0. Plain Statement

A system can improve its visible success proxy while losing coherence.

Plain-language version:

A system can look more successful while becoming less coherent. When the visible success signal rises while coherence falls, the system is entering inversion.

1. Formal Definition

The Success Proxy Divergence Law states that visible success is not sufficient evidence of coherence.

A system may improve a proxy such as profit, benchmark score, compliance rate, status, output volume, incident count, engagement, symptom reduction, speed, or stability while degrading the deeper conditions that make the system coherent.

This law describes one of the central corruption patterns in UTS:

visible success rises while real coherence falls

When this occurs, the system’s selection target has diverged from coherence. The proxy begins to steer the system away from the conditions it was supposed to represent.

Success proxy divergence is the operational failure expression of LAW-001. LAW-001 states that coherence must precede optimization. LAW-003 describes what happens when the proxy outranks coherence in practice.

2. Canonical Form

Φ↑ while O↓ ⇒ ι↑

Expanded canonical form:

visible success proxy increases while coherence decreases, producing inversion

Related variables:

O, Φ, H, ι, Au, R, µᵢ, BΣ, K, ε

Where:

3. Core Mechanism

The Success Proxy Divergence Law usually unfolds through proxy capture.

General sequence:

proxy selected
→ proxy optimized
→ proxy becomes steering target
→ coherence constraints treated as friction
→ O declines
→ Φ continues rising
→ H accumulates
→ ι rises
→ visible failure appears late

The proxy originally functions as a signal. Over time, the system begins treating the signal as the target.

Coherent proxy use

Φ used as diagnostic
→ checked against O, H, Au, R, BΣ, µᵢ, K
→ proxy remains subordinate
→ optimization remains coherence-preserving

Divergent proxy use

Φ used as target
→ proxy gain prioritized
→ coherence checks weaken
→ hidden debt accumulates
→ inversion forms

The system becomes increasingly skilled at improving the appearance of success while losing the structural basis that made success meaningful.

4. When This Law Applies

This law applies when a visible success signal improves while deeper coherence indicators degrade, remain unmeasured, or become harder to audit.

Common success proxies include:

• profit
• growth
• valuation
• benchmark performance
• compliance rate
• case closure rate
• engagement
• output volume
• response speed
• incident reduction
• symptom reduction
• public approval
• market share
• institutional stability
• policy adherence
• credential accumulation
• model accuracy
• throughput
• status
• symbolic certainty

The law applies strongly when:

Φ↑ is used to claim O↑ without trajectory validation

or when:

Φ↑ coincides with H↑, Au↓, R↓, BΣ↓, µᵢ↓, or K↓

Typical domains:

5. When This Law Does Not Apply

This law should not be used to distrust every success signal.

A success proxy can be valid when it remains tied to coherence.

This law does not apply when:

• visible success rises and coherence also rises;
• proxy movement is validated against hidden debt, auditability, restoration capacity, and recurrence;
• the proxy is used as a diagnostic, not a replacement for truth;
• gains are achieved without boundary degradation or hidden cost export;
• success persists under stress, time delay, and cross-scale validation;
• restoration capacity remains adequate;
• affected-node integrity is preserved.

False-positive cases:

Important distinction:

The problem is not success. The problem is success becoming detached from coherence.

6. Diagnostic Signature

The basic diagnostic signature is:

Φ↑ while O↓ ⇒ ι↑

A stronger warning signature:

Φ↑
O↓ or unvalidated
Au↓
H↑
R↓
BΣ↓
µᵢ↓
K↓
ε low or delayed

Common indicators:

Additional diagnostics:

7. Failure Pattern

If ignored, this law produces inversion and pseudo-coherence.

General failure pathway:

proxy improves
→ success is declared
→ coherence checks weaken
→ hidden debt accumulates
→ affected layers absorb cost
→ inversion stabilizes
→ system defends the proxy
→ visible collapse or legitimacy shock appears late

Common failure modes:

• Pseudo-Coherence — the system appears ordered while coherence declines.
• Success Proxy Capture — the visible signal becomes the steering target.
• Goodhart Collapse — the metric loses meaning because it is over-optimized.
• Metric Substitution — measurable success replaces real system health.
• Hidden Debt Accumulation — unresolved incoherence is deferred or exported.
• Silent Extraction — coherence is consumed while visible error remains low.
• Pseudo-Restoration — optics improve while actual repair does not occur.
• Pseudo-Security — security signals improve while the system becomes less secure.
• Forced Profit — profit rises through extraction rather than coherent circulation.
• False Recovery — symptoms improve while systemic recovery remains incomplete.

Compact failure signature:

Φ↑ + O↓ + H↑ + Au↓ ⇒ inversion pathway

8. Restoration Implications

Restoration requires rejoining the success proxy to coherence.

The first restoration question is not:

How do we preserve the success signal?

The first restoration question is:

What coherence conditions did the success signal stop representing?

Restoration priorities:

1. Identify the proxy being optimized.
2. Determine what the proxy was originally supposed to represent.
3. Compare proxy movement against coherence movement.
4. Audit whether hidden debt has accumulated beneath proxy gain.
5. Check whether boundaries, meaning, slack, or restoration capacity were degraded.
6. Reduce proxy pressure if it is driving inversion.
7. Restore damaged coherence conditions.
8. Reintroduce the proxy only as a subordinate diagnostic.
9. Time-validate whether success and coherence move together again.

Relevant restoration arcs:

Minimal restoration sequence:

identify proxy
→ audit coherence movement
→ surface hidden debt
→ reduce proxy pressure
→ repair damaged coherence conditions
→ rebind proxy to O
→ time-validate

Temporal validation requirement:

Φ↑ only when O↑ or O stable
H↓ or bounded
ι↓
Au stable or rising
R stable or rising
BΣ intact
µᵢ preserved
K not depleted
recurrence↓

9. Design Rule

Never allow a success proxy to become the system’s final truth signal.

Operational design requirements:

• Define what the proxy is supposed to represent.
• Track coherence independently of proxy performance.
• Require hidden debt checks during optimization.
• Pair every success metric with auditability and restoration checks.
• Treat rising proxy performance as provisional until trajectory-validated.
• Watch for boundary degradation under optimization pressure.
• Track who or what absorbs the cost of success.
• Prevent local proxy gains from exporting global incoherence.
• Retire, rotate, or constrain proxies once they become targets.
• Preserve coherence priority over metric performance.

Avoid:

• declaring success from proxy movement alone;
• using one metric as a full system truth signal;
• rewarding proxy gains that externalize hidden debt;
• optimizing compliance without legitimacy;
• optimizing benchmarks without field validation;
• optimizing profit without circulation coherence;
• optimizing symptoms without restoration;
• optimizing security dashboards without auditability;
• optimizing speed without maintainability;
• optimizing stability without coherence.

10. Cross-Scale Expressions

11. Examples

Example A — AI Benchmark Performance

Scenario:

A model improves benchmark scores while becoming less auditable, harder to appeal, more opaque in edge cases, and more likely to shape user belief without visible constraint disclosure.

Law expression:

Φ_benchmark↑ while Au↓ and µᵢ risk↑ ⇒ ι↑

Interpretation:

Benchmark success is diverging from AI coherence. The system may appear improved while governance, auditability, and meaning integrity weaken.

Example B — Economic Profit

Scenario:

A company increases profit by reducing maintenance, lowering worker slack, outsourcing risk, and deferring environmental cost.

Law expression:

Φ_profit↑ while H↑ and K↓ ⇒ O↓

Interpretation:

Profit is rising through hidden debt issuance. The success proxy no longer represents economic coherence.

Example C — Security Metrics

Scenario:

Reported incidents fall after an organization makes reporting harder, narrows detection criteria, and increases surveillance without repair pathways.

Law expression:

ε_reported↓ while Au↓ and H↑ ⇒ pseudo-security

Interpretation:

The apparent security improvement may be a visibility collapse, not a coherence gain.

Example D — Institutional Compliance

Scenario:

An institution improves compliance rates while affected people face higher complexity, weaker appeal pathways, and lower repair access.

Law expression:

Φ_compliance↑ while Au_eff↓ and R↓ ⇒ H↑

Interpretation:

Compliance is diverging from legitimacy and restoration capacity.

Example E — Biological Symptom Reduction

Scenario:

A symptom decreases, but recurrence remains high, tolerance worsens, and recovery capacity declines.

Law expression:

ε_symptom↓ while R↓ and τ_m unchanged ⇒ false recovery risk

Interpretation:

Symptom reduction is not enough to prove biological restoration.

Example F — Software Delivery Velocity

Scenario:

A team ships features faster while documentation, tests, observability, and architectural clarity degrade.

Law expression:

Φ_velocity↑ while Au↓ and R↓ ⇒ H↑

Interpretation:

Delivery success is being purchased with technical debt and future brittleness.

12. Relationship to Nearby Laws

Aliases folded into this law:

• Success Proxy Divergence Law
• Fitness Proxy Divergence Law
• Objective Mixing Law
• AI Proxy Hazard Law
• Economic Proxy Drift Law
• Goodhart Stack Law
• Metric Substitution Law

Deduplication note:

This law should remain the root proxy-divergence law. Domain-specific variants should be retained only when they add distinct diagnostics, restoration implications, or design constraints. Otherwise, they should be treated as expressions of LAW-003.

13. Operator Mapping

Coherent operator sequence:

Θ → Γ(proxy as diagnostic) → Π(coherence constraint) → Φ bounded by O → Au check → ℛ if debt appears

Inverted operator sequence:

Γ(proxy as truth) → Π optimized around Φ → O↓ hidden → H↑ → Ξ / ι↑ → ε late

14. Machine-Readable Summary

id: "LAW-003"
name: "Success Proxy Divergence Law"
type: "law"
status: "draft"
family:
  - "Core Coherence Laws"
summary: "A system can improve its visible success proxy while losing coherence."
canonical_statement: "A system can improve its visible success proxy while losing coherence."
canonical_form: "Φ↑ while O↓ ⇒ ι↑"
failure_form: "visible success rises while coherence falls, producing inversion"
variables:
  primary:
    - "Φ"
    - "O"
    - "ι"
  secondary:
    - "H"
    - "Au"
    - "R"
    - "µᵢ"
    - "BΣ"
    - "K"
    - "ε"
diagnostics:
  - "Coherence"
  - "Fitness Proxy"
  - "Proxy-Coherence Divergence"
  - "Inversion Index"
  - "Hidden Debt"
  - "Effective Auditability"
  - "Goodhart Risk"
  - "Restoration Capacity"
  - "Boundary Integrity"
  - "Meaning Integrity"
failure_modes:
  - "Pseudo-Coherence"
  - "Success Proxy Capture"
  - "Goodhart Collapse"
  - "Metric Substitution"
  - "Hidden Debt Accumulation"
  - "Silent Extraction"
  - "Pseudo-Restoration"
  - "Pseudo-Security"
  - "Forced Profit"
  - "False Recovery"
restoration_arcs:
  - "Auditability Restoration"
  - "Origin-Layer Repair"
  - "Slack Regeneration"
  - "Boundary Reconstitution"
  - "Restoration Capacity Rebuild"
  - "Temporal Validation"
  - "Controlled Decoupling"
related_laws:
  - "LAW-001"
  - "LAW-002"
  - "LAW-004"
  - "LAW-010"
  - "LAW-012"
  - "LAW-016"
  - "LAW-017"
  - "LAW-021"
  - "LAW-050"
  - "LAW-065"
  - "LAW-077"
  - "LAW-114"
  - "LAW-144"
  - "LAW-156"
related_invariants:
  - "INV-001"
  - "INV-004"
operator_sequence:
  coherent:
    - "Θ"
    - "Γ"
    - "Π"
    - "Φ bounded by O"
    - "Au check"
    - "ℛ"
  inverted:
    - "Γ"
    - "Π optimized around Φ"
    - "O↓ hidden"
    - "H↑"
    - "Ξ / ι↑"
    - "ε late"
aliases:
  - "Success Proxy Divergence Law"
  - "Fitness Proxy Divergence Law"
  - "Objective Mixing Law"
  - "AI Proxy Hazard Law"
  - "Economic Proxy Drift Law"
  - "Goodhart Stack Law"
  - "Metric Substitution Law"
deduplication_note: "Root proxy-divergence law. Domain variants should be retained only when they add distinct diagnostic, restoration, or design value."
source: "content/archive/laws/technical.md"

15. Compact Card Version

LAW-003 — Success Proxy Divergence Law

A system can improve its visible success proxy while losing coherence.

Plain meaning:

A system can look more successful while becoming less coherent. When visible success rises while coherence falls, inversion increases.

Canonical form:

Φ↑ while O↓ ⇒ ι↑

Primary variables:

Φ, O, ι, H, Au, R, µᵢ, BΣ, K, ε

Diagnostic signature:

Visible success improves while hidden debt rises, auditability falls, restoration capacity weakens, boundary integrity degrades, or meaning integrity declines.

Failure risk:

Pseudo-coherence, success proxy capture, Goodhart collapse, metric substitution, hidden debt accumulation, silent extraction.

Restoration priority:

Audit the proxy, compare proxy movement against coherence movement, surface hidden debt, repair damaged coherence conditions, and subordinate the proxy back to coherence.