The method is described as ordinary among a particular class of practitioner, and one instance is given in enough detail to be a protocol.
"Spiritual masters — and most often those without teachers themselves — often test cultivation practices that have been written about in ancient texts to verify them like this. My master, as an example, would practice all sorts of samadhi just to see the end result and compare it with ancient scriptures. For instance, at one time he cultivated the samadhi of non-knowing ignorance for a few weeks just to see what the effect really would be."1
The structure is a self-experiment and every element of it is specified.
A hypothesis from a text — the scriptures say this samadhi produces that result.
A treatment — cultivate it deliberately.
A duration — a few weeks, long enough to produce an effect and short enough to be reversible.
And a comparison — the observed end result set against what the scriptures claim.
Just to see what the effect really would be. No devotional motive, no expectation of benefit. The samadhi of non-knowing ignorance is not something a practitioner would want; it is being run because a text describes it and the description has not been checked.
Which is a strange and revealing choice of example. The technique tested is one nobody would pursue for its own sake — which removes the confound that ruins most self-experimentation in this territory, where the practitioner wants the result they are testing for. [n=1 REPORTED].
The qualifier in the first sentence is not decorative and it identifies who does this.
Most often those without teachers themselves.
Which makes the method a substitute rather than a preference. A practitioner with a living teacher does not need to run the experiment — they ask, and someone who has been through it answers. The text's claim gets checked against a person.
Without a teacher, the only available instrument is your own body. The text says something; nobody present can confirm or deny it; so you run it.
Three consequences follow.
The method is forced rather than chosen. It is what you do when the correction mechanism is missing, and its prevalence in a tradition is a measure of how few teachers that tradition has.
It is slow. A few weeks per claim, one at a time, in a literature containing thousands of claims. A teacher answers in a sentence what a self-experiment answers in a month.
And it is how a lineage restarts. A practitioner who tests the old texts and finds which parts hold is reconstructing, alone, what a transmission would have carried — which is presumably how anyone begins after a break. See The Dakini Who Invented a Method.
The reason for testing is stated as a claim about the inherited material.
"Great masters often test ancient teachings like this to see what is right and what is wrong because not everything passed down from the past, especially in cultivation matters, is valid… You cannot trust all the reports in ancient books but have to sieve the information through wisdom, personal experience and self-authentication."2
Not everything is valid — which is a moderate claim and the right one. Not that the tradition is worthless; that it contains errors, and errors do not announce themselves.
Especially in cultivation matters is the sharpening. Doctrine can be argued about; a technique's claims are empirical and were never checked by most of the people who transmitted them.
The warrant offered is an attribution: "Shakyamuni Buddha even said so himself, saying 'Don't believe me but just test everything I say, and then decide.'"2 🚩 SECONDARY WITHOUT PRIMARY — a widely-circulated paraphrase with no text behind it.
And the consequence is stated with unusual bluntness: "If most religions went by that scoring system then countless cherished religious dogmas would certainly fail the test."2
Scoring system is the operative phrase. The proposal is that claims be scored — kept or discarded on results — and the prediction is that a great deal would not survive.
Which is the same standard applied to techniques throughout this material, extended to doctrine. A claim about what a practice does to a body is checkable, and most of what is transmitted has never been checked by anyone in the chain. See Unverified Dogma Collects Dust.
The comparison used twice is a diet, and it is doing more work than an illustration.
"It is just like Mahatma Gandhi's practice of trying out different diets to see their effect on his body."2 And: "It is like selecting a diet that is personally best for you. The fact is that many different types of diet abound, but because of genetic individuality and your own personal habits you'll have to test them to determine which one will best work for you."4
The analogy makes a specific claim: the answer is person-dependent.
Genetic individuality and personal habits — two sources of variation, and they mean there is no single correct diet and therefore no single correct practice. The literature can tell you what the candidates are; it cannot tell you which one your body will respond to.
Which is a considerably stronger position than test what you read. It says the testing cannot be done once by anyone and transmitted, because the result does not generalise. Every practitioner has to run it again.
And it explains why so much conflicting advice survives. If effectiveness varies by constitution, then two masters reporting opposite results about the same technique may both be reporting accurately — and a tradition that accumulates both is not confused; it is recording a genuine variance.
The complementary claim: "If you don't try things out for yourself, sometimes you just cannot know the ins or outs of a particular samadhi or cultivation technique and what it will actually do for you."4 What it will do for you — the qualifier is the whole point. See Try a Method for One to Three Months.
The safety limit is stated in the sharpest sentence in this material, and it is the case study because of what it rules out.
"Bodhisattvas who practice medicine often try herbs out on themselves to see their effects, as another example, and in this way they can learn what the herbs do and which ones can help people in which manner. You must always be careful with such techniques and employ your wisdom to the maximum because many individuals in the past who tested medicinal substances on themselves have died because of accidental poisoning. Just because you have samadhi does not mean you will not die from poison!"5
Just because you have samadhi does not mean you will not die from poison.
That sentence does something almost nothing else in this material does: it names a domain in which attainment confers no advantage whatsoever.
Everywhere else, attainment generalises. Purified channels produce health, effectiveness, wisdom, insight into policy, control of the elements, an incorruptible body. Here it stops dead at toxicology. A master with genuine samadhi, testing an unknown herb, is exactly as poisonable as anyone else.
And the historical claim behind it is grim and specific: many individuals in the past who tested substances on themselves have died. Not been harmed — died.
Which sets the real boundary on the whole self-experimentation method. It works because the practitioner is the instrument, and instruments break. A few weeks of an unpleasant samadhi is recoverable. An herb is not necessarily.
The distinction the passage draws is between reversible and irreversible tests, without naming it as such — and it is the distinction that separates a method from recklessness.
The framing that closes it: "Cultivation is all about science and wisdom. It is not about stupidity or superstition or blindly following others and tossing logic or safety out the window."5 Safety is on the list, alongside logic, as a thing the enterprise does not discard. [n=0 NAMED] on the deaths.
Run the method where it is cheap and refuse it where it is not.
Test techniques, not substances. The reversibility line is the operative one. A practice you can stop is testable; a thing you ingest may not be.
Give each test a fixed duration decided in advance. A few weeks is the model — long enough to produce an effect, short enough that you stop whether or not you have an answer. A test with no end date is an adoption.
Test one thing at a time. The whole logic depends on attributing an effect to a treatment, and two simultaneous changes attribute to neither.
Write down what the text predicts before you start. The comparison is against what the scriptures claim, and a prediction recorded afterwards is not a prediction. This is the single cheapest control available and it costs one line. See The Proximate Attainment Diary.
Prefer to test the techniques you do not want. The master's example is the samadhi of non-knowing ignorance — a state nobody would seek. Testing something you are hoping for is testing under the worst possible conditions, because the wanting supplies the result.
And expect the answer not to transfer. Constitution and habits vary; a result is about you. Which means your finding is not a correction to the literature — it is a finding about your own body, and reporting it as more than that is how the literature accumulated its errors in the first place.
This sits in cross-domain because the method cannot be understood without an epistemology of testing and a physiology of practice simultaneously — the claim is that a doctrinal proposition is checkable by running it in a body, so an account in either vocabulary alone describes half of it.
Take either half away and it collapses.
Epistemology alone gives the testing norm — sieve the reports, score the dogmas, believe nothing untested. But it has no account of what the test is, because ordinary verification has no procedure for checking a claim about a samadhi. There is no experiment, only the practitioner.
Physiology alone gives the practice — cultivate this state, observe what happens. But nothing tells you why you would run one you do not want, or what the observation is for. The comparison against the text is what makes it a test rather than a practice.
The bridge is the claim that a scripture's assertion is a hypothesis about a body, and that a body can therefore adjudicate it.
Which is what makes this more than a slogan about scepticism. It produces a specific and unusual research programme: run the states nobody wants, for fixed periods, against recorded predictions, in a subject whose constitution makes the answer local — and accept that the result does not generalise, so the programme never terminates and never accumulates.
That is a real methodology with a real defect, and neither domain identifies the defect alone.
The framing repeated throughout deserves examining, because it is doing something the surrounding material does not always earn.
"Cultivation is all about science and wisdom. It is not about stupidity or superstition or blindly following others and tossing logic or safety out the window."5
Four things rejected, and they are not the same kind of thing.
Stupidity and superstition — failures of the reasoner. Blindly following others — a failure of the relationship to authority. And tossing safety out the window — which is not an epistemic failure at all. It is a practical one, and its inclusion is the honest part.
Most appeals to science in this kind of material mean my claims are true. This one is being used to mean something narrower and more defensible: proceed by test, keep what survives, and do not injure yourself.
Which is a workable standard, and it is worth noting where this material meets it and where it does not. The technique claims are mostly of the right kind — checkable, with stated durations and markers. The historical and comparative claims mostly are not, and they are asserted in the same voice.
The distinction the framing implies and the material does not draw: a claim you can test on yourself, and a claim about what most traditions did. Only the first is subject to the scoring system being proposed, and the second is where the flags cluster. See Be a Lion, Not a Sheep.
Evidence. [PRACTITIONER SOURCE]. 🚩 SECONDARY WITHOUT PRIMARY — the Buddha quotation is unlocated, Gandhi's dietary experiments are invoked without a source, and no scripture is named for any tested samadhi. 🚩 SINGLE SOURCE. [SPECULATIVE]. The strongest item is the specific case: a named technique, a stated duration, a stated purpose.
Tension — test everything, and only a master can tell. The self-authentication standard is proposed as the sieve. The same material holds that practitioners cannot distinguish projections from perception, misread markers routinely, and that only a master knows for sure. The instrument being recommended is one the material elsewhere calls unreliable.
Tension — results are person-dependent, and the stages are universal. The diet analogy makes effectiveness vary by constitution and habit. The stage sequence is elsewhere said to be identical for everyone, like puberty. If the destination is universal and the routes are constitutional, that is coherent — and it is never stated, so the two claims sit unreconciled.
Tension — masters without teachers, and the teacher is decisive. The method is attributed to those without teachers, and the surrounding material treats finding a great teacher as the thing that ended years of waste.
Open question — what happens to a failed test? Nothing describes a case where a scripture's claim was checked and found false, or what was done about it.
Open question — how is an effect attributed? A few weeks of a samadhi produces changes; so does a few weeks of anything. No control is described and none is proposed.
Two positions on inherited material run through this material and they are both held at full strength.
Test it; most of it would fail. Not everything passed down is valid; sieve it through wisdom, experience and self-authentication; countless cherished dogmas would fail the scoring system. The old material is a hypothesis set.
And use the proven ones; do not invent. Use the tantric visualisations that have worked for countless people; start with what has the most success stories rather than exotic methods; the search for novelty is a major cause of failure. The old material is accumulated evidence.
These are not contradictory in principle — one is about evaluating, the other about selecting — but they pull opposite ways for a practitioner deciding what to do next. Test the unusual claim, or run the standard practice? Both are recommended, and the answer depends on whether you are a master reconstructing a lineage or a beginner accumulating years.
The convergence is on what neither permits: continuing a practice because it is traditional. Under one you must have tested it; under the other you must know it has worked for large numbers. Both require evidence, from different sources, and both rule out the same default.
Which locates the real distinction. The testing method is for people who have run out of things to do — no teacher, standard practices exhausted, and a literature full of unchecked claims. It is not an alternative to the standard programme; it is what comes after it. See Semblance Dharma and Ramakrishna Misread as Homoerotic.
The Reverse-Engineering Method — business.
The self-experiment method has a hypothesis, a treatment, a duration and a comparison — and no instrument. The craft version of the same problem built one, and the contrast identifies exactly what is missing.
The reverse-engineering apparatus is a seven-part procedure for taking someone else's essay apart: which archetype the author occupies paragraph by paragraph, where tension builds and releases, what emotional sequence the reader is moved through, how credibility accumulates, and how each paragraph hands off to the next. Six of the seven are analytical instruments. And it opens with a preparation ritual — a fixed procedure performed before analysis begins.
That is the same project: a finished artefact, received from someone else, taken apart to find out what actually does the work.
Three things fall out.
The craft version disassembles; the cultivation version only runs. A practitioner who cultivates the samadhi of non-knowing ignorance for a few weeks and compares the end result to a scripture has performed one comparison at one level of resolution — did it match, roughly. The essay method breaks the artefact into six named dimensions and checks each separately. Which means a failed match tells the craft analyst where it failed and tells the practitioner only that it did.
The preparation ritual is the missing control. It exists because analysis contaminated by the analyst's expectations is worthless, and a fixed procedure performed first is the cheapest guard against it. The self-experiment has no equivalent — nothing performed before the test that would prevent the practitioner finding what the scripture told them to look for. And this material's own diagnosis is that expectations supply experience.
And the craft field can accumulate because its findings are about the artefact. Six dimensions, applied to any essay, yielding results that transfer. The diet analogy says cultivation findings do not transfer — they are about this body — which means every practitioner starts from zero and the literature never improves.
The insight neither field produces alone: self-experimentation without an analytical apparatus is not a weaker version of testing. It is a different activity. What the craft method has is not more rigour but decomposition — the artefact broken into parts that can each be checked. A samadhi tested as a whole against a description of a whole is a single subjective comparison, and no amount of care makes one of those into evidence. Which is why the method's own strongest example is the one where the practitioner wanted nothing: removing the expectation was the only control available.
The unresolved thing is whether the sieve can operate at all, on this material's own account of the sieve.
The proposal is a scoring system: run the claim, compare the result, keep or discard. And the three filters named are wisdom, personal experience and self-authentication.
Every one of them is undermined elsewhere in the same body of work.
Personal experience is described as systematically unreliable — practitioners mistake sensations for attainments, states are manufactured by expectation, and the semblance dharma is people finding descriptions in ancient books and matching themselves to them. Which is precisely the operation the sieve consists of.
Self-authentication is stated to be beyond ordinary practitioners: only a master will know for sure, and regular human beings cannot distinguish a projection from genuine perception.
And wisdom is the least specified of the three and is available only in proportion to attainment — which is the thing being established.
So the sieve is proposed for people without teachers, using instruments the material says need a teacher to calibrate.
There is a version that works, and it is narrower: the sieve is reliable for gross results and unreliable for fine ones. Did the samadhi produce the dramatic effect the text claimed, or did nothing much happen — that is answerable by anyone. Whether a subtle state matched a subtle description is not.
Which would mean the method can only falsify strong claims, and the strong claims are a small part of what any tradition transmits. Countless cherished dogmas failing the test presumes a discriminating instrument, and the instrument described can only detect the absence of something large.
That is still worth having. It is not the scoring system being advertised, and nothing here distinguishes them.