Fourth of the implications — the arguments built on the model, as distinct from the model itself. The mechanism posts set out the collapse and what stops it; the coercion post describes groups that engineer it; the metaphysics post is the frame that steers it; the leader post is the person at the top. This one is the evidence — and it is the closest thing in the series to a test of something the model claims.
Status. This post presents primary-source
passages found via Source
Library, a library of AI-translated historical texts. The
translations are machine-generated and I have not checked them against
the originals; the mapping from each tradition’s terms to the model’s
features is my interpretation, not the texts’. Content
is CC-BY-SA-4.0 (attribution: Source Library). The full passage
collection with citations is at
source-library-passages.md.
On the quotations below. The passages from Source Library are reproduced under CC-BY-SA-4.0, with attribution to Source Library and a link to the source page for each. That licence and its share-alike term travel with the quoted passages; the post’s own prose is separate from them.1
The series has leaned on one assertion repeatedly, and until now it has been an assertion:
The traditions encoded the control law, even though they did not know the mechanism.
That is the claim behind the guardrails in the mechanism post, the inversion in the coercion post, and the entire safeguards post. If it is true, it should be checkable — because it makes a prediction you can go looking for:
If the traditions encoded the control law, they should have named the failure, described what it does, and prescribed what to do about it — independently, in different vocabularies, without any of them having the mechanism.
So I went looking in the primary sources. Not for mystical poetry — for the warning literature: the places where a tradition tells its own practitioners what goes wrong.
They all have one. Seven traditions, each with its own term for the failure, each with a written remedy. In every case the term is named — which is the first piece of evidence, because you only name a thing that happens often enough to need a name.
| tradition | term | approx. |
|---|---|---|
| Theravāda (Buddhaghosa) | vipassanupakkilesa — the “corruptions of insight” | ~5th c. |
| Korean Sŏn (Chinul) | 風狂 — “wind-mad” / Zen sickness | 12th c. |
| Chinese internal alchemy | 走火 — “Fire Deviation” | 1609 (in print) |
| Orthodox Hesychasm | prelest — spiritual delusion | compiled 1793 |
| Syriac asceticism | “he disguises himself as an angel of light” | medieval |
| Catholic (John of the Cross) | discretio spirituum — discernment of spirits | 16th c. |
| Tibetan Mahāmudrā | nyams vs rtogs pa — experience vs realisation | compiled, undated |
And when you read them, they describe the same five things — four of which are the features this series derived from a model.
The single most striking passage in the whole search is Theravāda, and it says something it would be very uncomfortable to invent:
In truth, the corruptions of insight do not arise to: 1. A noble disciple who has already attained the realization. 2. One who practices wrongly (beginning with a collapse of morality). 3. One who has abandoned the meditation practice. 4. An indolent person (even if they practiced correctly from the beginning). They arise only to the child of good family who practices properly, applies effort repeatedly, and has begun the practice of insight.
Not the one practising wrongly. Not the one who quit. Not the lazy one. The one doing it right, and hard.2 That is the mechanism post’s least intuitive result — exposure rising with competent effort — stated 1,500 years earlier.
And the tenth and final corruption on that list is nikanti: attachment, craving. The corruption is clinging to the states themselves. The trap named as a trap.
Chinul’s Korean Sŏn manual gives the instruction most directly:
If you fear that distracting thoughts are flying up in confusion, you absolutely must not fight with them. The more you fight them, the more intense they become. Many people at this stage do not know whether to advance or retreat, and they cannot help but become “wind-mad” — a term for mental breakdown or Zen sickness, ruining their entire lives. You should go to the place where the distracting thoughts arise and gently lay them down. Turn your body around, get down on the floor, and walk…
Three things at once: a named catastrophic outcome; the warning that effort intensifies it (the rescue post’s finding that more force is the wrong move); and a remedy that is physical and outward — get up, walk — rather than more inward effort.3
The Chinese alchemical tradition catalogued the injury and its cause. A 1609 encyclopaedia carries a woodcut illustration of the failure — three figures in agitation — under the name 走火, “Fire Deviation,” and includes the risk factor:
If a practitioner lacks proper guidance or maintains an improper focus, this energy can “deviate” from its natural pathways, leading to symptoms that range from physical pain and illness to psychological obsession or madness.
Lacks proper guidance. That is the holder requirement — the model’s finding that the collapsed state cannot initiate its own exit, so the input has to come from outside — named as the mechanism of injury, in a Ming-dynasty medical warning.4
Two traditions address the hardest part — that the state hides itself.
The Syriac manuscript states the structure plainly, quoting the tradition: “he disguises himself as an angel of light, to lead astray, if possible, even the elect.” The deceptive state presents as the light. And it does not stop there — it supplies criteria to tell them apart:
…these are the signs of the spirit of truth: first, humility, gentleness, and love… And when you see a man who says that “the Spirit is in me,” and he exalts himself, and loves honor and titles… know that he is a false prophet.
The test is not how the state feels. It is whether the structure terminates at the person or past him — which is exactly the leader post’s diagnostic.5
John of the Cross goes further, and states the model’s consequence as a rule:
Although the soul holds what she understands to be as certain and true as has been said, she still cannot neglect to follow what her spiritual teacher commands, even if it is very contrary to what she feels.
The practitioner’s own certainty cannot be trusted, and the check comes from outside. That is the mechanism post’s “sensor destroyed by the failure it detects,” written as a discipline, by someone who had clearly watched it fail.6
The finest-grained version is Tibetan. Mahāmudrā’s analysis of progress treats the signs of advancement as having two species, and repeats the distinction for every stage:
The third part, the signs, has two aspects: the signs of experience and the signs of realization. The sign of experience is the arising of an experience in the mind… The sign of realization is the realization that all phenomena are…
Experience and realization, kept apart. Not a warning — a taxonomy, maintained because the two feel identical and only one is trustworthy. The tradition is not saying “be careful”; it is saying these are different categories, and the difference is the whole game. That is the appraisal trap — mistaking the state for the goal — as a filing system.
The same text grades it explicitly: the ladder starts at mere experience —
Small One-pointedness is the mere experience of the mind abiding… Great One-pointedness is when that very experience mixes with the wisdom that realizes the essence.
Experience alone is the bottom rung. Nothing about that passage would make sense to someone who had not noticed people mistaking the rung for the top.7
What it proves — and it is more than I expected. Seven traditions, separated by language, century, continent, and metaphysics, with no contact with each other and none with this model, independently describe:
Four of those five are the model’s derived features. And it addresses this project’s biggest acknowledged weakness: the model came from one case, so agreement with that case proves little. But the phenomenon and its features are not one person’s report. They are attested across a millennium of independent practitioners who each thought it important enough to write down, name, and prescribe remedies for.
What it does not prove — and this is the honest limit. Traditions recorded features, not dynamics. No passage contains a threshold, a fold, a ratio, or a timescale — because nobody had instruments. A tradition noting that a state presents as attainment is evidence about the phenomenon; it is not evidence for the model’s description of it. The transition could be a border-collision fold, a saddle-node, or something else entirely, and every passage above would read exactly the same.
The fifth feature — the no-critical-slowing signature, the thing that says a settling-time protocol cannot see the approach — appears in no tradition, and could not have. That one requires measurement.
So: this is qualitative corroboration of the target, not validation of the model. Worth having, and worth not overclaiming.
Two things, and both are useful.
First, it converts an assertion into a documented claim. The series has repeatedly said the traditions are the precedent, not the problem. That is now checkable rather than asserted: they named the failure, described its features, and — in the next post — wrote down the countermeasures.
And second, the convergence has a shape worth noticing. Every tradition in the table independently arrived at the same response: not “practise harder,” but name it, expect it, watch for it, and have someone outside the state check. A failure whose signature is “it feels like success” can only be caught by a structure that does not rely on the practitioner’s own read — which is precisely the definition of the guardrails the mechanism post derived and the coercion post found inverted.
Seven traditions, no mechanism, same control law. That is the strongest evidence in this series that the structure being described is real — and it is the floor under everything that follows.
Marking: quoted — every passage in §3–§7 is from the translation cited. Interpretation — the mapping of each tradition’s term to a model feature, and the claim that the five features converge. Not claimed — that any tradition validates the dynamics, thresholds, or transition type.
Passages were located through Source
Library (https://sourcelibrary.org), a library of AI-translated
historical texts, using its public API. Translations are AI-generated
and unchecked against the originals; the source language, date and page
are given with each passage in source-library-passages.md
so any of them can be verified. Source Library content is
CC-BY-SA-4.0 (attribution: Source Library); that
licence’s share-alike term attaches to reuse of the translations.↩︎
Buddhaghosa, Visuddhimagga, ch. 20 — Thai edition, p. 1111. https://sourcelibrary.org/book/visuddhimagga-path-of-purification-in-thai↩︎
Chinul (1158–1210), Susimgyŏl, chinsim chikkyŏl, p. 86 (1700 ed.). https://sourcelibrary.org/book/susimgyol-chinsim-chikkyol-chinul↩︎
Wang Qi & Wang Siyi, 三才圖會 (Sancai Tuhui), juan 74, p. 93 (1609). https://sourcelibrary.org/book/sancai-tuhui-juan-74↩︎
Syriac Theological Manuscript (Codex Vaticanus Syriacus 162), p. 41 (medieval). https://sourcelibrary.org/book/syriac-theological-manuscript-codex-vaticanus-syriacus-162-anonymous↩︎
Juan de la Cruz (John of the Cross), Die geistliche Bücher und Schrifften (German translation of the Ascent of Mount Carmel), p. 324 (1697). https://sourcelibrary.org/book/die-geistliche-bucher-und-schrifften-cruz↩︎
Thor bu Phyag chen rnal ’byor bzhi bshad (Four Yogas of Mahāmudrā), Tshamdrak Monastery, Tibetan, pp. 75, 93–94. https://sourcelibrary.org/book/thor-bu-phyag-chen-rnal-byor-bzhi-bshad-monastery↩︎