The Walls Move
Consciousness science and the oldest maps of the inner world are converging on one modest, enormous claim.
The traditions that mapped the inner world spent ten thousand years on the questions our technology has been asking for ten.
On a June evening in 2023, in a lecture hall in New York, a neuroscientist walked onstage carrying a case of wine and handed it to a philosopher, and several hundred consciousness researchers applauded.
The wine settled a wager twenty-five years old. In 1998, in a Bremen bar after the banquet of the field's own annual meeting, Christof Koch — then at Caltech, half of the partnership with Francis Crick that had put the neural correlates of consciousness on science's respectable agenda — bet David Chalmers a case of fine wine that the signature would be found within twenty-five years: a small set of neurons, a short list of properties, clear by 2023. It was the confident bet of a confident era, and twenty-five years was a long time. By June 2023 the tools had improved beyond anything either man imagined in that bar, and the answer had not arrived. "It's clear that things are not clear," Chalmers said from the stage — the original terms had turned on that word — and Koch went off and reappeared carrying six bottles of 1978 Madeira, grimacing, to laughter and applause. And then — this is the part to keep — he bet again.
Fields do not usually stage their own losses. Science is supposed to be the practice of arranging your beliefs so the world can correct them, and mostly the correcting happens in private, or slowly, or never; careers are long and concessions are rare. Here was a man renting a stage for his. Hold that scene, because this essay is about instruments for being corrected — and about who built them first.
Start with what the twenty-five years actually bought, because it was not nothing. The search did not find the seat of consciousness. What it produced instead — quietly, across thousands of unglamorous studies — was a redescription of perception itself — one that has moved from heresy to mainstream, with its own textbooks and review literature now, though whether it is a discovery or a description too flexible to fail remains an open fight the field conducts in print. Even its status is argued the right way.
The idea is old enough to have a nineteenth-century pedigree. In 1867 Hermann von Helmholtz, measuring what the eye actually delivers, concluded that it could not possibly account for what we see: the data are too poor, too gappy, too ambiguous for the rich stable world in experience. Perception, he argued, must be inference — the brain's best guess, constructed so fast and so far below awareness that the guessing is invisible. The modern version keeps his conclusion and adds the machinery: the brain is a prediction engine. The model of the world flows outward, from expectation toward the senses; the senses flow inward carrying mainly the errors — the places where the world disagreed. What you experience is not the world and not the raw data. It is the model, corrected on the fly.
The idea's unnerving modern name is controlled hallucination, and the neuroscientist Anil Seth, who made it famous, is careful not to claim it: he first heard the phrase from an older colleague, and his own footnote traces it to a 1990s seminar and no further — attribution hygiene, on the sentence he is known by, of exactly the kind this essay turns out to be about. Perception, on this account, is hallucinating continuously, with the hallucination disciplined by the senses rather than replaced by them. When the discipline slips — in dreams, on psychedelics, in his laboratory's machines built to slip it — the construction shows. When it holds, we mostly agree, and, as Seth puts it, "when we agree about our hallucinations, we call that reality." The agreement is not evidence that we see the world directly. It is evidence that we run similar models.
If that sounds like philosophy, it can be demonstrated on a tabletop. In 1998 — the same year as the bet — two researchers in Pittsburgh sat volunteers down with one hand hidden behind a screen and a life-sized rubber hand on the table in front of them, and stroked both with paintbrushes in synchrony. Eight of the ten reached for ownership language without being asked; one reported looking at the dummy and "thinking it was actually my own." Later laboratories put a clock on the feeling — it arrives, on average, in under half a minute — and found that when the rubber hand is threatened, the brain of a person who has claimed it responds with measurable anxiety, as for its own flesh. Body ownership, the most intimate certainty there is, turned out to be an inference — a running guess about which objects the model should claim — and the guess can be edited with two paintbrushes. The model is not a metaphor.
Now follow the thought one step further, because this is where it stops being a laboratory curiosity. If perception is a running guess, then the guess has habits. The priors that shape it are not chosen: your childhood, your language, your diagnosis, your worst year — all of it is load-bearing structure in the model, deciding what the guess expects before the evidence arrives. These are not tinted glasses that a sufficiently honest person could take off. They are the walls of the room you see from. And from inside, the walls are invisible for the simplest possible reason: they are not in your visual field. They are your visual field.
That thought has begun to organize serious money. Elizabeth Koch — a writer and publisher who has been unusually public about her own lifelong anxiety — built a foundation on exactly this premise: that each of us lives inside what she calls a Perception Box, and that its walls can be expanded — that some real fraction of human suffering is not circumstance but architecture. Her foundation funds open rounds of neuroscience on precisely the unglamorous questions — anxiety, chronic pain, emotion regulation, identity and the stories a self tells about itself — forty-five projects across four cycles since 2022, the latest round nine million dollars across eleven, chosen from nearly three hundred submissions by double-blind external review. Its chief scientist is Christof Koch, the man with the wine, who is no relation to the founder — a coincidence both of them are presumably tired of explaining. The money is Koch money in the older sense, though: she is Charles Koch's daughter, and the critique has already been written, more than once — a fortune entangled with the politics of keeping circumstances as they are, now funding a science that studies adjustment to them. It is a fair thing to weigh, and this essay reports it rather than referees it. What is worth keeping from the arrangement is the premise itself, which is neither wellness talk nor resignation: not your suffering is an illusion, and not your suffering is the world's fault, but the third thing — some of it is built into the walls, and walls can be renovated.
An essay in this series audits what it admires, so: how solid is the science this hope leans on?
The honest answer is that the field is young, loud, and fights in public — and that the fighting is the best thing about it. For decades, the two leading grand theories of consciousness ran in separate literatures making incompatible claims. Integrated information theory, Giulio Tononi's mathematical account, holds that consciousness is a graded, intrinsic property of certain kinds of causal structure. Global workspace theory, championed experimentally by Stanislas Dehaene, holds that consciousness is broadcast — a moment of content winning access to a brain-wide stage. Each camp had its own experiments, its own journals, its own confirmations. Then the field did something rare: it arranged an adversarial collaboration. The rivals agreed in advance on experiments that could wound either theory, registered their predictions before the data existed, and had the work run by laboratories with no stake in the outcome, results to be published whichever way they fell.
The results arrived in 2025, and both theories took wounds. The prefrontal "ignition" that workspace theory predicted at a stimulus's offset did not appear; the sustained synchrony in the back of the brain that integrated information theory predicted did not appear either. Neither camp folded — theories do not die of one wound, each had other predictions that held, and both have published their answers — but the misses were preregistered as misses, and the record now shows them. Meanwhile the same dispute turned uglier elsewhere: a hundred and twenty-four researchers signed an open letter calling integrated information theory pseudoscience — and researchers with no stake in the theory called the label inflammatory, Chalmers likened the letter to "dropping a nuclear bomb over a regional dispute," and the theory's authors answered in print that the real crisis lay in their accusers' paradigm. An audit does not need to pick a side to report the shape of the thing: a young science metabolizing its disagreements where everyone can watch.
And that shape is the finding. Twenty-five years in, the most transferable product of consciousness science is not a theory of consciousness. It is a set of procedures — the preregistered prediction, the adversarial collaboration, the public bet paid onstage — instruments for arranging your own commitments so that the world can correct them. Instruments, in other words, pointed at the experimenters' own boxes. Which raises the question this series keeps arriving at from different directions: who else built instruments like that, and how long ago?
Georges Dreyfus spent fifteen years inside the Tibetan monastic universities re-established in exile in India, and became the first Westerner to earn the Geshe Lharampa degree, the highest of the four ranks; the institution at the center of his account is not the meditation hall. It is the debate courtyard. Daily, for years — ten hours a day in the seats of old, by the book's account — monks pair off into assigned roles: a challenger on his feet, punctuating each question with the debate's famous clap; a defender seated and accountable, restricted to a menu of three allowable answers, hounded with why, why, why the moment he hesitates. The curriculum is not learn the doctrine and repeat it. It is watch your certainty fail in public, daily, until you can feel it beginning to fail before it does. And Dreyfus, no romantic, is exact about what the courtyard is not: a game as much as a fight, cooperative as well as combative, and a genuine though limited realm of inquiry — there are conclusions a monk cannot debate his way out of, and the system produces, he reports, an enormous conformity alongside its brilliance. Which gives it the same shape as the laboratories: a real procedure for catching conviction in the act, run inside walls of its own. Preregistration in maroon, run daily for centuries.
The mapping went as fine as the testing. The Abhidhamma — Buddhism's oldest systematic psychology — reads less like scripture than like a working taxonomy: consciousness sorted into eighty-nine classes, a hundred and twenty-one in the expanded scheme, by root, by feeling, by object; every moment of experience parsed into constituents. Whether those charts record meditative observation or scholastic system-building is itself a live argument among scholars, and the honest reading keeps both. What is beyond dispute is that the rival schools disputed the taxonomies themselves, category by category, for centuries — the Theravādins counting eighty-two constituents of experience where the Sarvāstivādins counted seventy-five, each side keeping formal books of the points of controversy. The inner world treated as chartable, and the charts treated as contestable: that is the pair that matters.
And the charting was never for its own sake. The tradition's most famous diagnostic comes from a short discourse in the Pali canon: when pain strikes an untrained person, it says, they are struck as if by a dart — and then, hard upon the first, by a second: the grief, the resistance, the casting about for any pleasure that will drown it. The instructed person is struck by one dart. The first belongs to the world. The second is manufactured inside the model — and the discourse lays the difference not on grace or luck but on instruction. Two and a half millennia before anyone imaged a brain, that is a precise, falsifiable claim about which fraction of suffering is architecture.
The traditions that mapped the inner world spent ten thousand years on the questions our technology has been asking for ten.
The convergence is not that the monks anticipated the neuroscience, which is the kind of claim that flatters everyone and informs no one. It is narrower and stranger than that. Both efforts concluded that seeing is constructed. Both discovered that the construction is invisible from inside. And both decided that the answer to an invisible construction is not a doctrine but a procedure — the debate, the taxonomy, the adversarial test, the bet — some repeatable arrangement that catches the model in the act.
An audit keeps its own books, so two fences before the close. The meeting point of these lineages — contemplative science — has had its hype audited from inside: a 2018 review by fifteen researchers, several of them contemplative researchers and practitioners themselves, found the clinical literature young, its definitions loose, its control conditions weak, its adverse effects under-reported; the corrective work since then has been the field growing up in public, which is the same good sign wearing different clothes. And ten thousand years of mapping produces mythologies along with maps; the traditions' claims have to be checked one at a time, like anyone's. What this essay carries is the checked minimum.
But the checked minimum is enormous. You have never seen the world — only your model of it, corrected on the fly. The model's walls were built without your consent, by your species, your language, your history, your worst year. The walls are load-bearing; there is no seeing without them, and no one has ever lived outside a box. And the walls move — not by wishing, and not by being told to think positive, but by procedure: training, repetition, company, the right kind of opposition. Every line of that paragraph is now defensible from both ledgers, the one written in journals and the one written in courtyards, and neither ledger got there by trusting itself.
Koch, for what it is worth, spent a career as the field's most cheerful mechanist, drifted for two decades, and then, in his sixties, published a book occasioned by an experience that briefly took the walls down altogether; what changed most, he has said, was not the science but his fear of death, which did not come back. His second wager with Chalmers runs to 2048. Neither man behaves like someone waiting to be proven right. They behave like men who have found a procedure for being wrong in public and remaining friends — which may be as good a picture of an expanded box as the field has yet produced.
The wager underneath the wager is much older than either of them: that a mind can be shown its own walls, and that the showing moves them. Every tradition that mapped the inner world staked itself on it. Ten thousand years of practice says it pays. Twenty-five more seems a reasonable time to check.
— OUROS Field Researchedited by Pablo Quiroga
sources
- The Koch–Chalmers wager, verified against the contemporaneous record and the 1998 tape: made in June 1998 in a Bremen bar after the banquet of ASSC 2 — the field's own meeting, which is why the 2023 settlement at ASSC 26 closes a circle — with the terms recovered from a recording made the next day (Per Snaprud's New Scientist account, 23 June 2018): a small set of neurons with a small list of intrinsic properties, "clear in 25 years"; the stake a case of fine wine. Settled 23 June 2023 (draft 01 said the 25th — corrected) at ASSC 26, hosted by NYU, at the "25 Years of Consciousness" evening session where the Cogitate results were presented; Chalmers's onstage line, quoted in the body, turned on the original terms' word "clear"; per Chalmers's own count the payment was six bottles of 1978 Madeira (Madeira is Portuguese wine — the press accounts agree, they do not compete), handed over to laughter and applause, Koch grimacing; the renewed bet runs to 2048 on the same terms. Reported in M. Lenharo, "Decades-long bet on consciousness ends — and it's philosopher 1, neuroscientist 0," Nature News, 24 June 2023 (in print at Nature 619: 14–15), with John Horgan's eyewitness account in Scientific American. Nobody disputes who won. Corrected by this sweep: Crick was at the Salk Institute, never Caltech — the founding NCC paper (Crick and Koch, Seminars in the Neurosciences 2, 1990: 263–75) is a cross-institution byline, and the body no longer places the partnership at Caltech.
- Cogitate Consortium et al., "Adversarial testing of global neuronal workspace and integrated information theories of consciousness," Nature 642 (2025): 133–42, doi:10.1038/s41586-025-08888-1, published 30 April 2025 (draft 01 said vol. 640 — corrected). Design verified: predictions preregistered jointly with the theory proponents before data collection; run by theory-neutral teams; 256 participants across fMRI (120), MEG (102) and intracranial recordings (34). The two failures the body names are the paper's own headline challenges — none of the 655 prefrontal electrodes showed GNWT's predicted onset-and-offset ignition profile, and IIT's predicted sustained content-specific posterior synchrony did not materialize — and the body also now says each theory kept predictions that held (prefrontal category decoding for GNWT; posterior decoding of conscious content for IIT), which the paper reports. Funder confirmed: Templeton World Charity Foundation (its ~$30M Accelerating Research on Consciousness program; Tiny Blue Dot appears only as Koch's affiliation — the two must never be conflated). Both camps answered on the record: the GNWT proponents' formal response is Sergent, Naccache, Dehaene et al., Neuroscience of Consciousness (2025), arguing offset ignition presupposes attention to the offset; the IIT camp's supplement attributes the synchrony null partly to sparse posterior coverage. No correction to the paper itself as of September 2026; the consortium released its data publicly. Dehaene vetted clean; Baars (not named in the body) clean, though he signed the 2023 letter — an adjacency noted for the interview file.
- The letter, exactly: IIT-Concerned (Fleming, Frith, Lau, LeDoux et al.), "The Integrated Information Theory of Consciousness as Pseudoscience," PsyArXiv, 15 September 2023, doi:10.31234/osf.io/zsr78 — 124 named signatories (Dennett, Bengio, Churchland among them), organized around Hakwan Lau. Its argument, in the letter's own order: the trigger was pre-peer-review media celebration of the COGITATE results; the charge is that IIT's core claims — as opposed to hand-picked auxiliary predictions — are untestable, with the theory's panpsychism-adjacent implications offered as evidence; the stakes named are coma care, organoids, AI and abortion. The peer-reviewed descendant appeared under a different title and restructured authorship: "What makes a theory of consciousness unscientific?," Nature Neuroscience 28 (2025): 689–93. The replies, same issue: Tononi et al., "Consciousness or pseudo-consciousness? A clash of two paradigms" (694–702), which counter-charges that the accusers' computational-functionalist paradigm is the one in crisis; and Gomez-Marin and Seth's non-partisan assessment (703–06). The 2023 reactions the body draws on: Seth (no IIT partisan) calling the label "inflammatory"; Chalmers's "dropping a nuclear bomb over a regional dispute"; Mudrik — whose consortium had just falsified an IIT prediction — saying the letter damaged the field. Draft 01's "the counterblast argued the letter itself was the unscientific act" was corrected by this sweep as an overstatement: the dominant published rebuttals called the label wrong and harmful, not the letter unscientific, and the body now attributes the counter-charge to the theory's own authors, where it belongs.
- Anil Seth, Being You: A New Science of Consciousness (London: Faber & Faber / New York: Dutton, 2021) — both editions verified. The provenance, from the book itself (p. 87): "controlled hallucination, a phrase I first heard from the British psychologist Chris Frith," with Seth's own footnote tracing the origin to a 1990s seminar by Ramesh Jain — "I have tried to trace it back further, but without success"; the print anchor for the Jain attribution is Rick Grush's BBS 2004 note ("I owe this phrase to Ramesh Jain, who produced it during a talk at UCSD"). So the body credits Seth with making it famous and shows him declining to claim it — the verified fact, and the better one. The TED 2017 sentence is quoted exactly: "when we agree about our hallucinations, we call that reality" (not "call it"). Corrected: "delirium" was not among his cases — his are dreams, psychedelics, and his lab's VR "hallucination machine," which the body now follows. The framing is contested in print, carried here: Baggini's otherwise-admiring WSJ review worried "hallucination" invites the misreading that we have no grip on reality (a misreading Seth answers in the book directly), and Evan Thompson's "Reality is not a controlled hallucination" (IAI, 2026) argues the framing is a category error. Seth vetted clean, with a strengthener already used in source 6: he co-authored the strongest critique of the rubber-hand paradigm from inside his own group.
- Hermann von Helmholtz, Handbuch der physiologischen Optik (Leipzig: Voss, 1867) — the parts appeared 1856, 1860 and 1866, the complete volume in 1867, the standard citation; the doctrine sits in Part III, §26, "Concerning the Perceptions in General," and "unconscious inference" (unbewusster Schluss) is his own coinage, which he defended in the text against critics rather than retracting. The body's compression — Parts I–II measure what the eye delivers; Part III concludes perception must be inference — follows the structure of the work. (The famous quip about returning a defective optical instrument to its maker belongs to his 1868 popular lectures, not this book, and is not used.) The modern synthesis: A. Clark, "Whatever next? Predictive brains, situated agents, and the future of cognitive science," Behavioral and Brain Sciences 36 (2013): 181–204, doi:10.1017/S0140525X12000477. The body's original "hardened into textbook" was corrected by this sweep: the framework has a dedicated textbook (Parr, Pezzulo and Friston, Active Inference, MIT Press 2022) and a large review literature, and it also has serious standing critiques — Litwin and Miłkowski, Cognitive Science 44 (2020) (unification by fiat, just-so stories); Bowers and Davis, Psychological Bulletin 138 (2012); Anderson and Chemero's "brain GUTs" commentary in the same BBS issue; Walsh et al., Annals of the NYAS 1464 (2020), a review of the neurophysiological evidence co-authored by Clark himself because the evidential question is real; Furutachi and Hofer, Annual Review of Neuroscience 49 (2026). The body now says the fight is open, which is the truth and also the essay's point. Helmholtz and Clark both vetted clean.
- M. Botvinick and J. Cohen, "Rubber hands 'feel' touch that eyes see," Nature 391 (19 February 1998): 756, doi:10.1038/35784 — verified against the printed page. The affiliations are the University of Pittsburgh and Carnegie Mellon jointly, so "two researchers in Pittsburgh" is the exact honest phrasing. Ten subjects, ten minutes of synchronized stroking; the paper measured no onset time — eight of ten spontaneously used ownership language, and the quoted phrase is a subject's own from the paper. The onset clock belongs to the follow-up literature (means of roughly 11–14 seconds for ownership; Lloyd 2007 for touch-referral), and the threat findings are later work too: K. C. Armel and V. S. Ramachandran, Proc. R. Soc. B 270 (2003), skin-conductance responses to "injuring" the adopted hand, and H. H. Ehrsson et al., "Threatening a rubber hand that you feel is yours elicits a cortical anxiety response," PNAS 104 (2007) — the body attributes both to "later laboratories" accordingly. The standing critique, carried rather than hidden: P. Lush, "Demand Characteristics Confound the Rubber Hand Illusion," Collabra: Psychology 6 (2020): 22, and Lush et al., Nature Communications 11 (2020): 4853 — the reports may partly reflect expectation and trait suggestibility; Ehrsson's group rebutted in Nature Communications (2022) and the dispute is live, with the mainstream position still multisensory integration. Worth savoring: the Lush Nature Communications critique is co-authored by Anil Seth — the man whose framing the illusion is always cited to support, attacking a load-bearing paradigm from inside his own Sussex group. The essay's thesis, wearing a lab coat. The illusion's existence as a reportable phenomenon is not seriously doubted.
- Tiny Blue Dot Foundation — verified against the foundation's own site and releases. The mission, verbatim from the site: "Our mission is to develop Perception Box modification techniques that are safe, effective, reliable, measurable, widely accessible, and easy to use"; the press boilerplate adds that people "live in their own unique Perception Box™ whose walls can be expanded to minimize suffering, increase acceptance of self and others" — expanded is their verb, and the body now uses it where it describes her framework ("the walls move" remains this essay's own language). The site lists Elizabeth R. Koch as Founder (some of the foundation's own releases say Co-Founder without ever naming another — the body says "built," which is safe); the entity dates to 2014, the open grant program to 2022. Christof Koch's title is Chief Scientist, and the no-relation fact is in the foundation's own boilerplate. The numbers, verified: 45 projects across four cycles since 2022; the fourth round (August 2026) awarded $9M across 11 projects from "close to three hundred" submissions after "a rigorous double-blind external review process" — their words; grants run to $900,000 over three years; a fifth-round RFP opened September 2026, its named domains Agency and Independence · Interoception and Emotional Regulation · Identity and Self-Narrative · Values, Meaning, and Belonging (the body's topic list is drawn from these and from funded-project descriptions; draft 01's "belief updating" was not their language and was replaced). The foundation is described in its own releases as the scientific research arm of the Perception Box framework developed by Elizabeth R. Koch; Unlikely Collaborators is her separate organization.
- Christof Koch, Then I Am Myself the World: What Consciousness Is and How to Expand It (New York: Basic Books, 2024) — verified exact; the title is a line from Tristan und Isolde, credited in the book's own introduction. The experience is a smoked 5-MeO-DMT session (around 2020, in his early sixties): a felt dying — "no more Christof, no more ego" — not a medical emergency, which is why the body says "an experience that briefly took the walls down" and never "near-death." What changed, per his own interviews: the fear of death gone, a metaphysical drift toward idealism completed — and the drift is the corrected part: it predates the book by a decade (he was already calling IIT "a modern variant of panpsychism" in his 2012 Confessions of a Romantic Reductionist), so the body says "drifted for two decades" rather than crediting one afternoon. His empirical program continued unchanged: the Allen Institute, the consciousness-measure work, the foundation. Reception: Science called it "magisterial"; the sharpest criticism (Horgan's affectionate "Famous Brain Scientist Loses His Mind") targets IIT, not the book's honesty.
- Georges B. J. Dreyfus, The Sound of Two Hands Clapping: The Education of a Tibetan Buddhist Monk (Berkeley: University of California Press, 2003) — draft 01 carried an invented subtitle ("The Curriculum of…"), the same failure mode 009's sweep caught on Whitehouse, corrected here. Verified: fifteen years of Geluk training in the exile institutions in India (the Institute of Buddhist Dialectics in Dharamsala and the three seats as re-established in the south — the body now says "in exile in India" so no reader pictures Tibet); Geshe Lharampa, the highest of the four geshe ranks, 1985 — the first Westerner, a primacy no scholarly source contests (Michael Roach's promotional "first Western geshe" is 1995 and cannot displace it; phrased precisely to avoid the association, since Roach's organization carries genuine scandal). The debate mechanics follow the book: defender and challenger roles (pp. 211–13), the defender's three allowable answers and the "cir, cir, cir" ("why, why, why") hounding, ten hours a day in the pre-1959 seats. Corrected: the clap's traditional symbolism per Dreyfus (p. 217) is wisdom and method meeting, not a gauntlet — the combative image was the draft's, and is gone; "every afternoon" flattened sessions that also ran mornings and into the night, now "daily." The limits paragraph is his own settled position: debate as "a genuine though limited realm of inquiry," both competitive and cooperative, with "an enormous degree of conformity" among Geluk scholars (p. 311), bounded inquiry, and Batchelor's jab quoted and only partly answered — the body carries the fence because he does. Vetted clean personally; his 1998 Shugden article is the one entanglement, and this essay does not cite it.
- Bhikkhu Bodhi, ed., A Comprehensive Manual of Abhidhamma: The Abhidhammattha Sangaha of Ācariya Anuruddha (Kandy: Buddhist Publication Society, 1993) — verified: first BPS edition 1993 (revised issue 1999; BPS Pariyatti edition 2000), presenting the eleventh–twelfth-century compendium with translation and explanatory guide. The 89 and 121 cittas are the manual's own headline numbers (its Table 1.1 is titled "The 89 and 121 Cittas at a Glance"); root (hetu), feeling (vedanā) and object (ārammaṇa) are three of Chapter III's six literal classification categories, so the body's dimensions are the text's own. The inter-school disputes are, if anything, understated in the body: the Kathāvatthu — one of the seven books of the Theravada Abhidhamma Piṭaka itself — records over two hundred disputed points, and the Theravāda and Sarvāstivāda schemes genuinely diverged (82 versus 75 constituents). Draft 01's "lab notebook of colleagues comparing observations" was corrected by this sweep: whether the Abhidhamma records meditative observation or scholastic system-building is a live scholarly debate (Gethin's work on the mātikās and Braun's The Birth of Insight carry the observational side; the disputational, exegetical reading is equally established), and the body now says so and rests its weight on what is undisputed — the contested charts.
- The second dart: Salla Sutta ("The Dart"), Saṃyutta Nikāya 36.6 (Vedanāsaṃyutta), in Bhikkhu Bodhi's translation (The Connected Discourses of the Buddha, Wisdom Publications, 2000). Verified against the Pali and two mirrors of Bodhi's rendering, and corrected on three counts from draft 01: the simile is sequential — one dart, then a second striking hard upon it (anuvedhaṁ; the compound is dvisallena, "by the double dart") — not two arrows at once; Bodhi's word is "dart," which the body now uses; and the sutta's own inventory of the second dart is sorrow-grief-lamentation, aversion (paṭigha — the body's "resistance" is textual), and seeking escape in sensual pleasure, because the untrained person "knows no escape from painful feeling other than sensual pleasure" — draft 01's "commentary" and "story" were absorbed from modern mindfulness popularization and are gone. The difference-maker in the text is instruction: the uninstructed worldling (assutavā puthujjano) versus the instructed noble disciple (sutavā ariyasāvako) — so "the discourse lays the difference on instruction" is the text's claim, with the caveat (in the file, not the body) that the instructed noble disciple is instruction-plus-attainment, not a weekend technique.
- N. T. Van Dam et al., "Mind the Hype: A Critical Evaluation and Prescriptive Agenda for Research on Mindfulness and Meditation," Perspectives on Psychological Science 13, no. 1 (2018): 36–61, doi:10.1177/1745691617709589 — verified exact: fifteen authors (Van Dam then at Mount Sinai), several of them career contemplative researchers and practitioners (Saron leads the Shamatha Project; Lazar, Vago and Britton are contemplative scientists; Olendzki is a Buddhist scholar), which is the body's phrasing. All four attributed findings are in the paper: only ~30 per cent of intervention research past NIH Stage 1; no accepted technical definition of "mindfulness"; weak or absent active controls (with the AHRQ meta-analysis's moderate-evidence ceiling); adverse events actively assessed in fewer than a quarter of trials against 100 per cent in pharmacology. The published pushback was Davidson and Dahl's commentary in the same issue, with a collegial reply — not Kabat-Zinn, and the essay claims no more. The "growing up in public" sentence is carried by what followed: Britton's group defining and measuring meditation-related adverse effects (Clinical Psychological Science 9, 2021), Farias et al.'s systematic review (Acta Psychiatrica Scandinavica 142, 2020, ~8 per cent adverse-event prevalence), and the preregistered 8,376-student MYRIAD trial publishing its null results in 2022.
codex, proposed by this transmission: the walls · the wager