Method profile 08Didactic reasoning frameworkOperational record projection

DIAL-4

Keep evidence, implication, utility and bounded validity from collapsing into one undifferentiated claim

DIAL-4 separates four reasoning modes: Deduction, Inference, Abatement and Legitimation. The framework asks what the source supports, what it suggests, what burden it may reduce, and what remains valid after scope control and verification. ChangePlane implements a related claim-record surface, but its six recommended claim types are not a one-to-one encoding of the four conceptual modes.

Four modes, four burdens

Each mode asks a different question and protects against a different category error.

Evidence

Deduction

What can be responsibly concluded from the source? Direct support is required.

Meaning

Inference

What does the source reasonably suggest or imply beyond its direct findings?

Utility

Abatement

What burden, cost, dependency, friction or uncertainty may be reduced?

Bounded validity

Legitimation

What survives explicit scope control, method discipline and verification?

Interactive reasoning-mode topology

Drag and inspect the four lenses around the source. The arrows express reasoning dependency, not an obligatory linear workflow.

Mode

DeductionInferenceAbatementLegitimation

Utility is not proof.A claim that something may reduce cost or friction remains an Abatement claim even when the practical benefit is compelling.

Synthetic claim-classification laboratory

Mark the function of a claim. Mixed signals produce a split recommendation rather than a false single label.

DEDUCTION

Suggested reasoning mode

This laboratory does not read or validate a real source. It demonstrates the category rules only.

DIAL-4+ grading surface

The conceptual framework adds confidence, evidence quality and scope. These axes affect handling but do not create calibrated truth probabilities.

RETAIN WITH CAVEAT

Handling suggestion

Conceptual modes versus ChangePlane claim records

The ChangePlane record operationalizes claim separation, but it does not encode the four framework modes as a closed enum.

No normative one-to-one mapping.source_fact, interpretation, implication, decision, inference and verification are a recommended operational vocabulary. They do not replace Deduction, Inference, Abatement and Legitimation as the conceptual four-mode model.

ChangePlane record builder

Construct a browser-local representation of the implemented Dial4Claim shape.

DIAL-4

Record representation

Protocol-tag switch.ChangePlane reports DIAL-4+ when confidence and/or strength is set. Supporting evidence, weaknesses and uncertainty are present on both record shapes.

The DIAL-4P naming collision

The same label denotes two different systems. The expansion or repository context is mandatory.

Possibility

DIAL-4P Possibility

Combines DIAL-4 claim type, DIAL-4+ claim strength, Diátaxis document form and 5PP possibility governance.

ChangePlane

DIAL-4P Projection/Productization

Renders an existing trace into a derived artifact whose source_truth is structurally false.

Development lineage

Public explanation, private first-party provenance and executable operational records are different evidence classes.

Structural and analytical views

These graphs describe the framework and its record surfaces. They are not efficacy measurements.

Analytical coding

Reasoning burden by mode

Values encode the profile's source-derived interpretation of each mode's emphasis.

Source structure

Conceptual and operational surfaces

Counts are not maturity scores.

Reasoning modes

Deduction, Inference, Abatement and Legitimation.

Recommended claim types

ChangePlane operational vocabulary.

Conceptual strength axes

Confidence, evidence quality and scope.

Record grading/support fields

Confidence, strength, evidence, weaknesses and uncertainty.

DIAL-4P meanings

Possibility and Projection/Productization.

MVP-4 new tests

Source-reported mixed protocol increment.

MVP-4 total tests

Source-reported repository total at that increment.

Independent effect studies

Identified in the inspected source set.

Evidence classification

Implementation and provenance support structural claims. They do not establish calibrated reasoning quality.

Claim	Evidence class	Status
DIAL-4 names four distinct reasoning modes	First-party framework source and public explanation	Supported
ChangePlane stores DIAL-4 and DIAL-4+ claim records	Source implementation, schemas and committed tests	Supported at pinned commit
ChangePlane's six claim types exactly implement the four modes	Semantic equivalence	Not supported
DIAL-4+ conceptual axes are fully encoded in ChangePlane	Schema equivalence	Not supported; explicit evidence-quality and scope fields are absent
Protocol records can enter a seal-bound trace	Implementation and committed protocol tests	Supported within documented seal boundary
DIAL-4 improves accuracy or reduces overclaiming in practice	Outcome efficacy	Not independently established

Evaluation interpretation and boundaries

The framework is best understood as claim hygiene and didactic separation, not an automatic truth engine.

Problem definition

Technical writing and AI output often place direct findings, implications, business benefits and bounded verdicts in the same paragraph at uniform confidence. The reader cannot tell what kind of claim is being made.

Mechanism hypothesis

Explicit claim typing, strength grading and scope-aware legitimation should make category errors, hidden inferential jumps and utility-as-proof easier to detect. This is a design hypothesis.

Possible evaluation

Measure inter-rater agreement on claim types, unsupported-deduction rate, inference-to-fact conversion, utility-overclaim rate, caveat survival, scope leakage, reviewer correction time and downstream decision quality.

Failure modes

Labels can become decorative, categories can be assigned inconsistently, a legitimation pass can merely restate conclusions, confidence can be uncalibrated, and a record schema can create a false impression of rigor.

Explicit non-claims

A DIAL-4 label does not prove that a claim is true.
Deduction in this framework is evidence-bound claim handling, not a formal-logic proof system.
Inference does not become deduction because confidence is high.
Abatement does not prove feasibility, maturity or return on investment.
Legitimation does not add evidence that was absent from the source.
DIAL-4+ confidence is not automatically calibrated.
ChangePlane's claim-type vocabulary is free-form and not a closed semantic validator.
The two DIAL-4P meanings are not interchangeable.
Committed tests do not establish real-world reasoning outcomes.
No independent annotation-reliability or efficacy study was identified.

Sources and reproducibility

Claims remain bounded to the sources and commits named below.

Concept source pinfbratten/control-center-ops @ 7485ee6ade639c3928f043c268995269e75aca54

Framework source04-Knowledge/Systems/Prompts and Prompting/Prompts library/DIAL-4.md

Development provenanceKB/inbox/user-input/DIAL-4 - DIAL-4+ - and DIAL-4P.md

Public explanationadaptivearts.ai/blog/from-reasoning-mode-to-possibility-state/

ChangePlane pinfbratten/changeplane @ 32cb7840915df7f0d527044e37242cd6123801e7

Implementation incremente37da0129d2541f28bdb68a786089b6d0a1b3e3d

Record modelsrc/changeplane/protocols/models.py

DIAL-4 schemaschemas/changeplane.protocol.dial4.v1.schema.json

DIAL-4+ schemaschemas/changeplane.protocol.dial4plus.v1.schema.json

Teststests/test_protocols.py

This page performs no source retrieval, model call, repository write, ChangePlane command, protocol insertion, projection rendering or seal verification. All laboratories use synthetic browser-local state.