FallaciesCausal
Post Hoc
Also called Post hoc ergo propter hoc · False cause · Correlation is not causation
Concluding that because one thing followed another, the first thing caused it.
Definition
Post hoc treats “B happened after A” as proof that “A caused B.”
The Latin means “after this, therefore because of this.” Sequence is one piece of evidence for causation, since causes do come before effects. But many things come before any given event, and most of them had nothing to do with it.
Form
- A happened.
- Then B happened.
- Therefore A caused B.
What’s missing: an account of the mechanism, a comparison with cases where A didn’t happen, and a check of other things that changed at the same time.
How to spot it
“Ever since X, Y has been happening.”
A single before-and-after story is offered as proof.
Credit or blame lands on whoever happened to be in office when a number moved.
Nobody asks what else changed during the same period.
There’s no comparison group: places, people, or times without A.
Examples
01“Crime dropped after the new mayor took office”
A campaign credits a mayor for a fall in the city’s crime rate during their term.
What went wrong
Crime rates often move on regional or national trends. If similar cities saw the same drop without this mayor’s policies, the mayor’s role is unclear. The fair question is how this city compared with places that didn’t adopt the same policies.
02Vaccines and autism
Signs of autism are often noticed around the same age that children get several routine vaccines. Some parents concluded the vaccine caused the condition.
What went wrong
The timing is a coincidence of development schedules. Large studies comparing vaccinated and unvaccinated children found no link. A 2014 meta-analysis covered more than 1.2 million children. The 1998 paper that started the scare was retracted in 2010. The before-and-after pattern looked strong, but the comparison didn’t support it.
03The lucky socks
A fan wore the same socks to three straight wins, then forgot them and the team lost.
What went wrong
Four games is a tiny sample, and there’s no plausible way socks affect a game on television. It’s harmless superstition, and it’s the same reasoning error as the serious cases.
Why it persuades
People are built to look for causes, and sequence is the easiest clue to notice. Stories make it worse: “First this, then that” feels like an explanation. Checking it properly means comparing against cases you never saw, which takes data and effort, while the before-and-after story is right in front of you.
How to counter it
1Ask what else changed
Most events have several things that happened just before them. Name one.
What else changed around the same time that might explain it?
2Ask for the comparison
Causal claims need a counterfactual: what happened where A didn’t occur?
What happened in places that didn’t do this? If they saw the same change, it probably wasn’t this.
3Ask how it would work
A plausible mechanism doesn’t prove causation, but having none is a warning sign.
How would A actually produce B? What’s the step in between?
When it's not a fallacy
Sequence is real evidence when it comes with more support. Controlled experiments, natural experiments, a clear mechanism, a dose-response relationship, or a result repeated across many settings all strengthen the claim. When a light comes on every time you flip a switch, and you know about the wiring, concluding that the switch causes the light is correct.
The fallacy is in treating timing alone as enough. Austin Bradford Hill’s criteria, published in 1965, are still a good checklist for when it isn’t alone.
Sources & further reading
- Austin Bradford Hill, “The Environment and Disease: Association or Causation?” Proceedings of the Royal Society of Medicine 58 (1965): 295–300.
- Luke Taylor, Amy Swerdfeger and Guy Eslick, “Vaccines Are Not Associated with Autism: An Evidence-Based Meta-Analysis of Case-Control and Cohort Studies,” Vaccine 32 (2014): 3623–3629.
- Bradley Dowden, “Fallacies”, Internet Encyclopedia of Philosophy.
End of entry · Back to top