The Beneficial Heuristic Test
Judge a rule by the behavior it produces, not whether it is literally true
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 68%
Sutherland's recurring model separates a belief's literal truth from its behavioral consequences. A rule that is factually wrong, or whose reason is unknown, can still be worth obeying if adhering to it reliably produces beneficial behavior. Miasma theory of disease was wrong, yet it funded sewers, waste management and well-ventilated hospitals. Telling children that treading on pavement cracks risks a bear attack is false, but it keeps them watching their feet and staying off the road. The corollary is Chesterton's fence: do not tear down an inherited rule just because you cannot see its purpose, and do not attach a reason to a good rule, because a reason becomes a loophole. The test is empirical: measure the benefit of following the belief, not the belief's accuracy.
Origin
Sutherland draws on Chesterton's fence, David Rock, Kahneman and his own advertising practice, framing heuristics as socially beneficial customs that get formalized as law without requiring correct reasoning.
Core principles
- 01A belief can be false or unprovable and still be worth holding if acting on it produces good outcomes.
- 02Correct reasoning is not required to reach a correct conclusion.
- 03Socially beneficial customs often precede and outlast the reasons given for them.
- 04Giving a rule a stated reason creates a loophole people will exploit to drop it.
- 05An old rule whose purpose you cannot see may still be load-bearing (Chesterton's fence).
How to run it
- 1
Isolate the rule
State the belief, custom, superstition or policy you are tempted to evaluate or discard.
- 2
Switch the question
Stop asking whether the rule is true and ask instead what behavior believing it reliably produces.
Pro tip A myth that changes behavior is doing real work even when the mechanism it claims is fiction.
- 3
Measure the behavioral benefit
Assess empirically whether the behavior the rule triggers is beneficial, harmful or neutral.
Pro tip The benefit is often tangential to the stated reason, like a dietary rule improving gut health for reasons unrelated to the myth given.
- 4
Keep what works
Retain rules that produce good behavior even when the reasoning is wrong or unknown.
Watch out Do not confuse this with endorsing the false claim; you are endorsing the behavior.
- 5
Withhold the reason
Avoid attaching a neat justification to a good rule, because a stated reason becomes an exploitable loophole.
Pro tip The strongest rules are Chesterton's fences: obeyed without a why, so no one argues the why no longer applies.
Watch out If you justify a hygiene rule by hygiene, people drop it the moment they believe hygiene is handled.
In the wild
Sutherland wrote in the Spectator about telling children that treading on pavement cracks risks a bear attack. There is no link between cracks and bears, but children fear bears instinctively, so the rule makes them watch their feet, avoid tripping, and stay on the pavement away from traffic. It is transport-safe behavior that is otherwise impossible to mandate.
→ A false belief reliably produced safer child behavior no direct instruction could achieve.
Sutherland recounts arguing with Steven Pinker that although germ theory is the correct theory of disease, the earlier and wrong miasma theory led to beneficial actions: funding sewerage, waste management and well-ventilated hospitals with big windows.
→ Wrong reasoning drove correct, life-saving public-health decisions.
Religious prohibitions on pork, and rapid burial of the dead in Muslim and Jewish law, encoded strong hygiene benefits long before anyone could explain parasites or decomposition. Quarantine's forty days was a heuristic for how long a threat takes to manifest.
→ Customs delivered public-health outcomes without the underlying science being understood.
Common mistakes
Demanding literal truth before obeying
Discarding a rule because its stated reason is false throws away the beneficial behavior the rule reliably produced.
Attaching a reason to a good rule
A stated justification becomes a loophole; people argue the reason no longer applies and abandon the rule.
Bulldozing a fence you do not understand
Assuming an inherited rule is pointless because you cannot see its purpose ignores that its author may have had a good reason.
Is it for you?
Best for
Evaluating norms, superstitions, dietary laws, safety rules or policies where behavior matters more than the literal justification.
Not ideal for
Domains where the truth of the claim itself is the whole point, such as a scientific or engineering fact.
From the transcript
“in order to have these successful heuristic rules they need to be capable of being chesterton's fences in other words something you obey without knowing…”
“the reasoning can perhaps be wrong but you can still arrive at a conclusion which is correct”
“by adhering strictly to stepping within the cracks you will as a byproduct of that belief you will adopt a more transport safe behavior”
From the episode
The Psychology Of Transport, Google Maps & Bear Attacks - Rory Sutherland - #401
Rory Sutherland