one that cannot be seen into, whose reasoning is unavailable, whose inputs are unrecoverable, which simply happened and now stands. This error is not forgivable, because it forecloses the only remedy that matters: the ability to find the mistake and fix it.

An institution that has built itself to be unaccountable has not protected itself from error. It has only protected its errors from view, which is a different and worse position. The wrong decisions are still happening; they are simply now invisible, accumulating without correction, until they surface as a pattern too large to deny. The institution that could see its own mistakes catches them small. The institution that cannot see them discovers them at scale, all at once, in the form of a crisis. The error you are allowed to make is the one you can find. The unaccountable institution has traded the forgivable error for the unforgivable one and called it protection.

What standing actually rests on

This reframes what it means for an affected person to contest a decision. The reflex treats contestation as an attack on the institution's competence — a demand that it admit it was wrong. But the right to contest is not the right to be proven correct. It is the right to have the decision examined — to make the institution show its reasoning, expose its inputs, and demonstrate that the decision, even if it turns out to be mistaken, was made by a process that can be inspected and corrected. A person who contests a decision and learns it was wrong has been served by the system working. A person who contests a decision and cannot even find out how it was made has been denied the thing contestation exists to provide.

And this is precisely why the record matters more than the outcome. An institution that keeps faithful records of its decisions is not exposing itself to challenge; it is building the only foundation on which it can survive challenge. The record does not have to show that every decision was right. It has to show that every decision was made by a process the institution is willing to stand behind and let others see. That willingness — to be wrong in the open, to have the reasoning examined, to correct what examination reveals — is the substance of legitimacy. It is worth more than a perfect record, because a perfect record is impossible and everyone knows it, while an honest, inspectable, correctable record is achievable and is exactly what trust is made of.

The institution that grasps this stops defending its infallibility, which it never had, and starts defending its accountability, which it can actually build. It keeps the record not despite the risk that the record will reveal a mistake, but because of it — because a mistake revealed and corrected costs far less than a mistake hidden until it cannot be hidden anymore. The right to be wrong, held openly and answered for, is the only right an honest institution needs. The false right to never be questioned is the one that destroys them.

— Dispatches · Summit Cognitive