Radical Uncertainty Inquiry
Isolate mechanisms and unknowns before declaring a cause or solution
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 96%
Radical Uncertainty Inquiry replaces the demand for a simple answer with a structured investigation of competing explanations. Begin by identifying all plausible factors, then isolate the mechanism through which each could have produced the observed result. Distinguish factors that merely appeared alongside the outcome from those supported as causal. For every unresolved claim, ask what information would be required to endorse or reject it. The output is not permanent indecision; it is a calibrated conclusion that shows what is known, what remains uncertain, and which evidence would resolve the uncertainty. The method is especially useful for financial crises, policy disputes, and other complex systems where several forces interact and counterfactuals cannot be directly observed.
Origin
Peter Conti-Brown described himself as a radical uncertainist while examining competing explanations for the financial crisis on Masters in Business.
Core principles
- 01Treat certainty as something earned through evidence.
- 02Separate plausible contributing factors from proven causes.
- 03Identify the mechanism connecting each factor to the outcome.
- 04State what evidence would change or resolve your view.
How to run it
- 1
Frame the contested question
Express the dispute as a specific causal or decision question rather than a vague disagreement.
Pro tip Define the outcome and relevant time period before debating explanations.
Watch out A broad question can hide several materially different claims.
- 2
Map competing explanations
List the factors that credible participants believe contributed to the outcome. Include explanations you initially dislike.
Pro tip Treat explanations as candidates rather than mutually exclusive camps.
Watch out Do not elevate the most memorable factor into the sole cause.
- 3
Isolate each mechanism
Describe how each candidate factor would move through the system to produce the outcome. Identify intermediate steps and feedback loops.
Pro tip Use an explicit inputs-to-process-to-output chain.
Watch out Correlation without a mechanism is not a sufficient causal account.
- 4
Inventory the unknowns
Mark assumptions, missing measurements, and unobservable counterfactuals within each mechanism.
Pro tip Phrase each uncertainty as a question that evidence could answer.
Watch out Do not disguise missing evidence with confident language.
- 5
Define decision-changing evidence
State what you would need to observe to favor one explanation over another, then calibrate the conclusion to current evidence.
Pro tip Record what evidence would make you reverse your view.
Watch out Uncertainty should improve inquiry, not become an excuse to avoid every conclusion.
In the wild
A policy analyst considers low central-bank rates, global savings, securitization, underwriting failures, and derivatives deregulation. Instead of naming one culprit, the analyst maps how each factor could expand credit or weaken risk controls, identifies missing evidence about relative credit availability, and states which observations would distinguish a primary cause from a contributing condition.
→ The analyst produces a defensible causal map and a targeted research agenda rather than an oversimplified verdict.
Common mistakes
Demanding one master cause
Complex systems often produce outcomes through interacting forces, so selecting one vivid factor can erase the operative mechanism.
Using uncertainty as paralysis
The method requires calibrated judgment and evidence-seeking, not indefinite refusal to form a view.
Is it for you?
Best for
Researchers, investors, policymakers, and leaders evaluating contested events or uncertain decisions.
Not ideal for
Situations requiring an immediate operational response under a clear and established rule.
From the transcript
“I am a radical uncertain tist.”
“what would I have to know in order to horse one or another?”
“by really digging deep, isolating the mechanisms, ask yourself what you don't know, what you'd need to know in order to answer that question.”
From the episode
Peter Conti-Brown Discusses U.S. Financial History (Podcast)