The Self-Refutation Test
Take an idea literally and follow it through; bad ones destroy themselves.
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 80%
Naval names this recurring move in Deutsch's work a self-refutation. You take a theory at face value, apply it to its own case, and watch whether it survives. The precautionary principle is the clean example: civilisation has never followed it, so adopting it now would be an unprecedented and therefore unprecautionary act. Solipsism is the deeper one: take seriously the claim that everything is your dream, and you immediately need an explanation for where the details of the dream, the bus, the passengers, the person in the yellow suit, came from, which reintroduces the external world you were trying to eliminate. Deutsch reframes the technique: it is not a special refutation method at all, it is just what it means to take theories seriously rather than as forms of words one learns to say.
Origin
Naval identifies the pattern across Deutsch's books and calls it a Deutsch refutation. Deutsch rejects the framing of it as a method, saying it is simply the consequence of taking a theory seriously instead of treating it as a phrase to repeat.
Core principles
- 01Most bad ideas are refuted from the inside, not by outside evidence.
- 02Taking a theory seriously means treating it as a claim about reality, not a form of words.
- 03If a rule was never followed before, adopting it now usually violates the rule itself.
- 04A theory that has to forbid a question is protecting itself, not explaining anything.
- 05Cheap internal criticism should come before expensive external testing.
How to run it
- 1
Restate the claim literally
Convert the slogan into an explicit statement about reality. Deutsch's complaint is that many positions survive only as forms of words that people learn to say.
Pro tip If you cannot restate it without the jargon, that is already a signal.
- 2
Apply the claim to itself
Ask what the theory says about its own adoption, its own evidence and its own advocates. Universal rules are the easiest to break this way.
Pro tip Rules containing words like always, never or all are the highest-yield targets.
- 3
Follow the consequences one step further
Advocates usually stop at the conclusion they like. Keep going: if everything is a dream, what explains the contents of the dream?
Pro tip Ask what would still need explaining even if the theory were true.
Watch out Stopping at the appealing conclusion is exactly how a self-refuting idea stays alive.
- 4
Watch for forbidden questions
Note any point where the theory's defenders declare a question meaningless or not allowed. Deutsch treats that as a way of shutting up the questioner rather than an answer.
Watch out You get used to it is a social move, not an explanation.
- 5
Discard or keep, then say why
If the theory undermines itself, drop it and record the reason. If it survives, you now have a sharper statement of it to test against rivals.
Pro tip Write the surviving statement down; the sharpened version is worth more than the original.
In the wild
Naval applies the test to the precautionary principle, the rule that you should not adopt anything whose consequences are not fully known. Civilisation has never operated that way, so switching to the principle now would itself be an unprecedented change with unknown consequences. Following the rule therefore breaks the rule, and the principle collapses on its own terms without any empirical dispute about any particular technology.
→ A widely invoked policy rule is defeated from the inside, before any argument about specific risks is needed.
Deutsch takes solipsism seriously instead of dismissing it. If all of this is his dream, then he is dreaming the bus and everyone on it, including a person wearing a yellow suit he had never thought of before seeing. He now needs an explanation for how the contents of the dream came about, which requires exactly the kind of complex autonomous structure that solipsism was meant to do without.
→ Solipsism reintroduces the external world under a new name, so it explains nothing and defeats itself.
Common mistakes
Treating it as a debating trick
Deutsch resists calling it a method of refutation. Used as a gotcha it becomes point scoring; used properly it is just the discipline of taking any theory, including your own, literally.
Accepting a question ban as an answer
When defenders respond that a question is not meaningful or not allowed, the natural reading is that the theory cannot answer it. Deutsch describes this as shutting up the questioner rather than explaining.
Expecting external evidence to do the work
Waiting for data to settle a self-refuting idea wastes time and money, because the contradiction is internal and can be found in minutes.
Is it for you?
Best for
Anyone stress-testing a policy, philosophy or strategy where running an empirical test is slow, costly or impossible.
Not ideal for
Choosing between two internally coherent rival explanations, which needs a genuine experiment rather than internal criticism.
From the transcript
“you talk about the precautionary principle since civilization has never followed the precautionary principle if we start following it now we're no longer being precautionary…”
“rather than thinking of it as a method of refutation is to think this is just what it means to take theories seriously”
“that's really why solipsism destroys itself and in philosophy in physics too most ideas destroy themselves”
From the episode
David Deutsch: Knowledge Creation and The Human Race, Part 2