The veil of ignorance
The veil of ignorance is a thought experiment for fairness: design the rules of a society — or any deal — as if you didn’t know which position in it you’d occupy. Not knowing your wealth, talents, or identity, you’d choose rules fair to every position.
How it works
Stripping away knowledge of your own stake removes the self-interest that warps judgments of fairness. Behind the veil you must weigh the worst-off outcome as if it could be yours, so you tend to pick rules that protect everyone, not just the lucky.
How to use it
- Testing fairness: ask “would I accept this rule if I didn’t know which side of it I’d be on?”
- Designing systems, contracts, or splits where you want a result that’s defensible to all parties.
- Reducing self-serving bias in policy or product decisions that distribute benefit and burden.
Worked example
Asked to cut a cake fairly, the person who cuts takes the last slice — they’ll divide it evenly because they don’t know which piece they’ll get. The veil generalises this: choose the rule before you know your place, and self-interest produces fairness.
Where it fails
It’s an idealisation — real people can’t fully shed self-knowledge, and critics argue the “original position” quietly assumes particular attitudes to risk. It clarifies fairness; it doesn’t settle every disagreement about justice.
Frequently asked
- What is the veil of ignorance?
- A thought experiment in which you choose the principles of a fair society without knowing your own place in it, so the rules are fair to everyone, especially the worst-off.
- Who came up with the veil of ignorance?
- Philosopher John Rawls, in his 1971 book A Theory of Justice, as part of the “original position”.
- How do you use the veil of ignorance in everyday decisions?
- Ask whether you’d accept a rule or split if you didn’t know which side of it you’d be on — a quick test that strips out self-interest.
Related
Editorial synthesis © ReadGlobe 2026, drawing on the mental-models tradition (Charlie Munger, Farnam Street) and the primary sources for each model. · Last reviewed 2026-06-30.