The Scientific Demarcation Test
Separate real knowledge from woo by asking: can you measure it and repeat it?
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 72%
Ferrie's test collapses a hard epistemic question into a practical filter borrowed from how working scientists decide they 'know' something: a claim counts as knowledge only if it can be measured and the measurement repeated by others to the same result. Run any claim through three gates — Is it measurable (can you put a number on it)? Is the procedure repeatable and shareable? Does it connect to the tested scientific corpus? A claim that fails isn't necessarily false, but it's untethered from prediction and should be held as belief or metaphor, not fact. Crucially, even claims that pass are treated as useful models rather than universal truths, to be abandoned the instant new evidence stops fitting. The test is domain-agnostic: it works on quantum-healing, vibrational-frequency, diet, and entanglement claims alike.
Origin
Articulated by quantum physicist Chris Ferrie on Modern Wisdom while explaining how to spot 'quantum bullshit' — misuse of physics terms in spirituality and pseudo-medicine.
Core principles
- 01In science, to 'know' something means you can measure it and repeat the measurement.
- 02If a claim can't be measured with a shareable procedure, it isn't scientific.
- 03Not-scientific doesn't automatically mean wrong, but it usually means useless for prediction.
- 04Models of reality are useful, not universally true; keep the ones that pay off.
- 05Abandon an assumption the moment it stops fitting new evidence.
How to run it
- 1
Restate the claim precisely
Force the vague claim into a specific sentence. 'Emotions vibrate at a frequency and I raised mine' has to become an exact assertion before it can be tested.
Pro tip Keep asking 'what do you mean by that?' until the claim is concrete or collapses.
- 2
Ask if it can be measured
Check whether the thing can be assigned a number or quantity. If you can put kilojoules or hertz on it, it becomes something you can make predictions with.
Watch out Colloquial and technical uses of a word (energy, frequency, resonance) feel similar but aren't the same claim.
- 3
Ask if the procedure is repeatable
A measurement only counts if you can tell someone the exact procedure and they can reproduce the result. Unrepeatable one-offs are anecdote, not evidence.
Watch out Confirmation bias makes a single lucky outcome feel like proof; ignore the misses at your peril.
- 4
Check the connection to tested knowledge
Ask whether the claim links back to the existing scientific corpus or floats free of it. Real effects connect to the rest of what we can measure.
Pro tip The whole modern world (GPS, lasers, MRIs) is downstream of measurable quantum physics — real effects leave a working technology trail.
- 5
Classify and hold loosely
If it passes, adopt it as a useful model, not eternal truth. If it fails, keep it as belief or metaphor. Either way, drop it the moment it stops earning a return.
Pro tip Attach an implicit 'your mileage may vary' to every model you use.
Watch out Locking into a story and forcing all new evidence to fit it is how smart people end up down wrong paths.
In the wild
Williamson met a woman who said she had 'raised her frequency.' He kept asking what she meant. Each answer looped back on itself — emotions vibrate at a frequency, so she changed her frequency — until she finally said 'I believe in quantum physics.' Run through the demarcation test, the claim never produced a measurable, repeatable procedure; it was a walled-off retreat, not knowledge.
→ The claim collapsed under the 'can you measure and repeat it?' filter, revealing belief dressed as science.
A relationship author wrote that 'using the power of the quantum field, honing your emotions and intentions to create the frequency or vibration of that desire, you will draw the things you want to you.' Restated precisely, it's a specific, testable claim about drawing outcomes — and it fails measurement and repeatability, surviving only on confirmation bias when the occasional 'win' gets remembered and the losses forgotten.
→ A comically specific claim that is unmeasurable and unrepeatable, kept alive only by cognitive bias.
Common mistakes
Treating not-scientific as automatically false
Ferrie stresses that failing the test doesn't make a claim wrong — it makes it unverifiable and probably useless for prediction. Overclaiming falseness weakens your credibility.
Counting the hits, ignoring the misses
Winning $20 after doing the ritual feels like proof, but the hundred failed attempts are forgotten. Repeatability guards against this confirmation bias.
Clinging to a model after it stops paying
Even good models are only useful stories. Refusing to abandon one when evidence turns is how you get led astray.
Is it for you?
Best for
Anyone evaluating health, spiritual, or 'science-backed' claims before acting on them.
Not ideal for
Questions of taste, meaning, or values where measurability was never the point.
From the transcript
“if you can measure and perform repeatable experiments and those experiments give similar or the same results then we can say that we know something”
“unless you can measure it and you can tell someone the procedure by which you measured it and they can repeat it then it's not…”
“we should accept these lies if they give us mileage but as soon as they stop giving us a return for believing it then we…”
From the episode
7 Ways To Ruin Your Life With Lies From Quantum Physics - Chris Ferrie - #648
Chris Ferrie