Research

The Fourth Law

A system may change anything about itself except its correspondence with reality and its continuity of purpose.

Classical logic has three laws: identity, non-contradiction and the excluded middle. They describe things that stay still. The Fourth Law of Logic is the framework CRIS works from for things that change: it holds that a system stays itself through change as long as it keeps corresponding to reality and keeps its purpose.

The formula

What each symbol means:

SymbolNameMeaning
EᵢExistenceA real-world condition, measured by how far it has drifted from a stable state (its entropy)
IᵢIntelligenceThe knowledge, people, tools and resources applied to that condition
↔CorrespondenceWhether the intelligence actually matches the condition
ΣAccumulationAdding up correspondences across every condition, over time
CContinuityThe result: how well a system holds together as it changes

In Recursion Without Rupture, the same symbols are read for self-improving systems: Eᵢ as observation and Iᵢ as reasoning. This page reads them more generally, as a real-world condition and the intelligence applied to it. The two readings are being aligned.

The continuity simulation

The simulation below shows the idea for one condition at one moment. Pick a condition, change the intelligence and the support around it, and see whether they match. It is a picture of the framework, not a measure of any real place. Every run has its own link, so you can share it or come back to it.

Illustrative simulation of the CRIS continuity framework. Not an established scientific law.

Model 1.0 · Illustration. Runs on this page. The data file, build tests and run links are in place, awaiting sign-off as a tested illustration. The model has not yet been reviewed by people outside CRIS.

Simulation inputs

Starting values are illustrative.

Deficit (E > I)
68.5% effective intelligence as a share of the condition’s entropy
Margin: −28 points (effective intelligence 61 minus entropy 89)

They are not yet enough. More capability, coordination or resources would be needed. A picture of the idea for one condition, not a measure of any real place.

One condition · one moment · illustrative numbers. How the model works · Model 1.0

Stage 1 · Observe

Identify a real-world condition (Eᵢ, Existence)

The condition being evaluated: Water scarcity. Its starting value is an illustrative starting point, not a measurement.

Stage 2 · Observe

Measure its entropy

Entropy is how Eᵢ is measured here: how far the condition has drifted from a stable state.

89
entropy of the condition, from 0 (stable) to 100
Stage 3 · Reason

Intelligence available (Iᵢ, Intelligence)

What can be brought to bear: Human expertise, Engineering teams, Scientific models, Sensors, Data collection, Compute resources, Community participation. Stewardship, verified compute, infrastructure and governance strengthen or weaken it.

61
effective intelligence (base 61)
Stage 4 · Verify

Correspondence (↔)

The condition is compared with the intelligence applied: Eᵢ ↔ Iᵢ = 89 ↔ 61.

Stage 5 · Verify

Outcome

Deficit (E > I)

They are not yet enough. More capability, coordination or resources would be needed.

Stage 6 · Accumulate

Continuity score (C), for one condition

68.5%
For one condition, the correspondence is shown as a percentage. The formula’s sum across conditions and over time is not modelled here.

About the framework

How CRIS systems are designed to use the idea. Nothing in this panel is computed by the simulation, and none of these systems is running today.

Formerly stage 7

Inside CRIS systems

The same comparison is designed to run inside CRIS systems: a condition is detected, interpreted and checked for continuity before a response is given.

Reasoning Core Status to be confirmed
CRL Engine Status to be confirmed

Formerly stage 8

Across places

The framework is designed so that a deficit in one place could be met by capacity in another, through shared evaluations, for example on hosting such as LOGOS-C, under a written agreement.

LOGOS-C Related organisation

Formerly stage 9

At every scale

The same test applies at every scale; only the inputs change. The simulation does not compute results at any of these scales.

Individual · Family · Community · Institution · City · Nation · Continent · Planet

Formerly stage 10

Where it sits

Fourth Law → continuity mathematics → LOGOS-C → Reasoning Core → CRL Engine → Botswana Global Testnet Initiative → SCVX → CIX.

CIX Status to be confirmed
LOGOS-C, the Botswana Global Testnet Initiative and SCVX Related organisation

LOGOS-C, the Botswana Global Testnet Initiative and SCVX. An organisation also led by our Founder. Any work with CRIS is under a separate written agreement on arm’s-length terms. How we work with related organisations. SCVX states its own licence position at scvx.global.

The simulation in pictures

Sixteen pictures explain what the simulation does, what it assumes, how it is checked and how it is built. The model in full is on How the model works.

What changed on this page. Each of the seven blocks, and what this update did to it.

The Fourth Law page: seven blocks, and what this update did to each1 The leadK1 · now hedged2 The formulakept3 The symbol tableK8 · one name per symbol; book note4 Try itreplaced by a plain introduction5 Banner, panel, ten stagesK2–K7, K9–K12 · result card, six stages, run links6 The bylaws paragraphkept, tied to the loop7 Where to go nextdrafts marked phase P3changed in this updatekept as it was
The Fourth Law page: seven blocks, and where the twelve findings sat.

Who reads it. Five kinds of reader, and what each now gets.

Curious visitorArrives fromhome pageNeedswhat the scoremeans afterone runNow getsthe result cardResearcheror reviewerArrives fromthe book, acitationNeedsevery assumption,written downNow getsHow the modelworksBuilderArrives fromdeveloper hubNeedsreproduce a runfrom a versionand inputsNow getsrun links;changelogPartneror officialArrives froma meetingNeedsthat it isillustrative; whichsystems are realNow getsbanner; About theframeworkBook readerArrives fromRecursion WithoutRuptureNeedshow the tworeadings of thesymbols relateNow getsthe symbol note
Who reads the simulation: five readers, and what each now gets.

The formula and the model. What the simulation uses for each symbol, and what it does not model.

C = Σ ( Eᵢ ↔ Iᵢ )This pageSimulationThe bookEᵢExistenceentropy slider0–100observationIᵢIntelligencesliderthen four modifiersreasoning↔Correspondenceeffective I ÷ E × 100cappedverificationΣAccumulationnot modelledone condition onlyverified improvementsCContinuitythe capped ratioas a %purpose that holdsThe two readings are being aligned. This page reads Eᵢ as a real-world condition; the book reads it as observation.
Formula and model: what the simulation uses for each symbol. The red box is what it does not model.

Inputs and assumptions. What you set, and what the model assumes, now in the open.

What the reader setsWhat the model assumes, now shownCondition (sets the starting E)Entropy of the condition (Eᵢ)Available intelligence (Iᵢ)Stewardship · modifierVerified compute (CTU) · modifierInfrastructure · modifierGovernance · modifierEqual weights: 0.25, 0.25, 0.25, 0.2550 is neutral; ±50% at mostMatch band: ±2 pointsScore cap: 100Entropy floor: 1Starting values: illustrativeDerived bars removed (stage 2)All of these are read from data/fourth-law-sim.json and printed on the model page.
Inputs and assumptions: what the reader sets, and what the code assumes, now in the open.

One run. Four lines of arithmetic, worked with the default run.

Four modifiersSCH, CTU, Infra, Goveach 50 hereSupport sweighted mean= 50Effective II × (1 + 0.5(s−50)/50)= 61Scoremin(100, I_eff ÷ E × 100)= 68.5%Outcomemargin -28 vs ±2= deficitWorked here with the default run: water scarcity, E = 89, I = 61, every modifier at 50.The same four lines run in the build and in your browser, from one core file, so the two never disagree.
One run, in four lines of arithmetic: sliders to support, effective intelligence, score and outcome.

The ten stages and the loop. The six stages that compute follow the same loop as the book and the proposed bylaws: observe, verify, reason, accumulate, benefit and continue.

The ten stages, beforeThe six-stage loop (book; proposed bylaws)1 Identify a condition2 Measure entropy3 Intelligence available4 Correspondence5 Outcome6 Continuity score7 How the runtime processes it8 Coordination across places9 Every scale10 Where it sitsObserveVerifyReasonAccumulateBenefit · not shown in the simulationContinue · not shown in the simulationStages 7–10 → About the framework, with status from data
Ten stages and the six-stage loop: a proposed mapping (Q4). The simulation keeps the six that compute.

The result card. What each part of the result means.

DEFICIT (E > I)68.5%effective intelligence as a shareof the condition’s entropyMargin: -28 pointsA picture of the idea for one condition,not a measure of any real place.One condition · one moment · illustrative · Model 1.0outcomescoremargin (new)plain readinglimits and version
The result card: outcome, score, margin, a plain reading and the limits. The only part of the section announced to screen readers.

How it is checked. Three levels, and what would count against each.

Level 1 · CodeDoes the code do what the page says?The worked runs and edge cases are build testsin place: tests pass on every buildAgainst it: A test failsLevel 2 · ModelIs the model a fair picture of the formula?Outside review; sensitivity runs on each assumptionplanned for phase P3; reviewers to name (Q6)Against it: One assumption decides most outcomesLevel 3 · IdeaDoes continuity, measured this way, predict how real systems fare?Predictions written before data; a published methodnot claimed on this pageAgainst it: Predictions fail against real data
Validation: three levels of checking, and what would count against each.

Reproducible runs. One data file and one core feed every use.

data/fourth-law-sim.jsonmodel 1.0: conditions,params, weights, testssim-core.jsone core, no randomnumbersServer renderworks without scriptsYour browserevery slider moveBuild testsseven casesModel pageprints every valueRun link#simulation?c=…&v=1.0Same inputs and the same model version always give the same result.
Reproducibility: one data file and one core feed every use.

How a claim reaches this page. Strong claims go to a paper, not here.

1 DraftResearch andEngineering2 Tag its leveldescriptionto standing3 Team checkand build tests4 Outside reviewmodel claims only5 Sign-offFounder; counselfor standing wordsDescriptionteam checkCode claimteam check, tests passModel claimoutside review, note, FounderFramework claimFounder sign-offStanding claim (law, proven, scientific, predicts)Founder and counsel; usually a paper
Review path: five steps from a drafted claim to the page, and the five levels of claim.

Connections. Theory above; systems and related organisations below, each labelled.

Theory and links inHome page →The book →The bylaws →Mission →The simulation#simulation · an illustrationCRIS systems, with status from dataRelated organisations, with the related-party lineReasoning Coreto be confirmedCRL Engineto be confirmedCIXto be confirmedLOGOS-CrelatedBGTIrelatedSCVXrelatedNamed only in About the framework, as how systems are designed to work; the simulation calls none of them.
Connections: theory above; systems and related organisations below, each labelled.

Maturity. The simulation’s own ladder.

ConceptThe formula and words, no runnable modelIllustration ← todayRuns on the page; labelled illustrativeTested illustrationVersion, data file, build tests and run links, signed offReviewed modelOutside review of the model against the formula, publishedResearch toolUsed with real data in published workMoves up one rung at a time, with a changelog row naming who moved it. Read from data.
Maturity: the simulation’s own ladder, shown under the banner.

Where the detail lives. The simulation stays here, with four pages beneath it.

#simulationstays on the Fourth Law pageHow the model works →/simulation/model/Three runs, read →/simulation/reading-results/How it is checked →/simulation/validation/Model changelog →/simulation/changelog/Each page links back with “Back to the simulation”. The home page link to #simulation is checked by the build.
Section map: the simulation stays, with four pages beneath it. Select one to open it.
data/fourth-law-sim.jsonversion · maturity · bannersymbols: one name eachconditions (source: null)params · modelloop mapping · scalestests · changelog · claimsThe corecomputes every runThe model pageprints every valueThe build testscheck every caseThe maturity lineunder the bannerA change to model, conditions or params needs a new version and changelog row; the build compares a fingerprint.
Data model: one file read by the core, the model page, the tests and the maturity line.
BannerMaturity lineInputs panelResult cardSix stagesAbout the frameworkPicturesFour pagesabove every resultmodel version and rungcondition, six sliders, reset, run linkoutcome, score, margin, reading, limitseach tagged with its loop stepstages 7–10, status from datasixteen figuresmodel, runs, checks, changelog
The rebuilt section: each block and why it is there.
1 Data filebuilt2 Core readsdatabuilt3 Build testsbuilt4 Resultcard, linkbuilt5 Stages andAboutbuilt6 Live regionbuilt7 Lead andnotedrafted; Q28 Four pagesbuilt9 Buildchecksbuilt10 Publish1.0waiting11 Outsidereviewwaitinga failed test or review goes back to the core
Build pipeline: a failed test or review goes back to the core.

How the Fourth Law governs CRIS

Our proposed bylaws would adopt the Fourth Law of Logic as the framework that guides the Board's judgment, through five operators across six stages and a written Fourth Law Review for major decisions. They are approved by the Board for implementation. The simulation’s six stages follow the same loop. Read the Fourth Law in the bylaws.

Where to go next

Owner: Research and EngineeringNext review: 2027-03-01