*(Historical record, filed 14 July 2026, before the v2.0 rename: "resolution" herein is the framework term since renamed "cohering", and claim labels are as of v1.6. Preserved unedited.)*

# The Container Problem
**Working note v0.2 · Space Immanence · 14 July 2026**
**Supersedes v0.1 (committed alongside). Incorporates by reference: `Container_Redteam_Study_B_Items_v0.1.md`, `Container_Redteam_Study_C_Constraints_v0.1.md`, `Container_Redteam_Governance_v0.1.md`, and the red-team report v0.1. Where this note and a companion document conflict, the companion document wins for its own domain (items, C constraints, governance).**
**Status: pre-theoretical claim, post red-team. Still circulated to be attacked.**

## 0. What changed from v0.1, and what forced it

The mechanism is now conditional, not default (forced three ways independently: the internal contradiction M2, the scalar-dominance analysis in the C constraint document §5.2, and the clinical record on representational limits). The trajectory is demoted from discriminating signature to candidate signature and gains the tradition-contrast commitment (M3, M6, M9). Stage 1 is retyped from observed result to testable prediction (M8, M12). The priority claim is narrowed to one sentence, given a neighbour map, and given its own falsification condition with a search protocol (M1, M5, M7, M9, M16, governance §1.4). All drawer conditions are rewritten to the governance document's tightened wordings, with the missing conditions added (M14). "Darkening" is retired (§2). C's downgrade to conditional-existence form is accepted explicitly (constraint doc §5.2). The governance quarantine is adopted in full, including the P2 sign-error block (governance §§1–2).

One sentence from v0.1 is retained as the template for everything here: container collapse at the far end of practice is a consistency check we must pass, not credit we can claim.

## 1. The claim, conditional form

Experience is sometimes reported as happening inside something: a field or space of awareness that contents are presented to. Call this the container structure. The programme makes two claims, graded separately.

**C1 (existence).** There is a non-spatial container factor in first-person report, separable from the perceptual spatial frame and from body-boundedness, under the pre-registered machinery of the items document. This is an empirical existence claim about a factor, not a claim about what the factor means.

**C3 (mechanism, conditional remainder-modelling, hereafter CRM).** Any system that models by selective illumination produces at every moment a remainder: content that funds the current figure but is not shown as figure. By default the system does not model this remainder at all, or models it minimally as a point-index (an egocenter: "more, relative to here"). The medium encoding (homogeneous, surrounding, empty, in the operational senses frozen in the C constraint document §4.2) is predicted to become optimal only conditional on the remainder itself becoming a target of modelling, that is, under sustained ground-directed attention. Formally: there exists a non-rigged, independently justified task-and-rate regime, characterised by high introspective demand, in which the medium encoding is uniquely optimal against all live rivals and exhibits all three signatures under frozen tests. This is an existence claim over regimes, not a universality claim, and it is the honest downgrade the constraint document forces: compression alone can never select the medium, because a medium is a scalar plus an index set plus enclosure topology and every cost function prefers the scalar; only task demands can pay for structure.

The conditional form is not a retreat made under fire alone. It is independently motivated by the clinical record: the blind spot is perceptually completed, not experienced as a gap; the neglected hemifield has no experienced edge; Anton's patients confabulate content rather than perceive a void; inattentional blindness leaves no phenomenal placeholder. The empirical default for the unrepresented is nothing. The one literature in which absence is positively represented (absence perception, Farennikova's template-mismatch account) requires an actively deployed expectation directed at where the missing thing would be, which is structurally the CRM trigger: the remainder gets encoded only when attention queries it. And a predicted transition (the λ-parameterised E8-to-E6 shift in the constraint document's Appendix A) is a sharper, more falsifiable object than a default.

Two scope refinements. First, "cannot itself be shown" is weakened to "cannot be shown as figure": where ordinary experience does present its remainder, it presents it as heterogeneous and relevance-graded (James's fringe, Gurwitsch's theme, thematic field and margin). CRM therefore needs a two-level story: a structured near-remainder (the fringe), and a far remainder that is not presented at all and is CRM's actual target. The boundary between them, and the handoff, must be specified before C's target is well-posed; this is an open hole (§8) that closes at C freeze or C does not run. Second, C is constitutionally asymmetric: even a maximal positive result exhibits a data structure with three formal properties, and "read phenomenologically as a container" is a bridge posit no optimisation can test. C retains full kill authority over the mechanism and capped, necessary-condition confirm authority. This asymmetry is stated now so a positive result cannot be inflated later.

## 2. Naming and provenance

The framing that seeded this programme is Dōgen's, via Okumura's Realizing Genjōkōan: when one side is illuminated, the other is dark. That provenance is aesthetic lineage and is disclosed as such, and it is explicitly not evidence.

The placeholder name "darkening" is retired in this note, one step beyond the governance document's two-register recommendation, for a substantive reason: the clinical record shows the metaphor carried a false implicature. "Dark" suggests the ground is experienced as dark by default, and the finding is that the ground is not experienced at all until attention makes it a target. A mechanism name that pre-loads the phenomenology it is supposed to derive, and that imports the framework's favourite classical ally into every sentence, is a liability the naming-risk audit already priced. The mechanism is called CRM throughout. Per the governance rules, no mechanism vocabulary of any register appears in instruments, item wordings, preregistrations, or the public machine layer; Study D's manipulation is named "unstructured-field viewing" in all study documents.

## 3. Priority: the clearing, fenced

The claim, at its full and final narrowness: no prior work derives a homogeneous, surrounding, empty phenomenal medium from the remainder of selection under resource constraints and task demands. Every broader priority sentence from v0.1 is withdrawn.

The neighbour map, one discriminating sentence each:

* Place (1956) names the phenomenological fallacy (properties of the representation read as properties of an inner field) but supplies no generative mechanism; CRM is a candidate mechanism for the error Place named.
* Dennett (1991) diagnoses the Cartesian Theater as theorist's error and folk seeming, and his "absence of representation is not representation of absence" is the canonical objection CRM absorbs by going conditional; the diagnosis is not a sub-personal derivation of the empty-medium format, which is the gap CRM occupies.
* Jaynes (1976) derives the container from metaphor acquisition (linguistic construction); CRM is sub-personal and non-linguistic, and the two are discriminated empirically by the tradition-contrast test and B's cross-family convergence design, not by argument.
* Metzinger, transparency and the tunnel (2003, 2009) derives phenomenal immediacy from unavailability of construction processes and the tunnel from bandwidth; the modelled variable is not the selection remainder.
* Metzinger, MEO and the contraction principle (2020, 2024) is the nearest neighbour: empty-space phenomenology derived from the system modelling its own epistemic capacity, tied to tonic alertness. CRM differs on the modelled variable (remainder of selection, not capacity) and on scope (the ordinary, untrained container, not MPE). Candidate discriminator, offered not established: MEO predicts container phenomenology covaries with tonic-alertness manipulations; CRM predicts it tracks ground-directed attentional demand at fixed alertness. This discriminator is underdeveloped (§8) and the per-neighbour entry must be finished by someone trying to prove convergence.
* PCM (Rudrauf, Williford, Friston) formally derives the perspectival frame under projective geometry: content-structured, perspectival, neither homogeneous nor empty, and precisely the first factor B controls for. PCM-style projective encodings are in C's frozen encoding space (E8/E9 regimes) so the comparison is run, not asserted.
* Graziano (AST) derives an "ethereal, vaguely localisable" awareness-attribution from a control model of attention. v0.1's sentence that the schema "has no reason to reshape with contemplative training" is retracted as unverified and likely false; AST's schema is plastic, contemplative training is attention training, and an extended AST is scored as a live rival in its own vocabulary, with the trajectory no longer billed against it.
* Laukkonen & Slagter (2021) derive a staged contemplative continuum ending in pure awareness from precision-weighting and shrinking counterfactual depth; they do not derive the container format at the intermediate stage. Candidate discriminator: precision accounts predict that opacity reveals process structure (heterogeneous, dynamic); CRM predicts a homogeneous empty medium at the intermediate stage; B's facet structure is in principle sensitive to the difference.
* Limanowski & Friston (2018) formalise the trigger operation itself (introspective attention rendering transparent states opaque); what remains CRM's own is the claim about the content of what opacity reveals for the far remainder.
* Husserl, James, Gurwitsch describe the structured near-remainder; CRM's target is the far remainder, and the handoff is the §8 open hole.

**Priority falsification condition (new).** The move is claimed, and this programme's novelty case dies, if any prior work derives a spatially surrounding phenomenal medium from a representational-completion or remainder-modelling requirement. Search protocol, binding: (i) systematic search of PhilPapers, Scholar and PubMed under frozen strings (phenomenal space derivation; space of awareness model; background of consciousness mechanism; horizon generative model; unattended content encoding; remainder representation attention), logged with dates; (ii) full forward-citation chase from Place 1956, Dennett 1991, Jaynes 1976, Metzinger 2020 and 2024, the PCM papers, and Laukkonen & Slagter 2021; (iii) before any public priority statement, at least one hostile external referee from the predictive-processing or Metzinger orbit reviews the claim with instructions to find the prior derivation. The method humility is earned: this programme's own narrative contains two failed "nobody has" claims in one week. Discovery at any time fires the condition and is logged publicly.

## 4. The trajectory, demoted and typed

The sequence point, then container, then collapse is a candidate signature: as of this note, a post-hoc arrangement of three known results, held to the same discipline v0.1 applied to the MPE collapse endpoint. It is also, honestly, a two-mechanism trajectory: embodiment and egocentric modelling give the point; the MPE literature gives the collapse; CRM's sole novel work is the stage-two transition, and the note claims credit for exactly that and nothing else.

**Evidence typing (correcting v0.1).** The stage-one citations measure folk theory, not phenomenology: Forstmann & Burgmer's participants localised consciousness on brain images for a generic person; Starmans & Bloom measured third-party closeness judgments; only Bertossa et al. is first-person, and its protocol presupposes a locatable answer. Stage one is therefore restated as a prediction B's non-meditator free-report arm must confirm, not a result. Additionally, the folk-theater finding cuts both ways: the folk construct is a located presentation venue, which satisfies presented-to structure before any training. So stages one and two are distinguished by format (point-located venue versus unbounded homogeneous medium), not by the mere presence of presented-to structure, and the items document's located-theater outcome (point, then located theater, then collapse) is pre-registered as a named alternative, not a failure.

Rivals, stated in their own vocabularies:

* **Curriculum.** The classical formless-attainment arc makes boundless space the literal object of the fifth absorption, transcended in turn: the point-to-space-to-beyond-space sequence is a two-millennia-old syllabus delivered procedurally and cross-linguistically, and the MPE sample anchoring the collapse endpoint is dominated by lineages that teach it. The null rival: practitioners report what their manuals teach, in the order taught. A cross-linguistic vocabulary control is structurally blind to this confound.
* **Construction (Jaynes).** If metaphor acquisition builds the phenomenology, metaphor inventory and experience strata are the same variable and B's planned dissociation collapses; the Lakoff regression alone cannot separate them.
* **Extended AST and precision-reallocation accounts** reach a staged sequence with plausible tuning; discriminators are at the format level (§3).

**The discriminating commitment (new, binding).** The one test that separates CRM from both curriculum and construction: practitioners of ground-interrogating trainings that never use space instruction should still develop the container stage on CRM, and should not on the rivals. Design: space-lean arms drawn from Mahasi-style pure noting and Rinzai kōan training; space-rich arms from Dzogchen, Mahamudra and open-monitoring lineages with explicit spacious-awareness instruction. Two cautions are design requirements, not footnotes: modern contemplative culture is cosmopolitan, so lineage-contamination instruments (books, apps, retreats across traditions, extending MV-V) are mandatory; and a historical-text arm (pre-modern practitioner reports from space-lean lineages, predating cross-pollination) is the cleanest available natural experiment and should be scoped. If only the instructed develop the container stage, the trajectory belongs to pedagogy and drawer-condition 2 fires.

**Ordering commitment.** The trajectory's within-person claim gets kill authority only from the longitudinal or randomised-training arm (meditation-naive participants, pre and post container measurement). We commit to that arm rather than striking the trajectory from B: without it the ordering claim is permanently unfalsifiable, and a claim we cannot afford to test is a claim we do not get to make. Cross-sectional strata results in B are reconnaissance for the ordering claim, never verdicts on it.

## 5. Studies

**A. MPE-92M reanalysis.** Reconnaissance only; no kill or confirm authority; never citable as support (pinned by governance drift-test 4). Purpose: item harvesting and curve reconnaissance, with the stated caveat that the instrument's contemplative lexicon (including the Luminosity factor) meeting any light-adjacent theory vocabulary is contamination one step removed, which is one more reason §2 retires that vocabulary.

**B. The residue instrument.** Kill authority over C1 (existence) only. Fielded exclusively through the items document's machinery, which this note adopts wholesale: the two metaphor-family facets (dative and venue) with cross-family convergence as the signal, since neutral wording is impossible and structure must adjudicate; NSC-FC before all Likert content; three-form planned missingness; randomised order arms with only the NSC-first arm counting toward the existence verdict if teaching effects exceed threshold; cognitive-interview gates with reading-coded deflation ceilings; the jangle-bait and method-factor controls; the matched-abstraction ownership/agency positive control that voids the run if it fails to separate; incremental validity against the boundlessness pools (OAV/5D-ASC, EDI, NADA-T), since a residue factor collinear with unity items is the collapse pole renamed; the Lakoff regression with a pre-registered mediation threshold; Korean as the test language with full invariance gating; and the §4.3 decision table as the sole arbiter of verdicts, including the located-theater and genuine-absence rows. The known-groups meditator arm makes a naive-arm absence interpretable rather than convenient.

**C. Formalising the mechanism.** Kill authority over C3; confirm authority capped. The constraint document is adopted as binding and must be hash-committed, frozen, before any formalisation, implementation or pilot; work preceding the freeze is exploratory by definition. The two-regime structure is the study: regime one (remainder not queried) predicts a point or indexical optimum; regime two (ground-directed attentional demand) predicts the medium. Tier 1 (the analytic exchangeable-source model) runs before any code, and a Tier-1 failure kills the mechanism without a training run. The Appendix A λ-transition is the only part of C that touches the trajectory, and only if separately frozen. One disclosure: the author has proposed a candidate ecological justification for the task ensemble (attentional steering: a system planning where attention goes next needs a representation of where else the figure could land, which is an index set enclosing the current figure). Because task-ensemble selection is the master degree of freedom (D6), this proposal is submitted as hostile input for red-team scrutiny like any other, enters only through the frozen justification document, and its author-origin is flagged.

**D. Dissociations.** Later; lab; own drawer condition. Body-model perturbation versus unstructured-field viewing, crossed magnitude predictions with pre-registered power analysis, and a Ganzfeld-phenomenology literature pass first: if existing reports already describe frame or container destabilisation under homogeneous fields, that arm is retyped as retrodiction before anything is claimed. Scope sentence per the report: acute perturbation data (including psychedelic datasets) become probative only after B's three-factor separation exists; D then either commits to a psychedelic-adjacent prediction or states why not.

## 6. Falsification conditions

Fixed now, in the governance document's tightened form. The idea, or the named limb, goes in the drawer if:

1. **Existence.** No separable third factor emerges in the introspectively experienced strata under the pre-registered discriminant criterion (CFA nested-model comparison, thresholds frozen before fielding), with the item pool frozen after one adversarial revision round; a second null after that single permitted revision fires regardless of item-quality objections. A factor appearing equally strongly in naive strata fires condition 2, not a save. All verdicts route through the items document's decision table, including its voiding gates.
2. **Trajectory.** The pre-registered ordered contrast (naive below intermediate-trained) fails at the pre-set minimum effect size, or equivalence testing (TOST at a pre-registered SESOI) shows the naive-to-trained slope flat; the collapse limb is a separately pre-registered prediction whose failure kills the collapse limb only. Cross-sectional survivorship is acknowledged as unresolvable in B; the condition fires on B's data anyway, and the ordering claim additionally dies if the longitudinal arm fails its pre-registered contrast. The tradition-contrast result feeds this condition: container-stage development confined to space-instructed lineages fires it.
3. **Vocabulary/construction.** The cross-linguistic arm is mandatory before C1 exits pilot grade. Pre-registered model comparison: container-factor loading regressed on individual introspective experience versus a frozen language-level container-productivity index built corpus-side before any B data are seen. Fires if the language-level predictor dominates under the pre-set criterion and the individual predictor adds nothing at threshold.
4. **Mechanism.** Under the frozen two-regime C: fires if the regime-two optimum is robustly non-medium under the constraint document's decision rule, or if any of its seven failure-reporting conditions hold, including formalisation failure, which is published as a result, not absorbed into redesign. Regime one producing a point-index is the theory's prediction, not a failure.
5. **Dissociation.** D's crossed magnitude predictions fail at pre-registered power.
6. **Priority.** The §3 condition, with its search protocol.
7. **Staleness.** If B's cognitive-interview gate has not been entered within 24 months of this note, the programme's public status reverts to dormant. A drawer that can be indefinitely deferred is a shelf.
8. **Public semantics.** From the moment the programme goes public, drawering is a published state (`status: "drawered"` in container.json and state.json, drift-tested), and firing conditions 1 or 3 is additionally logged as negative evidence on S1's weak reading in the framework ledger, per the symmetric transfer commitment.

## 7. Governance

The quarantine is adopted in full: container.json as a separate machine object with its own namespace (C1–C4), the thick no-transfer clause, and the drift-test suite of governance §2.4, including the Study A containment pin. One evidential edge exists, and only one: B to S1, weak reading only, both directions, supporting never establishing, S1's confidence capped at its current pin on confirmation, and the negative direction pre-committed. The P2 sign error is stated in this note because it is the transfer nobody would think to block: a confirmed CRM is a deflationary rival to P2, not support; it would explain the sense of inside-ness as a modelling strategy, and P2 would lose its explanandum to it. The firewall blocks credit in both the naive direction and the flattering one, and permits discredit to flow. The empirical child is allowed to kill parts of the metaphysical parent; that sentence is in the governance file so it stays meant when it becomes expensive.

v0.1's wiring sentence is rewritten as follows and the old form is withdrawn: CRM is a candidate generative mechanism for the intuition the v1.5 container-error diagnosis targets; confirming the mechanism would not confirm the diagnosis, and the two are graded separately.

## 8. Open holes, named so they cannot be quietly forgotten

The fringe/far-remainder switching boundary is unspecified and closes at C freeze or C does not run. The MEO discriminator (alertness versus attentional-demand covariation) is a candidate needing development, ideally by someone instructed to prove convergence. The task-ensemble ecological justification is unwritten; until it is frozen, C cannot start, and the author's steering proposal inside it is unreviewed hostile input. The tradition-contrast operationalisation (lineage purity criteria, contamination instruments, the historical-text arm's scope) is design work not yet done. And the bridge problem stands permanently: C can never test "read phenomenologically as", and every public statement about C must carry that asymmetry.

## 9. Key sources

As v0.1, corrected and extended: Place 1956; Dennett 1991; Jaynes 1976; Metzinger 2003, 2009, 2020 (PhiMiSci and Cognitive Neuropsychology), 2024; Gamma & Metzinger 2021 (PLOS ONE, MPE-92M); Graziano 2013, 2019; Webb & Graziano 2015; Rudrauf et al. and Williford et al. (PCM); Chalmers 2018 (JCS); Laukkonen & Slagter 2021; Limanowski & Friston 2018; Sandved-Smith et al. 2021; Sytsma & Machery 2010; Sytsma & Ozdemir 2019; Starmans & Bloom 2012; Bertossa et al. 2008; Forstmann & Burgmer (JEP: General); Ramachandran & Gregory 1991; Bisiach & Luzzatti 1978; Marshall & Halligan 1988; Mack & Rock 1998; Farennikova 2013; James 1890; Gurwitsch; Husserl; Lakoff & Johnson 1980; Kövecses; Choi & Bowerman; Longo et al. 2008; Pennycook et al. 2015; Graham et al. 2006; Okumura, Realizing Genjōkōan (aesthetic lineage, disclosed, not evidence).