# blog.jaye.ch $ cat safeguards.md▊

The Safeguards

the countermeasures: what the traditions wrote down, and why it works.

The full post

← home · the safeguards, with notes

What the traditions that survive the collapse wrote down — and why it is a safety specification, not folk wisdom

Fifth of the implications, and its resolution — arguments built on the model, as distinct from the model itself. It has been a dark run: the mechanism posts, who is at risk, the groups that engineer it, the metaphysics that steers it, the person at the top. This post is the other side of it. Not a defence of tradition — a reading of the traditions’ own codes against the mechanism the mechanism posts set out, and against what the clinical literature has independently arrived at.

Status. Where the previous posts reasoned from a model, this one is almost entirely citation: the safeguards below are written into the traditions’ own canonical material, and the clinical convergence is published. The final section’s claim — that the safeguards are the model’s requirements in another vocabulary — is the series’ argument, offered as such. This is not an endorsement of any tradition; §9 says plainly what it is not.


1. The claim

Every post so far has described a structure that fails. This one describes a structure that does not — and its central claim is that the difference is not vibes, sincerity, or the age of the lineage. It is a set of specific, written-down practices, each of which blocks one of the failure modes the series has documented.

The traditions that survive an encounter with this territory appear to survive it for reasons they can state. Not because they are old. Because they encoded countermeasures — and the ones that lost the countermeasures, or were captured by someone with an interest in removing them, are exactly the ones this series has been describing.

Here they are, cited, and mapped at the end.


2. Ethics comes first — and binds the teacher most of all

The most consistent structural feature across contemplative traditions is a claim about ordering: the ethical training is not an adjunct to the practice. It is the first limb.

  • In Buddhism the path is grouped into three trainings — ethical conduct (sīla), concentration (samādhi), and wisdom (paññā) — with ethics first, and the eightfold path’s first three limbs (speech, action, livelihood) entirely ethical.1
  • In Patanjali’s Yoga Sūtras the yamas are the first of the eight limbs — the restraints one begins with, before posture, before breath, before any state.2
  • In the monastic context this becomes a legal document: the Vinaya — a code of discipline that “operates as a moral and social mechanism ensuring transparency, accountability,” with procedures (confession, pārivāsa, reinstatement) applying to the community and its teachers.3

Read that against the series and the point is sharp. An ethics-first structure is not a moralistic flourish; it is an accountability structure, and its subject is the person with power. The empty leader of the previous post cannot function inside a code that gives his students standing to accuse him and a procedure to hear it. Ethics-first is a teacher-constraint before it is a student-discipline — which is precisely why groups that remove it are the ones where the harm concentrates.


3. The pace is metered, and no one skips

The second safeguard is a doctrine of gradualism so consistent it is almost a genre.

  • Buddhism’s gradual training (anupubba-sikkhā) is one of its oldest set pieces, and the Gaṇakamoggallāna Sutta (MN 107) is the canonical statement: the Buddha describes training as step by step, with two similes — a horse trainer schooling a thoroughbred progressively, and a known road to a destination, where the teacher can only describe the way and the traveller must walk it.4
  • Tibetan Buddhism built an entire literature on this: the lam rim (“stages of the path”) is a genre whose whole purpose is to specify what is taught when.
  • The Yoga Sūtras sequence the limbs, and pair practice with its counterweight — abhyāsa (sustained practice) and vairāgya (dispassion) — as a balance (YS I.12), not as intensity.

The model’s mechanism makes the reason obvious. The transition is a cliff at a threshold — 66.3 time-units of sustained inward drive in the model, with no gradual warning and no critical slowing down to feel. Gradual training is not caution for the timid. It is the only structure compatible with a cliff: you do not sprint at a threshold that gives no signal, and a tradition that has watched the edge for centuries does not tell beginners to.

Contrast the modern severed version directly: sit until it breaks, go deep, kill the ego. Every one of those is an instruction to remove the pacing — and pacing is the safeguard the clinical field has now rediscovered under another name (§7).


4. The community is plural, deliberately

Third, and the one that reads most clearly as an anti-capture measure.

The Buddhist formulation is startlingly direct. In the Upaḍḍha Sutta (SN 45.2), Ānanda suggests that “admirable friendship, admirable companionship, admirable camaraderie” is half the holy life. The Buddha corrects him:

“Not so, Ānanda. Admirable friendship … is actually the whole of the holy life.”5

The term is kalyāṇa-mittatā, noble friendship — and the structure it describes is plural. Not a guru, not a master: friends, and a sangha — a community with many relationships and, per the Vinaya, procedures among them.

Set that against the coercion post’s inversion and it is the same safeguard seen from the other side. There, exclusivity was the tell: a coercive group must become the only available attachment figure, because a holder that is one person is a holder that can be owned. A tradition that says the community, not the teacher, is the whole of the life has removed the single point of capture. It is the structural opposite of “the only route runs through this room.”


5. The teacher points past himself — by rule

Fourth, and the most directly targeted at the previous post’s problem: the traditions specify, in canon, that the teacher is not the thing.

Two formulations carry it:

  • The four reliances (catvāri pratisaraṇāni): rely on the teaching, not on the personality of the teacher; on the meaning, not the words; on the definitive meaning, not the provisional; on wisdom, not ordinary consciousness. The first of the four is the whole rule: the person is not the authority; the dharma is.6
  • The Kālāma Sutta (AN 3.65), the “charter of free inquiry,” lists what one should not accept on: “repeated hearing; nor upon tradition; nor upon rumor; nor upon what is in a scripture; nor upon surmise; nor upon an axiom; nor upon specious reasoning; nor upon a bias … nor upon another’s seeming ability; nor upon the consideration, ‘The monk is our teacher.’”7

Read the last clause again. The canon explicitly instructs practitioners not to accept a teaching because the teacher said so — and bundles the teacher’s personal authority into the same list as rumor and hearsay. And at the end of his life, in the Mahāparinibbāna Sutta (DN 16), the Buddha declines to name a successor: the teaching is the teacher now.8

This is the empty-leader safeguard, written down, two and a half thousand years early. The previous post’s diagnostic was follow the self-content — does the structure’s meaning, admiration, and coherence terminate at him, or past him? The four reliances and the Kālāma charter are a canonical instruction to make it terminate past him, and to treat any structure that redirects it as something to be tested rather than obeyed.


6. Discernment: a written protocol for noticing it has gone wrong

Fifth, and the piece modern practice most conspicuously lacks — a method for telling authentic movement from misleading movement, written as procedure.

  • Ignatius of Loyola, in the Spiritual Exercises (313–336), gives explicit rules for the discernment of spirits — distinguishing consolation from desolation, with concrete markers for each and, crucially, instructions about what to do in desolation. Its function is diagnostic and directional: not “push through,” but recognise the state and adjust course.9
  • John of the Cross gives the same territory a name and a stage — the dark night of the soul — and, in the tradition around it, a role: the spiritual director, a separate, limited, non-exclusive relationship whose job is accompaniment through exactly this.10
  • The Buddhist path supplies its own maps — the jhānas, the fetters, the stages — with the understanding that states are stages, not destinations, and that a teacher’s function is assessment rather than validation.

The mechanism post found that this territory gives no internal warning: the transition is a fold, the reporting faculty survives while the self does not, and the endpoint feels like arrival. A tradition facing that must supply an external discernment procedure, because it cannot rely on the practitioner’s own read. Ignatius’s rules are precisely that — a written procedure for exactly the situation the model describes, where your own sense of how it is going is the thing that has stopped being reliable.


7. The clinical field is re-deriving the same list

None of this is only ancient. The adverse-effects literature has, independently and in its own vocabulary, been arriving at the same safeguards:

  • Trauma-sensitive mindfulness (Treleaven, 2018) gives operational principles — stay inside the window of tolerance; shift attention to support stability; keep the body in mind — for the specific hazard the mechanism post describes: sustained inward attention is not neutral, and for some people it destabilises rather than settles. (This is pacing and attentional redirection, restated.)11
  • Informed consent is the recommendation of the adverse-effects research (Britton et al., 2021, and the Varieties of Contemplative Experience work): practitioners should know, in advance, that these states occur, what they look like, and what to do. (This is frame before you arrive, restated.)12
  • Spiritual emergency (Grof & Grof, 1989) argues that such crises are frequently transformative when properly held and destructive when not — the empirical version, in the psychedelic and transpersonal literature, of this series’ central claim that the same event has opposite outcomes depending on the scaffolding around it.13
  • The religious/spiritual struggles literature (Pargament; Exline; the APA’s own topic material) finds struggles to be common, normal, and outcome-dependent on context and support — not evidence of pathology and not evidence of progress.14

Four independent research programmes, four restatements of the same list. That convergence is the most encouraging fact in this series: the safeguards are not in tension with evidence — they are what the evidence keeps pointing at.


8. The mapping: the safeguards are the model’s requirements

Here is the series’ argument, stated once. Every safeguard in this post is one of the model’s required conditions, in contemplative vocabulary.

safeguard canonical source the model requirement it satisfies
Ethics first; the teacher is bound three trainings; yamas as limb 1; Vinaya removes the incentive to induce capture; accountability for the powerful
Pace metered; no skipping anupubba-sikkhā, MN 107; lam rim; abhyāsa/vairāgya the cliff has no warning — pacing is the only structure that fits a fold
Community plural, not exclusive kalyāṇa-mittatā, SN 45.2; the sangha a holder that cannot be owned; no single point of capture
The teacher points past himself four reliances; Kālāma Sutta; DN 16 the anti-empty-leader rule: self-content must not terminate at him
Discernment as procedure Ignatius, Exercises 313–336; John of the Cross; spiritual direction the state cannot read itself — an external read is required
The endpoint is not named apophatic theology; nettī netti removes the telos a frame would otherwise steer the collapse toward

Six safeguards, one function: they keep the collapse un-owned. The frame is not supplied by the person who benefits from it, the community cannot be monopolised, the pace cannot be accelerated by someone else’s urgency, the teacher cannot become the destination, the state cannot be read from inside, and the end cannot be named. Each is a closed door on one of the routes the series has shown to a bad outcome.

And that is what the traditions appear to be: not a method, but a record of the doors that needed closing — accumulated by people who kept notes on what happened when they were left open.


9. What this is not

  • Not an endorsement. This post specifies a structure, not a tradition. Traditions have harboured serious abuse — including celebrated teachers in celebrated lineages — and the safeguards above are precisely what such cases honoured in the breach. “Traditional” is not a safety property. Keeping the safeguards is.
  • Not a claim that the old ways are correct in their metaphysics. The metaphysics post argued the opposite for the quantum-mystical frame, and nothing here defends any doctrine’s truth claims. The claim is narrow: these practices are structurally protective against a documented failure mode.
  • Not a substitute for care. For a person in difficulty, the relevant instruments are clinicians and cult-aware professionals — not a reading of the canon.
  • The mapping is the argument. That the safeguards are the model’s requirements is this series’ reading; the traditions do not know about the model and the clinical field is not testing for it. That is the honest status of a claim offered, not established.

10. The test, and the point

The series opened with a mechanism and one measurement that would settle it — tolerance, not settling time. It closes with a second, and it is the practically important one:

Adverse outcomes should track the safeguards, not the technique.

Same practice, different scaffolding, different outcomes. That is a testable claim, and the existing evidence already leans its way: the same meditation-triggered phenomenology has opposite trajectories depending on framing and support (the appraisal difference in Pons et al. 2026)15; the adverse-effects research converges on informed consent and pacing; and the spiritual-emergency literature has argued the “held versus unheld” distinction for decades.

If that holds, the useful question about a practice setting is not what technique is taught but which of the six doors are closed: Is the frame given in advance? Is the community plural? Is the pace metered? Does the teacher point past himself? Is there a procedure for noticing it has gone wrong? Is the endpoint left unnamed?

None of those require belief. All of them are checkable by anyone, from outside, without expertise — which is the point of writing them down.


Notes

Marking: documented — every safeguard’s canonical source, and each item of clinical convergence. Argument — that the safeguards are the model’s required conditions restated, and that adverse outcomes track the safeguards rather than the technique. The argument is the series’ contribution, and §10 states the measurement that would settle it.

  1. The three trainings — sīla (ethical conduct), samādhi (concentration), paññā (wisdom) — as the grouping of the noble eightfold path, whose first limbs (right speech, action, livelihood) are ethical. https://en.wikipedia.org/wiki/Noble_Eightfold_Path↩︎

  2. Patanjali, Yoga Sūtras — the yamas as the first of the eight limbs (II.29); the practice/detachment balance abhyāsa–vairāgya (I.12). The yamas are explicitly the restraints one begins with, ahead of any state practice.↩︎

  3. The Vinaya — the monastic code of discipline, which “operates as a moral and social mechanism ensuring transparency, accountability, and mutual [correction],” with procedures (confession, pārivāsa) that apply to the community including its teachers. https://tricycle.org/buddhism-sangha↩︎

  4. Anupubba-sikkhā, the gradual training; the Gaṇakamoggallāna Sutta (MN 107) for the step-by-step account with the horse-trainer and road similes. https://accesstoinsight.org/tipitaka/mn/mn.107.horn — and the lam rim (“stages of the path”) genre, e.g. Tsongkhapa’s Lamrim Chenmo, whose purpose is to specify sequence.↩︎

  5. Upaḍḍha Sutta (SN 45.2): “Admirable friendship, admirable companionship, admirable camaraderie is actually the whole of the holy life.” https://accesstoinsight.org/tipitaka/sn/sn45/sn45.002.than.html — kalyāṇa-mittatā, noble friendship. https://suttacentral.net/sn45.2/en/sujato↩︎

  6. The four reliances (catvāri pratisaraṇāni): rely on the dharma, not on the instructor; on the meaning, not the words; on the definitive meaning, not the provisional; on wisdom, not ordinary consciousness. https://www.rigpawiki.org/index.php?title=Four_reliances↩︎

  7. Kālāma Sutta (AN 3.65), the “charter of free inquiry”: “Do not go upon what has been acquired by repeated hearing; nor upon tradition; nor upon rumor; nor upon what is in a scripture; nor upon surmise; nor upon an axiom; nor upon specious reasoning; nor upon a bias towards a notion that has been pondered over; nor upon another’s seeming ability; nor upon the consideration, ‘The monk is our teacher.’” https://accesstoinsight.org/lib/authors/soma/wheel008.html↩︎

  8. Mahāparinibbāna Sutta (DN 16) — the Buddha declines to appoint a successor, and directs the community to the teaching. (Standard reading: “the dharma is your teacher now.”)↩︎

  9. Ignatius of Loyola, Spiritual Exercises, 313–336 — the rules for the discernment of spirits (“consolation” and “desolation”), including instructions for acting in desolation. https://www.ignatianspirituality.com/making-good-decisions/discernment-of-spirits/introduction-to-discernment-of-spirits↩︎

  10. John of the Cross, Dark Night of the Soul — the “dark night” as a named, expected stage; and the spiritual director as the accompanying role in that tradition.↩︎

  11. Treleaven, D. (2018). Trauma-Sensitive Mindfulness: Practices for Safe and Transformative Healing. W. W. Norton. Principles include: stay inside the window of tolerance; shift attention to support stability; keep the body in mind. https://davidtreleaven.com/↩︎

  12. Britton, W. B. et al. (2021). “Defining and Measuring Meditation-Related Adverse Effects in Mindfulness-Based Programs.” Clinical Psychological Science — and the informed-consent recommendation carried across the adverse-effects literature. https://journals.sagepub.com/doi/abs/10.1177/2167702621996340↩︎

  13. Grof, C., & Grof, S. (1989). Spiritual Emergency: When Personal Transformation Becomes a Crisis. — crises as frequently transformative when properly held and destructive when not.↩︎

  14. Pargament, K. I., Exline, J. J., et al. — the religious/spiritual struggles literature: struggles are common and their outcomes depend on context and support. https://www.apa.org/topics/belief-systems-religion/spiritual-struggles↩︎

  15. Pons, E. et al. (2026). “A cross-sectional survey on depersonalization/derealization and meditation-induced alterations of the self.” Scientific Reports 16:14673 — the same phenomenology, differing appraisal. https://www.nature.com/articles/s41598-026-51014-y↩︎