{"path":"research/parity-and-the-zone-interior.md","content":"# Parity, the Permissive Zone's Interior, and the Ruling That Deliberus Maps It\n\n**Date**: 2026-08-24 · **Type**: founder-listening session (Ruth Chang on *The Kids or Childfree Podcast*, ep. 56, March 2025 — first half; the kids-or-childfree decision as the worked example of a hard choice) + a ruling that extends the residue map's reach. **Provenance note**: the Chang positions below rest on the 3:AM interview (fetched, quoted in [philosophical-foundations-stress-test.md](philosophical-foundations-stress-test.md) §4), the episode's own summary, and secondary accounts of her AI-alignment work — her papers are not freshly read here; verify before citing outward.\n\n## The founder's assessment of Chang, verbatim spine\n\nListening to the first half of the episode, the founder sorted Chang's apparatus into what transfers and what does not:\n\n> \"I'm unconvinced about the applicability of the commitment-angle to what we're trying to do with Deliberus. But parity and looking inwards for most of the equation of life-transforming decisions makes sense. And of course for such decisions you do in practice have to commit for at least a while to let the decision bear real fruit and impact your life for better or worse, also it's probably more psychologically healthy than constantly second guessing your big life decisions. This can and should be modeled as claims inside Deliberus for those that would like to use it to make this class of decision.\"\n\nSo: **parity transfers** (it is already the permissive-zone residue's lineage), **commitment does not transfer as platform machinery** — it is the individual's move inside the zone, and Deliberus's relation to it is *representational*: commitment-reasoning (including the pragmatic case for committing — decisions must be inhabited to bear fruit; chronic second-guessing is corrosive) can be modeled AS CLAIMS for users who bring this class of decision, not built into the platform as a stance.\n\n## The ruling: the zone has an interior, and Deliberus maps it\n\nThe prior settled picture had the platform's work ENDING at the zone boundary: certify honestly that reason leaves several options standing, decline to fake a winner. The founder ruled past that boundary, verbatim:\n\n> \"If the internal process is also modeled as claims inside Deliberus, and it presumably is traversable, decomposable and representable in language, perhaps not fully but quite deeply, then it can at least be helpful for decisions within that zone, even if we need not force a definitive verdict and certainly not one that fits everyone. The internal process modeling should be clear which reasons and values weigh more for whom, but still be used to inform other people's internal/spiritual/values-reflections that come to explore a similar decision at a later date. Deliberus should not shy away from dealing with this space of reasoning.\"\n\nUnpacked into its commitments:\n\n1. **The zone's interior is claim-representable** — traversable and decomposable in language, \"perhaps not fully but quite deeply.\" The honest limit (\"not fully\") is stated inside the ruling itself; the attunement pole and the load-bearing unsaid mark where language runs out, and the ruling claims deep reach, not total reach.\n2. **No verdict is forced, and none would fit everyone anyway** — the zone certification stands. What changes is what happens after certification.\n3. **Attribution is the load-bearing requirement**: the interior map must show *which reasons and values weigh more for whom*. This is per-person weighing structure — the same `attributed_to` + weighing apparatus the graph already has, pointed at individual travelers of a zone rather than debate positions.\n4. **The map accumulates across travelers**: one person's articulated interior traversal becomes material for the NEXT person facing a similar decision — \"inform other people's internal/spiritual/values-reflections that come to explore a similar decision at a later date.\" The zone's afterlife is a growing, attributed library of traversals.\n5. **The scope word is deliberate**: \"internal/spiritual/values-reflections.\" The reflective register — not just argumentative prose — is inside Deliberus's remit. \"Deliberus should not shy away from dealing with this space of reasoning.\"\n\n## Why this is the contextual-exit ruling's sibling, not a new species\n\nThe 2026-08-20 ruling ([weighing-scheme-and-terminus-classifier-design.md](weighing-scheme-and-terminus-classifier-design.md) § Addendum 2026-08-20) established for the *fittingness* residue: the exit is an entrance — \"it's contextual\" is where the descent changes object, from a scalar weight to a mapping from situations to rankings. Today's ruling makes the same move for the *permissive-zone* residue: \"several options permitted\" is where the mapping changes object, from the shared argument structure to the **attributed interior** — a mapping from persons (their values, weightings, situations) to resolutions. Together they complete a pattern: **no residue type is a dead end; each has a post-certification object the map keeps articulating.** Fittingness → situated structure. Permissive zone → attributed traversals. (Empirical → evidence; axiom-choice → the axiom fork held openly; structural → the vantage difference named; is-bedrock → competing readings held.)\n\nThis also composes with Chang's own subtype analysis (the stress test's four-way split of the zone): for a **parity** leaf the interior map records will-based commitments — self-constitution, no mistake on either side; for a **permissivism** leaf it records which epistemic standards each traveler weighted; an **indeterminacy** leaf should dissolve under sharpening before interior mapping matters; **incomparability** is the one true dead end, and rare. The interior-mapping ruling gives the first two their affordance.\n\n## The Chang AI convergence, and where the ruling goes further\n\nChang's current alignment argument (NOT from the episode the founder heard — that one stays on the kids decision; this comes from the Ben Yeoh Chats episode summary and public accounts of her AI work, held loosely per the provenance note): optimization-shaped AI assumes every pair of options is better/worse/equal on some scale, so where human value exhibits parity, a maximizer *structurally must fake a winner*. Deliberus's declining-to-verdict design is the architecture-level agreement. But today's ruling is the part Chang's critique alone does not give: after refusing to fake the winner, the system keeps working — it maps how real people actually resolved the zone, attributed and reusable. Refusing false verdicts is the floor; the attributed interior library is the product. This is also the strongest available answer to \"isn't a residue an unsatisfying output?\" (the deprivation-curiosity worry in [curiosity-as-growth-fuel.md](curiosity-as-growth-fuel.md)): the zone certificate comes WITH a living annex of how others crossed it.\n\n## Build implications (parked, founder-paced — nothing here is scheduled)\n\n- **Ontology**: an interior traversal is claims + weighings + a resolution marker attributed to a person, attached under a zone-certified terminus. Nearest existing machinery: `attributed_to` normalization, the weighing CQ set, `CLASSIFIES_TERMINUS` grounds. A `resolution` lifecycle marker (\"this traveler committed to A, for these attributed reasons\") does not exist and would need design — it is a *record of a choice*, not a verdict, and must never propagate strength (the REPORTS non-propagation lesson applies).\n- **Privacy gradient**: interior/spiritual reflections are the most personal material the platform would hold; capture-immediately-publish-deliberately and the retrieval-weight machinery (raw/draft lifecycle) are the right existing rails. Solo use (L6, incentive-clean) is the natural first mode.\n- **The diarized live session** produces attributed per-person weighings as a by-product — the first real interior-traversal data would arrive through the very session already planned.\n\n## Addendum 2026-08-25 — the sacred-domain extension (founder verbatim, woven on instruction)\n\nThe ruling above reached its fullest form the next day, when the founder extended it across the most sensitive territory the platform could touch:\n\n> \"Just because we come into the domain of personal or contextual decisions, it doesn't mean we can't render helpful general considerations and reasoning and philosophy and heuristics that might be highly reusable for others. We just need to be wary of concluding premature verdicts or advice, since traversing across the fundamental boundary of talking about what's true or good regarding each sentient individual's body and spirit and their own relationship with these/with themselves is an extremely sensitive and I'd even say sacred subject. It can and should be possible to reason about everything, but we should be even more risk averse and epistemically humble in the personal/spiritual domain, since every conscious being's right to self-determination is for many one of the deepest values we can sense. But even this should not be artificially protected against analysis ultimately. Which of course, again, necessitates an equally deep attunement, if analysis is allowed to descend/map/traverse this deeply into the soul/spirit. I don't see any fundamental need to put up stop signs for ourselves in facilitating either/both however, just because it's sensitive and sacred. In a way it makes it even more important to shine the light of reason and clarity on all of this.\"\n\nIts commitments, named: **no stop signs anywhere** — nothing is artificially protected against analysis; **the proportionality principle** — depth of analysis must be matched by equally deep attunement, and epistemic humility scales with intimacy; **general considerations remain distillable** even from personal territory (philosophy, heuristics, reusable reasoning) — only *premature verdicts and advice* are what must not arrive; **self-determination is named among the deepest values**, which is the source of the heightened humility, not a fence around the territory; and sensitivity *raises* the importance of illumination rather than lowering it. This is the substance the July 10 sacredness-brake re-scope was waiting for (consent-and-timing, never epistemic exemption) — recorded as ratified-in-substance in the weighing design doc's 2026-08-25 addendum.\n\n## Addendum 2026-08-25 (night) — the fallible-self-knowledge caveat (founder verbatim, conviction-grade)\n\nAttaching to the sacred-domain ruling and to the proportionality principle's canon candidacy, the founder stated the epistemological caveat that governs ALL interior mapping:\n\n> \"I believe firmly that humans may be wrong about the causes of their own experiences. Not wrong about their fundamental experience, but about what objective causes were behind it. They may be wrong about counterfactual what-if branches of reality and experience that might have played out and resulted if they'd have acted differently or if variables and factors of reality at the time of or leading up to the experienced would've been different. They may be wrong about what would best satisfy a need or desire they're sensing. They may in short be wrong about correlates of internal direct experience and about the constitution and conditionality of their subjective world. We need to structurally keep this in mind in all future settings contexts and moments, in order to best assist in mapping what is yet beyond the horizon.\"\n\n**The distinction, named**: first-person authority is **incorrigible about the experience itself** (the felt datum — challenging it is a category error, and attunement is how it is honored) and **fallible about everything surrounding it** — causes, counterfactual branches, satisfiers, correlates, \"the constitution and conditionality of their subjective world.\" Those surrounding claims are ordinary empirical territory: decomposable, evidence-attachable, challengeable like any claim, whoever asserts them.\n\n**This resolves the sacred-domain ruling's internal tension precisely.** Why can analysis go anywhere with no stop signs, even in the soul's terrain? Because the genuinely protected core is TINY — the felt quality alone — and everything else in the personal/spiritual domain is fallible-surround. And why is deep attunement owed in proportion? Because the datum is the person's alone; the map around it is shared. Analysis maps the fallible surround; attunement honors the incorrigible core; the proportionality principle is what keeps them together.\n\n**The empirical literature firmly agrees with the founder** (survey-grade; verify before citing outward): Nisbett & Wilson's \"Telling More Than We Can Know\" (1977) — people confabulate the causes of their own behavior while accurately reporting the behavior; **choice blindness** (Johansson & Hall, Lund — Swedish research): people fluently defend choices they never made; **affective forecasting / miswanting** (Gilbert & Wilson): systematic error about what will satisfy; misattribution of arousal (the Dutton-Aron bridge study). Experience authoritative, self-interpretation fallible is not a philosophical stance the project must argue — it is the mainstream empirical finding.\n\n**Ontology consequence (buildable, parked)**: interior-traversal claims split by kind — *experience reports* (observer-privileged; the attunement pole's object; decomposition never attacks the datum) vs *self-interpretations* (first-person causal/counterfactual/satisfier claims — which the existing `epistemic_status` values causal_hypothesis/counterfactual already type, and which get NO elevated credence for being first-person). The weighing set's state-deficit and satisfier-belief questions are this caveat's shipped instruments — asking exactly where the experiencer may be wrong about themselves. The three-ingredient dissolution in decomposing-value.md (state-deficit / satisfier-belief / residual weighting) is its decomposition-side form: the first two ingredients ARE the fallible surround.\n\n**Canon line RATIFIED 2026-08-25 (\"Sounds good\") and added to the One-Breath convictions the same commit, per the upkeep rule:** *\"Analysis may go anywhere, and owes attunement in proportion to depth — the experience itself is the person's alone; its causes, counterfactuals and satisfiers are mappable territory where anyone, including the experiencer, can be wrong.\"*\n\n## Addendum 2026-08-25 — the naming thread, and three related rulings\n\n**\"Residue\" is under founder-approved renaming; \"threshold\" is the leading candidate.** The founder's reason, verbatim: *\"Calling it residue feels to me like it brushes aside the importance of internal reflection and reasoning.\"* The analysis that led here: \"residue\" names the point correctly relative to the dissolve-toward-shared-verdict operation (which genuinely terminates there, keeping the wager's accounting honest) but carries a subtractive connotation — the leftover in the filter — and frames the point entirely from analysis's side, when the core dialectic says it is precisely the entrance to attunement's work. \"Threshold\" names the mode-shift: one kind of work ends, another begins. The migration is NOT yet executed (the word is still open to better candidates); the term appears in the residue map, the wager fraction, the API and dozens of docs, so the rename is a deliberate small migration once settled.\n\n**The never-finished relativity.** Founder: convergence status cannot be finally known, because the mapping is open-ended — \"one can always contribute more considerations and facts... The world and the reasoning keeps evolving.\" Consequence one: every terminus verdict is doubly relative — to the current move-set (already in the definition) and to the considerations-contributed-so-far (now stated). Consequence two, founder-approved as reasonable: the residue fraction is honestly published as a **trend over time**, not a snapshot state. In practice this is cheap: `CLASSIFIES_TERMINUS` edges carry creation times, so the series is computable retroactively; the instrument change is an endpoint/rendering one (fraction-over-time plus honest denominators), and the falsification reading shifts from \"what is the number\" to \"which way does it move as the corpus grows.\"\n\n**Descent and mapping are two geometries, deliberately.** The founder asked why each word. Descent is inherited from the proof lineage (justification examined downward toward more basic things — the local *experience* of examining one claim); mapping is the cartography conviction (the global object produced). The corpus has already recorded that the underlying object is web-shaped (confirmation holism, evidence-division), so descent is the local phenomenology and the map is the global truth — the founder's own phrase \"descend/map/traverse\" treats them as one motion, which stands as the working resolution.\n\n**Display-word ruling: \"fittingness\" is used everywhere for now.** The founder prefers it over \"contextual\" precisely because of the remainder \"contextual\" drops — the normative claim that a situated response can genuinely *fit* (be apt, merited, non-arbitrary) even without a general rule. From the alternatives list, the founder chose **\"fit\"** (the plain noun, 2026-08-25: \"'Fit' sounds best then\") as the beginner-surface candidate — same root, keeps the aptness; \"fittingness\" remains the internal name.\n\n**The horizon reopening (2026-08-25, founder).** The rename candidacy widened: against the earlier objection that horizons are never reached, the founder's reading — verbatim: *\"embodied beings standing on the same ground in the same town generally see and easily map everything this side of the horizon, and explorers reaching beyond the current horizon to uncover new ones is a beautiful image... Anything beyond the horizon exists solely in the imagination/mind of sentient beings or in maps — which also corresponds beautifully to the fact that in Deliberus everything beyond the 1st person horizon concerns itself with what's in a person's mind... And we can explore and discover more of it as a common project and find that we share more landscape that's mappable than we may have anticipated\"* — dissolves the objection (it conflated the horizon-as-line, never stood on, with the terrain AT the horizon, which explorers reach while the line moves). The horizon metaphor natively encodes the never-finished relativity (observer-indexed, moves as you advance), the interior-mapping ruling (beyond the shared line lies what is in minds), and the wager itself (more shared landscape than anticipated); its philosophical lineage is Gadamer's *fusion of horizons*, the canonical term for understanding across minds. Costs noted: the temporal misread (\"on the horizon\" = upcoming) and vocabulary adjacency with the project's own frontier page. **The two-register settlement (founder, 2026-08-25, after the horizon ruling).** The descent/horizon metaphor mixture is a FEATURE, not a defect to tidy: real terrain has three dimensions — outward to the horizon, downward into bedrock and groundwater — and the two meet where the pun wants them to (a spring is where the deep surfaces; a river's source is at once the farthest point upstream and the place water rises from below). So the vocabulary map, settled: **depth and deepen STAY** (a claim's grounds run downward; what deepens is the examination, never the terrain); **horizon** = where the shared mapped surface currently ends; **upstream / sources** available for evidence provenance; **watershed** for where reasoning parts ways. Koan-grade phrasings (\"the horizon in the deep\") are licensed in the VOICE layer — landing prose, doc openings — never on button labels, where beginner-readability governs. All vocabulary is English on every product surface (Swedish stays chat-flavor only; the founder's opaque-name carve applies to library names like nit/spela, never to Deliberus's teaching vocabulary). No future session should \"fix\" the mixture into a single geometry — the metaphor switches exactly where the object switches, and that is the point. **RULED 2026-08-25: \"Horizon it is!!\" — the migration executed the same day** (code: `HORIZON_TYPES`/`is_horizon`/`horizon_map`/`/graph/horizon-map` with back-compat aliases and the legacy route kept for one deprecation cycle; living docs renamed with one supersession note in convergence.md; historical docs and filenames untouched as record).\n\n## Cross-references\n\n[philosophical-foundations-stress-test.md](philosophical-foundations-stress-test.md) §4 (the four-way zone split; small-improvement probe) · [weighing-scheme-and-terminus-classifier-design.md](weighing-scheme-and-terminus-classifier-design.md) § Addenda (the contextual-exit sibling) · [decomposing-value.md](decomposing-value.md) (picking-vs-choosing as the zone's mechanism description; state-deficit/satisfier screens as interior moves) · [convergence.md](../convergence.md) (residue afterlives; wager accounting where parity/permissivism are ceilings) · [analysis-and-attunement.md](../analysis-and-attunement.md) (the register the ruling pulls into scope) · [curiosity-as-growth-fuel.md](curiosity-as-growth-fuel.md) (the unsatisfying-residue worry this answers)\n"}