Hanlon's Razor

Hanlon's Razor and conspiracy theories. Conspiracy theories systematically violate Hanlon's Razor by attributing to coordinated malice what is usually explained by coordination failure, incompetence, or independent self-interested action. The COVID-19 pandemic generated many conspiracy theories that required implausible levels of coordinated deliberate harm across independent institutions in many countries. Hanlon's Razor points toward more mundane explanations.

When to apply the opposite. There are domains where assuming good faith is dangerous. In adversarial security contexts — penetration testing, intelligence analysis, fraud detection — assuming malice is the professional default. A penetration tester who assumes the network was designed well and the logs are accurate is a bad penetration tester. Context determines which prior is appropriate.

Institutional malice. Organisations can have structural effects that harm specific groups without any individual intending harm. Redlining, discriminatory algorithms, and unfair hiring processes can be the result of decisions that no individual intended as harmful. The razor is helpful here: attributing these to conspiracy obscures the structural nature of the problem. But structural harm is not the same as no harm — it still requires a structural response.

The charitable interpretation as principle. In philosophy of language, the principle of charity says: interpret an argument in its strongest form before criticising it. Hanlon's Razor is the behavioural equivalent: interpret actions in their most benign form before attributing harmful intent.

1941. Robert Heinlein's short story "Logic of Empire" contains the line: "You have attributed conditions to villainy that simply result from stupidity."

1980. Robert Hanlon of Scranton submits a version to Arthur Bloch's Murphy's Law Book Two. It is included and cited as "Hanlon's Law" with an attribution to Robert Hanlon.

Late 20th century. The saying circulates widely in technology communities, particularly in discussions of software design, corporate behaviour, and government policy. Its appeal is highest in contexts where the alternative (assuming everything bad is deliberately caused) is clearly unproductive.

Conflict reduction. Interpreting harm as malice when it is incompetence creates adversarial dynamics that are harder to resolve than the original problem. The natural response to perceived malice is retaliation; the natural response to incompetence is correction. Choosing the latter when it's accurate reduces unnecessary conflict.

Better diagnosis. Correct attribution leads to correct response. If a system harms people through bad design, the fix is to redesign the system. If it harms through malicious intent, the fix is to expose the intent and apply consequences. Applying the malice fix to a design problem, or the design fix to a malice problem, doesn't work.

Epistemic humility. The razor is a reminder that other people's internal states — their intentions and motivations — are not directly observable. The malicious interpretation is a strong claim about inner states. The incompetence interpretation is a claim about external behaviour. The weaker claim requires less evidence.

Online discourse. Social media environments tend to promote the malicious interpretation: misunderstandings are read as attacks, mistakes are read as deliberate provocations, and the adversarial interpretation spreads faster than the charitable one. Hanlon's Razor is a counterweight — not always right, but a useful corrective to the default escalation dynamic.

AI alignment. When an AI system behaves badly — generates harmful content, gives wrong answers, pursues misspecified goals — the temptation is to treat it as intentional. It is not. AI systems cause harm through misalignment, not malice. The correct analogy is incompetence, not villainy. The practical responses (retraining, realignment, specification improvement) follow from the correct diagnosis.