The executive may describe his beliefs sincerely while missing the cause of his own decision. Corporate power turns self-deception into a public risk.
Editorial essay · political philosophy · power and justification
The Reasons of Power
A philosophical exploration of how we treat stated reasons when an official or institution can harm others.
Two ways to get it wrong
Two errors. Different costs.
You can unfairly suspect the person speaking. You can also grant too much credit to the person who controls the factory, the money and the clean-up.
You wrong the speaker
You dismiss a sincere explanation, humiliate them and deny their standing as a rational participant.
You leave third parties unprotected
The institution keeps producing harm while interest, access and self-deception remain outside the explanation.
When the speaker can close the factory and pay for the clean-up, premature trust can prolong the harm to an entire town.
The “Toxic Waste” case
Sincerity does not complete the explanation
Rachel, a resident of the imagined town, accepts that the chief executive may believe what he says. She still follows profit, wealth and the structure of the company.
Her move seems unfair: she rejects another person’s sincere explanation without knowing him closely. But the executive enters the conversation with more than an opinion. His office controls resources, paid experts and the decision to keep the factories open.
Rachel does not say that the executive is lying. She rejects his self-explanation as a complete cause and organises the community against the economic structure keeping the factories open.
Philosophers call this dismissive psychologising: replacing stated reasons with biases, impulses or interests that the speaker would deny. Amy Flowerree and Nicholas Smyth show why it can humiliate and diminish. In ordinary conversation, a presumption in favour of sincere reasons protects something important.
The executive, however, has formal authority: an explicit office, resources and the capacity to impose consequences. If Rachel is wrong, she wrongs him. If the town stops at his explanation while he is self-deceived, the community may keep correcting arguments while the engine of the decision remains untouched.
An official explanation may be sincere and still be incomplete precisely where incompleteness does the most harm.
The argument does not license mind-reading or universal cynicism. It changes the starting point: around power, incentives and institutional history enter the explanation before the official self-portrait is treated as sufficient.
Interactive · no invented scores
The same sentence, a different institutional field
Change the speaker. The sentence stays fixed; what changes is the information that responsible interpretation must seek.
“I made the decision after listening to every side.”
They explain their vote in a private conversation.
What legitimately enters the frame
- Officeno formal office
- Resourcesonly their voice and vote
- Beneficiariesno direct institutional beneficiary
- Affected third partiesdispersed consequences, no unilateral power
A qualitative editorial model derived from the paper. It does not estimate sincerity, rank people on a scale or replace evidence about a particular case.
Propaganda produced through deference
What disappears when the story stops at the quote
Officials routinely ask for more arrests, penalties and funding in the name of safety. The stated reason reaches the headline before the institution’s record enters the frame.
Alec Karakatsanis calls reporting that reproduces police and prosecutorial perspectives as the default explanation of crime and punishment “copaganda”. Shields uses the critique to show how deference to official self-explanation can have propagandistic effects even when the journalist prints no lie.
New powers proposed in response to overdose crisis
“We need stronger tools to keep the community safe.”
The quote organises the story; the institution appears only as authority.
When the quote becomes the explanation, the official is mythologised as a well-intentioned protector, material structure disappears and the remedy looks like one more better argument. If budgets, access and beneficiaries sustain the policy, the distribution of power belongs in the story.
Corruption and self-deception become exotic exceptions.
Budgets, careers and industries leave the explanation.
A problem of power is treated as an exchange of arguments.
Journalism can quote an explanation while treating it as a hypothesis. The questions are observable: who benefits, what history the institution carries, what evidence conflicts with the stated aim, which alternatives were excluded and who bears the cost.
The warning in the other direction
Suspicion travels downward too easily
A rule designed to scrutinise power becomes dangerous when applied indiscriminately to people who control no institution, resources or rules.
Defenders of slavery portrayed emancipated people as incapable of autonomy. Opponents of women’s suffrage invoked women’s supposed emotionality. Colonial administrators described governed peoples as unable to govern themselves. A diagnosis of irrationality prepared the case for exclusion from decision-making.
Research on motivated reasoning remains valuable. But evidential thresholds have moral and political consequences. Shields invokes inductive risk: where an error can feed stigma and disempowerment, researchers must include that cost when setting the threshold.
Maya Goldenberg examined vaccine hesitancy without the convenient portrait of parents as ignorant enemies of science. Their reasons revealed a different sensitivity to risk and, above all, distrust of institutions. Understanding does not validate a false belief; it shows where intervention may work.
The asymmetry creates two duties. Before power, protect third parties from institutional self-deception. Before ordinary citizens, protect political participation from hasty diagnosis.
The exit condition
Institutions can earn the presumption
Rank and confident tone do not produce trust. A credible institution leaves auditable traces and makes corruption or self-deception harder to turn into policy.
Declared conflicts, transparent funding and real barriers.
Departures from consensus are explained publicly with evidence.
Consultation can change the outcome, not merely decorate it.
Failure changes criteria and practice, not just messaging.
Reporting abuse does not cost a career, liberty or safety.
Independent evaluation supports promises in its field.
Five questions for an official statement
- What can this person do because of their office?
- Who bears the cost if I accept the explanation as complete?
- What incentives does the institution create?
- Which facts conflict with the stated aim?
- What independent mechanisms would make the institution credible?
The answers cannot identify the “real motive” on their own. They show what evidence is missing and how dangerous it is to stop at the speaker’s statement.
“The world has suffered more from leaders and authorities than from the masses.”John Dewey, 1926, quoted by Shields on p. 34
Power changes the cost of taking someone at their word. An official’s stated reasons are one part of the explanation; office, resources, beneficiaries and affected third parties are the others.
Provenance and limits
The argument can be traced to the page
The primary source is a 36-page preprint. The locators below make every mechanism used in the essay verifiable.
Primary source. Toxic Waste case: pp. 9–13; donations and formal authority: pp. 14–19; media and propaganda: pp. 21–25; exclusion and inductive risk: pp. 26–33; conclusion: p. 34.
PhilArchive ↗The presumption in favour of sincere reasons and the high epistemic standard for psychologising; discussed by Shields on pp. 6–8.
DOI ↗The moral and interpersonal harm caused by psychological debunking; discussed on pp. 5–8.
DOI ↗The critique of police-centred reporting and the way official reasons organise a story; used on pp. 21–25.
The New Press ↗Vaccine hesitancy as a problem of institutional distrust rather than simple hostility to science; pp. 31–33.
University of Pittsburgh Press ↗An editorial essay by Marius Comper, made with Codex, based on Matthew Shields. The cases, the distinction between two errors and the trust criteria belong to the paper; the selection, sequence and interactive mechanisms belong to this adaptation. On 15 July 2026, PhilArchive still listed the paper as forthcoming in Episteme. The interactive is qualitative and does not claim to detect lies or real motives.