Hasty generalisation
Secundum quid, Généralisation hâtive
Drawing a general conclusion from too few cases, or from unrepresentative ones.
The essentials
Argument diagram
A few observed Xs have property P.
So all Xs have P.
Why the argument is invalid
A small sample is very sensitive to chance, and a non-random sample can be biased. Personal anecdotes are particularly misleading, because they are striking.
In practice
Example
“My grandfather smoked all his life and lived to be 95, so tobacco isn’t that dangerous.”
Defending yourself
How to respond
Ask how many cases were observed, how they were chosen, and what broader data say.
Don’t confuse
Not to be confused with: Cherry-picking
Hasty generalisation concludes from too few cases; cherry-picking, from cases chosen from among others that contradict the conclusion.
Going further
Source
John Stuart Mill, A System of Logic (1843), Book V, chapter 5, “Fallacies of Generalization”: following Bacon, he criticises induction by simple enumeration, which rests on fewer instances than it should.
Practise
In the app, each entry offers ways to practise recognising the concept. No sign-up: your progress stays on your device.