In crisis management, we teach that a crisis has a lifecycle. Something happens. You respond. You recover. You write the after-action report, you brief the board, and the event becomes a chapter in the organization’s history.

I’ve built my career on that model. I now think it’s missing its final phase.

The event ends. The fight over what the event was does not. And in the current information environment, that fight can run for years, reverse itself, and produce an outcome the response team never imagined. The crisis after the crisis is the battle over its history.

What I study

Alongside running my firm, I’m a doctoral candidate at Hamline University. My research sits where my professional life always has: on information, and what happens when it breaks. Specifically, I study regimes of truth, which is a term from the philosopher Michel Foucault that sounds more complicated than it is.

Every society has machinery for deciding what is true and who is entitled to say so. For a death, that machinery is the medical examiner, the courts, the forensic sciences, and the press. For a plane crash, it’s the NTSB. For your company’s worst day, it’s the investigation, the regulators, and the official record you produce. That machinery, plus the authority we grant it, is a regime of truth.

My research asks what happens when a community stops accepting that machinery and builds its own.

The case I just published

This month I posted a new working paper on SSRN. It examines something that happened in my own city.

Five years after George Floyd’s murder, a substantial share of Americans no longer believe it was one. A revisionist account, holding that Floyd died of a drug overdose and that Derek Chauvin was wrongly convicted, has become the dominant account within a large American community, and by 2025 it had organized itself into a campaign for a presidential pardon.

The drift is measurable: the share of Americans characterizing Floyd’s death as murder fell from 60 percent in June 2020 to 36 percent by March 2021.

The usual frame calls this misinformation: false claims to be fact-checked and corrected. My paper argues that framing mistakes the object entirely.

What consolidated is not a set of false statements. It’s a rival system for producing truth, a counter-regime, with its own authorized speakers, its own evidentiary hierarchy, its own media institutions, and its own mechanism of sanction. You cannot fact-check a system. You have to understand it as architecture.

Two findings from the paper matter beyond the case itself.

First: the counternarrative was built almost entirely from the official record’s own material. The autopsy’s equivocations. The charging document’s language. Framing choices in mainstream coverage itself. The revisionist narrative did not have to invent the raw material of doubt. It was handed to it. That is why fact-checking cannot dislodge it: an account assembled from the regime’s own record is not refuted by appeals to that record. The record already contains it.

Second: counter-regimes mature through identifiable stages. They begin in what I call counter-forensics, contesting the official evidence on the official evidence’s own terms. They end in counter-history, a self-authorizing account that no longer needs the official record at all, and that seeks ratification from power itself. By the time you notice the mature version, the window for engaging the early version has closed.

Why a crisis leader should care

I know how this sounds: interesting for political scientists, remote from the operations bridge. Here’s why I think it’s the opposite.

Your organization’s crisis will produce an official record. An investigation. A root-cause analysis. Regulatory findings. Press statements. And that record will contain what every official record contains: hedged language, preliminary findings later revised, cautious lawyerly phrasing, and internal disagreements that surface in discovery. In the old information environment, those artifacts stayed inert. In the current one, they are raw material, and there are now well-developed templates for assembling them into a counternarrative.

Four working conclusions I’d offer any leadership team:

The narrative phase outlasts the response phase, so your monitoring has to. Most organizations stop watching the information environment when the incident closes. The record shows the fight over meaning can be just getting started.

Your own equivocations are the ammunition. The hedge you write to be careful is the seam a counternarrative opens later. This is not an argument against honesty. It’s an argument for a first account that is as complete, as plain, and as fast as the facts allow, because gaps get filled by someone else’s architecture.

You cannot fact-check your way out. If a rival account of your crisis takes hold, rebuttal memos aimed at individual claims will fail, because the audience has changed whose evidence counts. Understanding who the rival account’s authorized speakers are, and what its evidence hierarchy is, tells you more than any fact-check.

Watch the early stage. Counter-forensics, the phase where a rival account is still arguing with your evidence, is the phase where engagement is possible. Counter-history doesn’t argue with you anymore.

The through line

Thirty years of crisis work taught me that crises are won and lost on information. The research is teaching me what that means when the information environment itself is contested ground.

The paper is on SSRN if you want the full argument, method, and evidence. It’s a working paper, which means it awaits peer review, and I’d welcome serious critique. That’s what working papers are for.

More of this to come. The research program and the practice keep turning out to be the same subject.