#Wholeness Language Unpacking (RPR-WHOLE)

A.6.H · Stable · Relational-precision specialization · Normative unless explicitly marked informative · Part A - Kernel Architecture Cluster

#About this pattern

This is a generated FPF pattern page projected from the published FPF source. It is canonical FPF content for this ID; it is not a FPF Reference product feature page.

#How to use this pattern

Read the ID, status, type, and normativity first. Use the content for exact wording, the relations for adjacent concepts, and citations to keep active work grounded without pasting the whole specification.

Type: Relational-precision specialization Status: Stable Normativity: Normative unless explicitly marked informative

At a glance. Use A.6.H when words such as whole, part, integrity, complete, turnkey, or end-to-end hide the exact object or relation on which a decision depends.

Use this when. Enter after A.6.P:4.11 has recovered the concrete candidate objects and the sentence needed by the receiving use, and that sentence genuinely asks about a whole, part, structure, integrity, coverage, or completion. A.6.H helps the practitioner expose the candidate whole or other bearer, its boundary when relevant, the independently identified parts or constituents, and the exact direct claim that must leave to its owner.

Not this pattern when. Do not enter merely because a source contains a trigger word. A characteristic or measurement claim goes to C.16.P/C.16; an evidence or assurance claim goes to A.10/B.3; an episteme identity or edition claim goes to C.2.1; a publication claim goes to E.17/E.24.PUB; a Method, WorkPlan, or Work claim goes to its direct A.3/A.15 owner. Stay in A.6.P when the candidate objects or the receiving sentence are still unknown.

What goes wrong if missed. A situation record, diagram, bundle, adjective, phase label, or coverage slogan becomes the supposed whole or relation. Parts, members, portions, phases, method factors, Work parts, evidence, and measured characteristics are then silently treated as one generic “part of” claim.

What this buys. A short identity-first route from overloaded prose to one or more direct claims with exact participants and owners, or to an explicit blocker when a needed predicate is absent.

What changes in practice. The practitioner stops annotating a wholeness bundle and instead writes the few direct sentences the next decision consumes: which entity, which relation and participants, which owner, and which stronger inference remains blocked.

Natural language compresses several different engineering questions into the same small vocabulary:

#Keywords

• wholeness
• integrity
• part-of
• boundary
• environment
• mereology
• completeness
• order/time
• publication-carrier and EntityOfConcern/Description distinction
• role-method-work.

#Relations

#Content

#Problem frame

Natural language compresses several different engineering questions into the same small vocabulary:

• What individual is being treated as one whole?
• Where is its boundary, and what lies outside it?
• Which independently identified objects are parts, constituents, members, portions, or proper temporal restrictions?
• Which relations among those objects actually obtain?
• Does a named use need a construction trace or a selected structure?
• Is the same whole being recognized again, or must it be reidentified?
• Is “complete” about performed Work, capability, specification, evidence, or another exact coverage claim?
• Is “integrity” a measured characteristic, an assurance claim, or a claim that an assembled entity remains one whole?

Those questions have different participants, predicates, and owners. A.6.H does not answer them by creating a common wholeness object. It keeps the source wording readable while making the load-bearing claims exact.

A word is load-bearing here when a requirement, invariant, interface statement, architecture choice, model relation, decision, test oracle, assurance use, or downstream action depends on its interpretation. E.10 owns shared wording-use discovery. A.6.H begins only after the current wholeness-family claim has been selected by value.

#Problem

Without an exact-object discipline, the following failures recur:

1. Candidate-whole ambiguity. “The whole system” is asserted before one candidate entity, boundary, or identity rule is recoverable.
2. Reference-level drift. One noun phrase alternates among a referent, a claim-bearing episteme, a publication form or carrier, intended Work, performed Work, and evidence.
3. Parthood overload. Physical components, conceptual constituents, collection members, measured portions, and temporal restrictions are written as one generic inclusion.
4. Order-as-structure. A method factor, step description, plan item, or performed occurrence is treated as a component because a diagram places it inside a box.
5. History-as-parthood. A v2, revision, edition, shift, retry, or monitoring window is routed through PhaseOf before episteme identity or Work-temporal law is applied.
6. Construction-by-list. A list of objects, repeated trace, or selected diagram is treated as proof that one whole or direct relation obtains.
7. Coverage-as-wholeness. “Complete”, “turnkey”, or “end-to-end” is treated as a whole-level property without a scope, covered items, direct coverage or completion predicate, or current Work state.
8. Integrity collapse. A measured characteristic, security or data-integrity term, evidence report, assurance claim, and structural-whole claim are all forced through mereology.
9. Change-by-vocabulary. Generic verbs such as recompose, rephase, or recomplete replace the exact changed object and direct changed relation.

The practical failure is non-decidability: another reader cannot tell which object is at issue, what relation is claimed, what evidence would bear on it, or which stronger use is blocked.

#Forces

#Solution

#Entry and result contract

Enter with:

• one exact sentence or decision that depends on wholeness-family wording;
• the concrete candidate objects recovered under A.6.P:4.11;
• the receiving use that would change if the wording were read differently; and
• any already known direct owner.

Return one of:

1. one or more readable direct claims, each naming the predicate or claim family, ordered participants, material qualification, and governing pattern;
2. an explicit owner-return when a measurement, evidence, episteme, publication, method, plan, Work, production, or completion owner already governs the question; or
3. an A.6.RCD missing-governor[...] result after the exact participants and needed sentence are known but no current direct predicate closes the use.

When evidence cannot yet select among several readings, keep the candidate objects, discriminating questions, and blocked receiving use explicit in ordinary prose. Do not turn that temporary uncertainty into a wholenessSituation, card, bundle, lifecycle record, or new U-kind.

#Apply the exact-object sequence

Use the following sequence only as far as the current sentence requires:

1. Recover the working question. State what a reader must decide, do, accept, measure, rely on, start, continue, or stop. The cue word selects no branch.
2. Name the subject and reference level. Distinguish the referent entity, claim-bearing episteme, publication occurrence, publication form, presentation carrier, Method, MethodDescription, WorkPlan, performed Work, and evidence carrier. Keep only the subjects current in this case.
3. Recover a candidate whole only for an actual whole claim. Identify the candidate individual, its direct identity owner, relevant boundary or delimitation, environment, and at least one interaction, dependency, or constraint across that boundary when the use needs it.
4. Identify the alleged parts independently. A label, location, list, graph node, file section, timestamp, or common name does not identify a part. Recover each component, constituent, member, portion, temporal restriction, Method factor, Work part, or other object under its own owner.
5. State every direct relation occurrence separately. Name exact participants and test the direct predicate. A relation obtains neither because the whole was recognized nor because a trace, view, or record lists it.
6. Add construction or selected structure only when the receiving use consumes it. C.13 may report already recovered parts, relations, constraints, and a construction rule. A.22 may identify one selected structure when its selection basis and identity discriminators are current. Neither creates the direct facts.
7. Recognize or reidentify the whole only when that question is current. Use A.1 for holon recognition and B.2 for a remaining whole-reidentification question after direct existing-whole explanations have been tested. A changed adjective or part list alone decides neither.
8. Separate coverage, completion, and performed Work. Name what is covered, under which scope and criterion, by which exact relation, and whether the claim concerns a MethodDescription, capability, plan, Work occurrence, production result, evidence set, or another subject. Use A.15.1/A.15.PROD or the exact direct owner; do not treat a plan as performed Work.
9. Stop at the direct owner. A.6.H owns the unpacking move, not the recovered whole, relation, characteristic, Work, evidence, or verdict.

#Classify integrity by the claim it carries

The word integrity never chooses mereology by itself.

If the source leaves these readings genuinely open, preserve the alternatives and block the named use until evidence discriminates them.

#Select the direct relation, not a generic part edge

#Unpack complete, turnkey, and end-to-end

Ask what the next reader may do because the claim is supposedly complete.

A sentence may require several rows. Write several direct claims; do not bundle them back into one “wholeness” record.

#Use wording as a cue, not as ontology

The following recurring expressions are useful review cues, not a second trigger registry:

• whole, entire, integrated, coherent, holistic — ask whether there is an actual candidate whole, a measured or assurance claim, or only rhetoric;
• part, piece, component, module, element, subsystem, includes, contains, comprises — recover the object and direct relation rather than accepting the noun;
• phase, version, revision, edition, lifecycle — apply the direct identity owner before any history label;
• complete, turnkey, end-to-end, fully specified — recover the exact coverage or completion claim;
• pipeline, workflow, process, step, stage — distinguish Method, MethodDescription, WorkPlan, performed Work, order relation, and publication representation;
• collection, group, team, set — distinguish membership, acting system, role assignments, and selected collection structure;
• context, environment, discipline as a whole — name the actual bounded context, episteme family, community, organization, or other subject before making a boundary or nesting claim.

When a cue occurs inside a defined term of art, retain the definition and direct owner. Open A.6.H only if the sentence also makes an unresolved whole, part, structure, coverage, or completion claim.

#Describe change through the object that changed

When a wholeness-looking story changes, name the exact object and direct relation:

• a different boundary or interaction claim returns to the boundary/delimitation owner;
• an added, removed, or differently related part returns to the direct parthood, membership, portion, or structure owner;
• changed episteme content, EntityOfConcern, or effective ReferenceScheme identifies another episteme under C.2.1, with edition continuity tested separately;
• a different publication form, occurrence, or carrier stays with publication and carrier owners;
• a changed Method, MethodDescription, WorkPlan, Work history, production result, or completion claim stays with its own owner;
• a changed coverage scope, criterion, evidence set, or assurance use repairs that direct claim rather than a generic completeness status.

Do not substitute a generic change lexicon for those objects and predicates. A readable verb is welcome when the exact direct claim remains recoverable.

#Guardrails

1. No situation record, card, bundle, adjective, table, graph, or trace is the whole or direct relation by presence.
2. No generic partOf closes a load-bearing claim when a direct relation kind or owner is required.
3. No order, plan, or Work history is structural parthood by position.
4. No membership is upgraded to component assembly or acting-system identity.
5. No cross-boundary flow or influence is treated as a part merely because it crosses the boundary.
6. No integrity reading is selected before the bearer, claim, and receiving use are known.
7. No plan, description, or publication is treated as performed Work.
8. No version, revision, edition, phase, filename, or provenance label decides identity or continuity.
9. No construction trace or selected structure creates its listed parts or relations.
10. No coverage statement becomes assurance, acceptance, readiness, or completion beyond its exact predicate and evidence.

#Archetypal Grounding

#Assembled pump

Source sentence: “After seal replacement, the assembled pump remains an integral whole.”

1. The subject is PumpUnit-37, not the maintenance record, drawing, or seal-replacement Work.
2. The pump's direct identity rule decides whether the same individual continued.
3. The current use names the pump boundary, impeller, casing, replacement seal, and the exact assembly or parthood relations on which operation depends.
4. If the decision consumes one selected organization of those relations, A.22 governs that structure; if it consumes a construction account, C.13 reports already recovered facts.
5. A.1 recognizes the candidate whole; B.2 opens only if the replacement leaves a genuine whole-reidentification question.
6. Calibration, seal replacement, and inspection remain separately governed Work and change facts. The adjective integral proves none of them.

#Laboratory pipeline

Source sentence: “The whole chromatography pipeline is turnkey, and the chemist owns the whole thing.”

The repair produces several claims:

• the reusable procedure is one exact Method or composite Method under A.3.1/B.1.5, with exact joins and exposed interactions;
• its procedure document is a separate U.MethodDescription episteme under C.2.1/A.3.2;
• “turnkey” becomes the exact specification, capability, Work, or evidence coverage claim needed by the receiving use;
• the chemist is an admitted holder only through the exact role assignment, commitment, permission, or responsibility relation that obtains; and
• an actual laboratory run is dated Work under A.15.1.

No one claim is a component relation merely because the source uses pipeline or whole.

#Paper, proof, and revision

Source sentence: “Section 3 is part of the proof, and v2 is part of v1.”

• Recover whether Section 3 is a constituent of the paper episteme, a publication-unit constituent, or a described proof step. Keep those subjects separate.
• Recover argument order under its direct owner rather than as physical containment.
• Compare the exact C.2.1 triples for the two labelled epistemes. Changed claim content identifies two epistemes. Assert EpistemeEditionRelation(E_v1,E_v2) only when its historical-continuation predicate obtains.
• If one unchanged episteme is needed only during a proper interval, PhaseOf(E@τ,E) may state that restriction. It does not connect v1 to v2.
• Drafting, review, and publication are Work and publication relations, not participants of the edition relation.

#Integrity measurement and assurance

Source sentence: “The structural integrity score is 0.82, so the system is assured.”

First recover the bearer, integrity Characteristic, Scale, measurement method, result episteme, evidence, and time stance under C.16.P/C.16. Then ask whether a named B.3 assurance claim is actually being made and recover its claim, evidence-use relation, scope, limitations, and relying context. The number does not create a candidate whole, a part relation, or an assurance result.

#Recognition and assurance stay separate

Recognition questions decide which objects and direct relations are current:

• Is there one candidate whole under A.1?
• Which independently identified parts, constituents, members, portions, or temporal restrictions participate?
• Which direct relations obtain?
• Does a selected structure or construction account matter to this use?
• Does the same whole persist, or is reidentification current?

Assurance questions decide what may be relied on:

• Which claim is being supported?
• Which evidence bears on it through which relation?
• What scope, limitation, time stance, and relying use apply?
• Does the evidence support recognition, relation truth, measurement, completion, or another claim?

Evidence can make an assertion inspectable without becoming constitutive of the whole or relation. Unknown support does not create a third identity or obtaining state.

#Bias-Annotation

• Governance bias. The pattern favors reviewable direct claims over rhetorically satisfying wholeness language. Ordinary prose and explicit unresolved alternatives mitigate this when a decision is not yet due.
• Architecture bias. It favors exact owners and small typed vocabularies over one reusable wholeness schema. The minimum-current-object rule mitigates unnecessary apparatus.
• Ontological/epistemic bias. It insists on separating referent, episteme, publication, Method, plan, Work, and evidence. This cost is paid only when the distinction changes the receiving use.
• Pragmatic bias. It favors early disambiguation to avoid downstream refactoring. A local direct sentence is sufficient; reusable declarations or structures are added only for a named receiver.
• Didactic bias. It uses recurring cue words and worked cases to teach the route, while E.10 remains the shared wording-use owner and the cue list creates no second registry.

#Conformance Checklist

#Common Anti-Patterns and How to Avoid Them

#Consequences

The practical test is simple: if “whole” matters, name the thing, the relation, and what the reader may do with the claim.

#Rationale

Wholeness language is useful because it compresses boundary, identity, relation, construction, coverage, and assurance into ordinary speech. The same compression becomes dangerous only when downstream work relies on one particular reading.

The minimal repair is therefore not a richer wholeness schema. It is an exact-object sequence that starts with the working decision, recovers only the objects that decision consumes, and exits to their direct owners. This preserves conversational economy while preventing a representation, record, label, or adjective from replacing an in-world object or relation.

The sequence also preserves two positive uses often lost in blanket cleanup. PhaseOf remains valid for a proper restriction of one unchanged enduring individual, including one unchanged episteme when its C.2.1 identity triple is fixed. And ordinary whole recognition remains useful when an exact candidate entity, boundary, parts, relations, and direct identity rule are genuinely current.

#SoTA-Echoing

Whenever fraction, percentage, or share is used as a part claim, recover the extensive measure μ and additivity basis before PortionOf; otherwise keep the value with its measurement, allocation, membership, or other direct owner.

#Relations

• Specialises: A.6.P after its 4.11 whole/part/integrity/coverage branch has recovered exact candidate objects and the receiving sentence.
• Returns direct mereology to: A.14; construction accounts to C.13; selected structures to A.22; holon recognition to A.1; and remaining whole reidentification to B.2.
• Returns episteme and publication questions to: C.2.1, A.3.2, E.17, E.24.PUB, and C.29 as selected by the exact subject.
• Returns Method, plan, and Work questions to: A.3.1, B.1.5, A.15.2, A.15.1, A.15.PROD, and B.1.4 only for bounded aggregation of already recovered temporal or order relations.
• Returns integrity, evidence, and assurance questions to: C.16.P, C.16, the exact measurement owner, A.10, and B.3.
• Returns absent relation governance to: A.6.RCD after participants, required predicate, and blocked receiving use are exact.
• Does not create: a wholeness situation, card, bundle, lifecycle kind, automatic edition series, universal part relation, coverage status, assurance verdict, or direct relation occurrence.

#A.6.H:End

---

Last Updated: 2026-08-04 — upstream FPF commit 7ba40a95 (github.com/ailev/FPF)