Thermodynamic Diagnostic

The Physics of Return

How does a system return from collapse? Two quantities answer this: τ_R measures the actual return time observed in data, and τ_R* diagnoses how favorable return conditions are at any point in phase space. Together with the drift cost closure Γ(ω) and the seam budget, they form the thermodynamic backbone of the GCD framework.

Moratio reditus — the delay of return. Continuitas non narratur: mensuratur.

§1 — τ_R: The Return Time

A Tier-1 structural object. Not a trend statistic — a hitting time. Return is measured, never assumed.

Definition (Kernel Specification, Def 10)

τ_R(t) := min{ t − u : u ∈ Uθ(t) }

Where Uθ(t) is the return neighborhood — the set of eligible prior indices where the trace state is coherent with the present state within declared tolerances. In plain language: how long has it been since the system last visited a state it recognizes?

What τ_R Measures

  • The minimum delay until re-entry into recognised territory
  • A discrete hitting-time — an event-time, not an average
  • Inherently discontinuous under perturbations (one-sample shift can change the minimum)
  • This is a feature: return is an event, and events are sharp

What τ_R Is Not

  • Not a trend, moving average, or smoothed estimate
  • Not a prediction — it is a measurement of what happened
  • Not always finite — ∞rec is a legitimate outcome
  • Not a diagnostic — it is a Tier-1 structural quantity that feeds the seam budget

Typed Outcomes (Three-valued — never boolean)

τ_R does not return a single number. It returns a typed value with three possible forms. Each form is a first-class semantic primitive, not an error state.

τ_R = finite
Return occurred within the declared horizon. The system found a prior state it recognizes. Budget credit: R · τ_R.
τ_R = ∞rec (INF_REC)
No return under the declared horizon. A typed refusal, not a rounding error. Budget credit: 0 (no return → no credit). The seam cannot close.
τ_R = UNIDENTIFIABLE
The event is not computable under the sampling/missingness regime. Verdict: NON_EVALUABLE. Neither pass nor fail — the third state.

The anti-cheat condition: If τ_R = ∞rec, budget credit is exactly zero, not "undefined." This prevents synthesizing continuity from structure alone. Return must be measured. Si τ_R = ∞_rec, nulla fides datur. Recusatio est exitus primi ordinis, non error rotundationis.

τ_R in the Seam Budget

τ_R appears in exactly one place in the mathematics: as the credit term of the seam budget. This is the only legitimate way τ_R converts return time into coherence credit.

Δκbudget = R · τ_R(t₁) − (Dω(t₁) + DC(t₁))
R · τ_R
Return credit
Time spent recovering, scaled by return rate
Dω = Γ(ω)
Drift debit
Coherence lost to drift
DC = α·C
Curvature debit
Cost of channel heterogeneity

§2 — τ_R*: The Thermodynamic Diagnostic

A Tier-2 diagnostic built from Tier-1 invariants. Not a measurement — a landscape map. It tells you how hard return is from any point.

Definition

τ_R* = (1 − ω) / (Γ(ω) + α·C + ε)

Higher τ_R* means easier return. At τ_R* → 0, the system is trapped: the cost of return grows without bound relative to available coherence.

τ_R* (Diagnostic)

  • A thermodynamic potential — a map of the cost landscape
  • Computed from kernel invariants (ω, C) and frozen constants (α, ε)
  • Classifies phase behavior without modifying anything
  • Tier-2: built on Tier-1, but does not modify it

τ_R (Measurement)

  • A measured hitting time — what actually happened
  • Computed from data: find the nearest prior state in the return domain
  • Feeds into the seam budget as the credit term
  • Tier-1: a structural quantity of the kernel

The Relationship

Think of τ_R* as a topographic map and τ_R as a GPS position. The map (τ_R*) tells you the terrain — where return is easy (high τ_R*) and where it is hard (low τ_R*). The GPS (τ_R) tells you what actually happened: did the system return, and how long did it take? The map doesn't move mountains; the GPS doesn't draw maps. They are complementary: the diagnostic informs, the measurement decides.

Component Anatomy

Each piece of τ_R* has a precise physical role. Click each term to explore its contribution.

§3 — The Drift Cost Closure Γ(ω)

The price of drift. How much coherence the system loses as it drifts away from fidelity. A frozen closure — not chosen, discovered.

The Formula

Γ(ω) = ωp / (1 − ω + ε) p = 3 (frozen), ε = 10−8 (frozen)

Key Properties

  • At ω = 0 (Stable): Γ(0) = 0 — no drift, no cost
  • At ω → 1 (near total collapse): Γ → ∞ — simple pole, regularized by ε
  • At ωtrap ≈ 0.682: Γ = α = 1.0 — the trapping threshold
  • Cubic growth: mild near zero, explosive near one

Why p = 3?

  • The unique integer where ωtrap is a Cardano root of x³ + x − 1 = 0
  • Produces the Watch→Collapse crossover at ω ≈ 0.30–0.40
  • Matches experimental trapping across all 20 closure domains
  • Discovered by the seam — not chosen by convention

Frozen Constants Trans suturam congelatum — frozen across the seam

Parameter Value Role
p3Contraction exponent — cubic slowing
ε10−8Guard band — regularizes pole at ω = 1
α1.0Curvature cost coefficient — unit coupling
tolseam0.005Seam residual tolerance — 1/200
λ0.2EWMA decay coefficient

§4 — The Seam Budget

The accounting identity that connects drift, curvature, and return. Every collapse-return cycle must reconcile.

Budget Model (Def 11)

Predicted dynamics (budget):

Δκbudget = R · τ_R − (Dω + DC)

Observed change (ledger):

Δκledger = κ(t₁) − κ(t₀) = ln(IC(t₁) / IC(t₀))

Reconciliation (seam residual):

s = Δκbudget − Δκledger Seam PASS when |s| ≤ tolseam = 0.005 and τ_R is finite

Seam Pass Conditions (Def 13)

Three conditions must hold simultaneously. Failure of any produces a typed result.

1. Finite Return
τ_R(t₁) is finite — not ∞rec. The system actually returned. If no return, there is no credit to reconcile.
2. Residual Closure
|s| ≤ tolseam = 0.005. The predicted budget must match the observed log-integrity change to within 1/200.
3. Identity Check
The ratio form must match the log-difference form. Cross-checks that no numerical artefact entered the budget.
All three pass → WELD | Any fails → FAIL or NOT_WELDABLE

§5 — The Phase Diagram

τ_R* classifies every point in the ω–C plane into a thermodynamic phase. These are not labels — they are structural diagnoses with budget consequences.

Five Thermodynamic Phases

◈
SURPLUS τ_R* < 0
System generates budget credit spontaneously. Return is "free" — the system naturally recovers. Lives in the Stable regime.
◇
DEFICIT τ_R* > 0
Return costs more time than the budget allows at current rates. The system needs external investment (higher R or lower ω) to close seams.
⬡
FREE RETURN τ_R* ≈ 0
Exactly on the break-even surface. Neither surplus nor deficit. The boundary where Δκ* = −Γ(ω) − αC.
◆
TRAPPED τ_R* > 0 and Γ(ω) > α·Cmax
Single-step escape is impossible. The drift cost exceeds the maximum achievable curvature correction. Intervention from outside the current regime is required.
◉
POLE ω > 1 − 10ε
Singular behaviour. Γ(ω) → ∞. Non-physical regime — the ε guard band prevents actual singularity.

Regime-Dependent Dominance (Theorem T1)

In each regime, a different term dominates the budget cost. This explains why the same system behaves qualitatively differently across regimes.

STABLE (ω < 0.038)
Δκ Dominates
Memory term drives τ_R*. Drift cost negligible (Γ ≈ 5.5 × 10−5).
WATCH (0.038 ≤ ω < 0.30)
α·C Dominates
Curvature heterogeneity drives cost. Γ grows but curvature is still the larger debit.
COLLAPSE (ω ≥ 0.30)
Γ(ω) Dominates
Cubic drift cost overwhelms everything. At ω = 0.70, Γ = 1.14 vs DC = 0.50 max.

§6 — Critical Thresholds

Special points where the system's qualitative behaviour changes. Each is discovered by the algebra, not prescribed.

The Trapping Threshold (Theorem T3)

Γ(ωtrap) = α ωtrap ≈ 0.6823 · ctrap ≈ 0.3177

Below ctrap (above ωtrap), the drift cost Γ exceeds the maximum achievable curvature correction. Single-step correction cannot produce surplus. The system is thermodynamically trapped — escape requires multi-step paths or external intervention.

Rcritical (Theorem T4)

Rcrit = (Γ(ω) + αC + Δκ) / tolseam

The minimum return rate for seam viability. Below Rcrit, the seam cannot close regardless of τ_R. This sets an absolute floor on the return rate.

Free-Return Surface (Def T4)

Δκ* = −Γ(ω) − αC

The zero-cost curve where τ_R* = 0. Points above this surface are in surplus (negative τ_R*); points below are in deficit. This is the break-even boundary of the phase diagram.

§7 — Extended Dynamics

Seven theorems (T10–T16) that add statistical mechanics to the budget identity. These reveal why the system is thermodynamically consistent.

§8 — The Arrow of Time

An asymmetric cost structure that emerges from the budget arithmetic — no second law is postulated.

Asymmetric Cost Structure (Theorem T7)

Degradation (ω increases)
  • Releases budget surplus as system loses coherence
  • Structurally free — cost ≈ 0
  • "Exothermic" in the budget sense
  • Spontaneous — no external work required
Improvement (ω decreases)
  • Costs time (R · τ_R) to reverse drift
  • Must accumulate budget to pay for return
  • "Endothermic" — requires work input
  • 200× more expensive than degradation at ω = 0.60

The Second Law Analog

The arrow emerges without postulate — it is pure budget arithmetic. Systems spontaneously degrade (release Δκ) but require work (time) to improve. Γ(ω) is convex: moving toward higher ω always costs less than moving away from it. This asymmetry is built into the cost function by the cubic exponent p = 3 — degradation follows ω3, but improvement must fight against 1/(1−ω). The arrow of time is an accounting identity, not a postulate.

Measurement Cost (Theorem T9)

Every observation incurs a cost. N observations of a stationary system produce N × Γ(ω) overhead, because each measurement is a collapse event that demands a return to close the seam.

No external vantage. There is no measurement point outside the system. The "positional illusion" is the false belief that you can observe without being measured, measure without incurring cost, or validate without being inside the system you validate. Γ(ω) is the irreducible price of being inside.

Optimal policy: observe as rarely as allowed. Each observation makes seam closure harder.

§9 — Interactive Tools

Explore the thermodynamic landscape. The heatmap shows τ_R* across the ω–C plane; the Γ(ω) curve shows drift cost; the point calculator computes exact values at any coordinates.

τ_R* Thermodynamic Diagnostic

τ_R* = (1 − ω) / (Γ(ω) + α·C + ε) measures how favorable return conditions are. Higher τ_R* → easier return from collapse. The heatmap shows τ_R* across the ω–C plane.

Point Calculator

0.150
0.100
1.000
0.000

τ_R* Heatmap (ω × C)

τ_R*
High Mid Low

Drift Cost Γ(ω) = ω³/(1−ω+ε)

§10 — Numerical Examples

Three worked examples showing how τ_R*, Γ(ω), and the budget behave in each regime.

STABLE Low drift, curvature-dominated budget
ω
0.01
C
0.05
Γ(ω)
—
τ_R*
—

Drift cost is negligible. Curvature cost (α·C = 0.05) is the primary budget debit. τ_R* is high — return is easy.

WATCH Medium drift, curvature still dominant
ω
0.20
C
0.10
Γ(ω)
—
τ_R*
—

Drift cost grows to ~0.01 but curvature (0.10) is still larger. τ_R* decreases — return is harder. The system is transitioning.

TRAPPED High drift, Γ dominates, single-step escape impossible
ω
0.70
C
0.50
Γ(ω)
—
τ_R*
—

Drift cost (Γ = 1.14) exceeds curvature budget. System is trapped: Γ(ω) > α, so no single-step correction can produce surplus. τ_R* collapses toward zero.

§11 — Verification

All results on this page are computationally verified. Every theorem, threshold, and identity has been tested to machine precision.

Test Coverage

  • 79 tests for core τ_R* computation (test_145)
  • 144 tests for dashboard diagnostics (test_146)
  • 57 tests for extended dynamics T10–T16 (test_147)
  • All 15 casepacks: CONFORMANT
  • Zero warnings across all diagnostic tests

Numerical Verification

  • Pole residue: 0.499999863 (target 0.5, error 2.5×10−9)
  • Barrier height: 1.00000000 (target α, error 2.2×10−16)
  • Separability: all cross-derivatives identically 0
  • Scaling: β1/p·⟨ω⟩ → 0.4852 at β=1000 (target 0.5)

Source Files

src/umcp/tau_r_star.py — Core diagnostic

src/umcp/tau_r_star_dynamics.py — T10–T16

src/umcp/frozen_contract.py — Frozen constants

src/umcp/seam_optimized.py — Seam budget

tests/test_145_tau_r_star.py — 79 tests

tests/test_147_tau_r_star_dynamics.py — 57 tests

paper/tau_r_star_dynamics.tex — Full paper

KERNEL_SPECIFICATION.md — Defs 10–13