My third academic paper is now available on SSRN: From Counter-Forensics to Counter-History: The Maturation of a Counter-Regime of Truth in the George Floyd Revisionist Narrative.

The paper starts from a fact that should be more startling than it is.

Five years after George Floyd was murdered on a Minneapolis street, a substantial share of Americans no longer believe he was murdered. Polling captured the drift early: the share of Americans characterizing Floyd’s death as murder fell from 60 percent in June 2020 to 36 percent by March 2021. In place of the adjudicated account stands a revisionist one. Floyd died of a drug overdose. The restraint was incidental. Derek Chauvin was wrongly convicted by a captured legal process under mob pressure.

That account did not stay at the fringe. It moved through cable punditry into professionally produced documentary, acquired the trappings of forensic and intellectual respectability, and by 2025 had organized itself into a campaign for a presidential pardon.

The paper asks how that happened. Not whether the revisionist claims are true or false, but how the account was built, who it authorizes to speak, and by what mechanisms it moved from fringe assertion to dominance within its community.

Why “misinformation” is the wrong frame

The instinct is to call this misinformation and treat it as a defect in the information environment, a pile of false claims waiting for a fact-check. I argue that framing mistakes the object.

Following Foucault, every society has a regime of truth: the institutions, procedures, and authorities through which official truth is produced. For a death like Floyd’s, that means the courts, the medical examiner, the forensic sciences, and the mainstream press.

What consolidated against the settled account is not a set of false statements inside that system. It is a rival system. It has its own authorized speakers: dissident officers, a defected journalist, physicians who break with the medical examiner. It has its own evidentiary hierarchy, elevating raw body camera footage over the jury’s verdict and a surfaced deposition over the autopsy’s conclusion. It has its own institutions of circulation and its own mechanism of sanction, the demand that state power ratify its verdict through a pardon.

A regime of truth is not refuted the way a false statement is. It has to be understood as an architecture. That is what the paper does.

The two findings

The first finding is what I call the recycling thesis. The counternarrative was assembled largely from the official record itself. The overdose frame did not originate in conservative media. It was drawn from the autopsy’s own equivocations, from the initial charging document’s language deflecting agency toward Floyd’s health, and from moral-disengagement framing that content analysis found in 88.9 percent of sampled mainstream coverage.

The revisionist narrative did not have to invent the raw material of doubt. It was handed to it. This explains a durability the misinformation frame cannot: an account built from the regime’s own record is not dislodged by appeals to that record, because the record already contains it. This is why fact-checking fails against it.

The second finding is a developmental model. Counter-regimes of truth mature through four identifiable stages: insurgent assertion, where punditry names the suppressed truth at scale; forensic mimicry, where the counternarrative borrows the state’s own evidentiary forms because it cannot yet authorize itself; self-authorizing counter-history, where the outlet’s own voice suffices and evidence relaxes into typology and insinuation; and the demand for sovereign ratification, where the counter-regime asks state power to seal its verdict.

How I studied it

The corpus spans six layers, from the original 2020 documents through the punditry, the documentary production, a respectability contest among heterodox intellectuals, and the pardon campaign.

The spine of the analysis is two documentaries released two and a half years apart by the same Minnesota outlet: The Fall of Minneapolis (2023) and Minnesota Mao (2026). I broke both films into segments and coded every claim on three dimensions: what persuasive strategy is at work, where the cited evidence actually comes from, and how the editing itself makes arguments the script never states.

The two films turn out to be one argument at two stages of maturity. The first film builds a case from the state’s own documents: the autopsy, the training manual, the trial record, hours of body camera footage. It attacks the state’s verdict with the state’s own paperwork, which is the signature of a counter-regime that still depends on the thing it contests.

The second film no longer argues the case. It treats the verdict as settled and moves on to what it means. What the first film had to prove, the second merely serializes.

Beyond this case

The model generalizes. Election denial after 2020 shows the full arc: insurgent assertion, a forensic-mimicry stage of affidavits and litigation that borrowed the courts’ evidentiary grammar, a self-authorizing stage that needed no adjudicated evidence, and ratification demands aimed at legislatures and officials. Vaccine revisionism fits partially.

The claim is not that every counternarrative follows all four stages. The claim is that where a counter-regime consolidates, the consolidation takes a specific, measurable form: a shift in where its claims get their authority, from the institutions it opposes to itself. Once that shift completes, the facts that might have corrected it no longer can.

What the paper does not do

The study takes no position on the truth or falsity of the counternarrative’s claims. That is a deliberate methodological stance, not a hedge. The question a discourse analysis can answer is how an account organizes meaning, authorizes speakers, and structures political action.

I also disclose my own position fully in the paper. I am a participant in Minnesota’s public life on the political right, and the outlet whose films I code covers that world. Those facts cut in different directions, and the paper treats them as liabilities to be constrained: the coding instrument was locked before analysis, every change is logged, and the provenance codes are checkable by any reader against the public record. The complete coding registers, codebook, and adjudication records are deposited here on this site.

The research program

This is the third study in a continuous program on how rival regimes of truth are constructed. It began with my MA dissertation at King’s College London, in the Department of War Studies, which analyzed how Donald Trump and the American alt-right construct a rival conception of truth.

A second paper, Regimes of Truth in the Digital Age, carried the framework to a regional actor, Royce White’s 2024 Senate campaign.

This paper adds what the first two lacked: a counter-regime constructed retrospectively, about a settled past event.

I am a doctoral candidate in Management and Public Service at Hamline University, and this work continues there.

The full paper is on SSRN. Comments and academic critique are welcome.