What Is a Black Swan Event? Nassim Taleb's Definition
A Black Swan is a rare, high-impact event that is: 1. Unexpected (by the observer, though not by everyone) 2. Has extreme impact (dominates outcomes) 3. Retrospectively predictable (after it happens, people claim they saw it coming)
Taleb's examples: 9/11, the 2008 financial crisis, the rise of the internet, World War I.
The term comes from the assumption that all swans are white — until you see a black one. The event violates an assumed pattern.
Key Properties
Black Swans are:
Rare: By definition, they don't happen frequently. Most events are normal variation.
Impact Dominated: One Black Swan can exceed the impact of all normal variation combined. One bad financial crisis can exceed the gains of an entire decade of growth.
Invisible to Standard Models: Statistical models built on historical data miss Black Swans because Black Swans, by definition, haven't occurred in the historical data.
Narrative Driven: After they occur, stories are constructed to explain why they were "inevitable" or "obvious." Before they occur, this inevitability is invisible.
The Problem
In Extremistan domains (finance, geopolitics, epidemiology), Black Swans are not negligible tail risks. They dominate outcomes.
The financial crisis wasn't a 1-in-1,000-year event. It was a regular feature of financial systems.
The pandemic wasn't a black swan if you were paying attention to history and virology. It was a black swan to most people because they didn't look.
Distinction from Turkey Problem
Black Swan = rare, high-impact event.
Turkey Problem = the epistemological error of mistaking accumulated safe history for evidence of future safety.
A turkey's problem isn't that Black Swans exist. It's that the turkey's data (1,000 days of feeding) is generated by a process that contains a hidden structural break (Day 1,001).
Go deeper:
For the full breakdown of Black Swans and how they relate to the Turkey Problem, read The Turkey Problem: Nassim Taleb's Epistemological Paradox.