Silent Evidence
Version 1.0.0 · Updated 2026-07-30
CORE DEFINITION
Developed by Nassim Nicholas Taleb in The Black Swan, building on the survivor bias literature and the World War II aircraft study by Abraham Wald. Silent evidence is the data we never observe because the cases that would have produced it were destroyed, failed or filtered out. The core proposition is that the surviving record is a sample selected on survival, so regularities read off it can be artefacts of the selection rather than laws. The key qualification is that the missing cases are usually unobservable by construction, so the correction must be reasoned rather than measured.
SCAFFOLDING EFFECT
Reduce cognitive load
- Sample audit: ask which cases could not have appeared in this dataset. - Graveyard study: examine the failures that shared the same traits as the successes. - Trait test: discard any factor that both the survivors and the dead have in common.
Anchor fast decisions
Selection acts before observation, so only cases that passed the filter can be counted. Any trait correlated with survival then looks like a cause of success, even when it is merely a condition for being seen. Because the excluded cases leave no trace, the bias cannot be detected from inside the dataset, and the only way to correct it is to reason about what the filter removed and go looking for those cases deliberately.
MINIMUM ACTION
In progress 0/1Practice this model in one real situation:
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more
Source support: Explicit
- en.wikipedia.orghttps://en.wikipedia.org/wiki/Silent_Evidenceverified
PRIVATE NOTES · Only visible to you
SAVED Q&A
ENTRY Q&A · Private saving available
Ask with a clear boundary
thinkingmodels answers from published entry context only.
Your question is sent to thinkingmodels. The answer uses public entry context only.
RELATED MODELS