Moral Luck, explained.
Moral luck concerns the tension between responsibility judgments and factors outside an agent's control. Resultant luck asks how differing outcomes affect judgments of otherwise similar conduct.
Why it happens
Two agents can take the same risk while chance determines whether a bad consequence follows. A changed evaluation can therefore reflect outcome information rather than a difference in the original choice.
Moral luck examines tensions between judgments of responsibility and factors outside an agent's control. Resultant luck concerns how outcomes affect assessment of otherwise similar conduct.
Read the result
Rate conduct before revealing deliveries, then rate again afterward. Compare actual rating changes. One damaged and one intact parcel are paired by design to isolate outcomes, not to estimate risk or diagnose bias.
A worked example
Identical skipped checks, different deliveries
Both couriers knowingly skip the same protective check under identical stipulated risk.
One artwork shipment is revealed as damaged and the other intact, with the assignment seeded for the demonstration.
Reassess responsibility and compare with the earlier ratings. Ask which reason changed and whether that reason concerns conduct, outcome, or both.
OPTIONAL DEEPER DETAILGo deeper: inside the model
Inside this model
Both couriers make the same stipulated negligent choice with identical risk. Record responsibility ratings from one to five before outcome disclosure. A seeded bit assigns one damaged and one intact shipment for a controlled contrast; after-ratings and differences are recorded without prescribing blame.
Where this idea is useful
A practical use
An evaluation can distinguish the quality of a decision from a fortunate or unfortunate result.
A common misconception
“A fortunate outcome proves that the original conduct was careful.”
A result can be lucky even when the action's evidence and risk remain unchanged. Conversely, a bad result alone does not establish that an otherwise reasonable decision was negligent.
What this explanation leaves out
- One damaged shipment is paired with one intact shipment by design, not sampled independently or used to estimate damage probabilities. There is no legal responsibility score, human injury, or psychological diagnosis.
Do equal initial actions require equal final judgments?
That is part of the philosophical question, not the game's answer. Different accounts distinguish blame, responsibility, repair, and legal liability in different ways.
When you evaluate a decision, what belongs to its quality and what belongs to the luck of its outcome?
Associated thinkers
Associations marked provisional are awaiting source review.
Further reading
Explore the original research or the teaching reference behind this experiment.