From Equal Outcomes to Hidden Structural Routes

🧩 From Equality to Route Closure

Start with something deceptively ordinary. Two products are equal: ab = cd. Grade-school arithmetic says that's the end of the story — the numbers match, done. But equality of the endpoint says nothing about whether the two routes that reached it are secretly the same route in disguise, or genuinely different paths that happen to land on the same value.

Common Refinement asks the sharper question: does there exist a hidden four-factor square — elements e, f, g, h with a = ef, b = gh, c = eg, d = fh — from which both routes can be reconstructed? If so, the equality isn't just numerical coincidence; it's the shadow of one shared internal structure. If not, two routes can reach the same number while remaining, in every structural sense that matters, different objects.

For the positive integers under ordinary multiplication, this equivalence holds unconditionally — it is the classical Euclid's-lemma machinery in a new structural language, and it is the fully route-closed control case against which every other regime is measured. The interesting question — the one the rest of this project answers — is what happens once the ambient structure stops being the positive integers.

Figure 1 — From equality to route closure. Equal endpoints do not guarantee identical internal organization. Common refinement asks whether apparently different factorization routes can be reconstructed from one shared four-factor structure. The central equivalence is primal factor traceability ⇔ global four-factor route closure.

📊 A Classification Across Structural Regimes

Once the question is sharpened, it becomes possible to classify exactly which systems close their routes and which do not — not by trial and error, but by an explicit numerical criterion in each regime.

The rank-one criterion, RCI(H) := m/γ — the ratio of the smallest nonzero element to the generator of the ambient group — was validated computationally rather than left as an abstract claim. Every one of 793 rank-one generator families (generators in the range 1–12, generator-set sizes 1–4) was checked exhaustively.

At higher rank, the picture generalizes: an affine monoid closes its routes exactly when ARD(H) := |Atoms(H)| − rank(H) equals zero — that is, exactly when the number of atoms matches the rank of the ambient group, with no redundant generating directions left over. ARD effectively measures the excess of atoms over group rank; every unit of excess opens a new possibility for dependence, and dependence is what makes route closure fail.

Figure 2 — Route closure across structural regimes. Positive integers provide the fully route-closed control; rank one is classified exactly by RCI(H) = 1; higher-rank affine closure requires ARD(H) = 0; and the audited transfinite realization replaces persistent obstruction with exact repair and well-founded residual descent.

🌍 The Geometry of Failure and Repair

Knowing that a system fails to close its routes is one thing. Knowing how the failure is shaped turns out to depend, sharply, on dimension.

Rank one: a finite core, then permanent repair

Among the 513 non-refinable rank-one families, 414 (80.7%) already show a non-primal element near the bottom of the system. But in every single case, the non-primal locus is confined to a finite obstruction core: a median of 18 non-primal elements (maximum 130 across the whole corpus), after which the system becomes permanently, cofinally primal. The proved Cofinal Primal-Tail Theorem guarantees this repair happens by 4c (four times the conductor) at the latest — and the empirically observed onset ratio across the corpus ranges from 2.1 to 3.5, with a median of 2.67, comfortably inside that bound.

Higher rank: the ray that never has to stop

At rank two or above, geometry hands failure a new tool it didn't have before: a nonzero facet direction along which a defect can be translated indefinitely while preserving its witness. Where rank-one obstruction is finite and eventually heals, higher-rank obstruction can propagate along a persistent affine ray — forever, if nothing intervenes. This is the cleanest place the project connects factorization theory to polyhedral geometry: cones, facets, lattice saturation, and unimodularity stop being abstractions and start deciding whether a defect terminates or runs on without end.

The transfinite mechanism: repair made rigorous

The audited transfinite candidate replaces persistent obstruction with something stronger than "the defect eventually goes away": an explicit repair mechanism whose extracted induction principle was formally checked. Every unresolved case is reduced by exact local resolution, to a strictly smaller residual, under a well-founded ordering, until a certified primal base is reached and the original object is reconstructed. All four ingredients are required — a repair that merely changes a verdict without shrinking the residual, or without terminating, is not a repair in this sense.

The extracted route-closure induction proposition was independently formalized rather than merely asserted: it reuses the external Conway-refinement repository's own primality and four-factor-refinement machinery, deliberately declining to introduce a parallel notion of refinement, and the kernel accepted every one of its 837 build jobs and every one of its 829 cached-replay jobs.

Figure 3 — The geometry of failure changes with dimension. Rank-one obstruction is finite and eventually heals; higher-rank obstruction can propagate indefinitely along a persistent affine ray; in the audited transfinite mechanism, unresolved structure is repeatedly reduced by exact repair and strict well-founded descent until a primal base is reached.

🔄 Route Closure Is a Coordinate, Not a Proxy

It would be convenient if route closure simply tracked other structural properties already in use across the UNNS Substrate — if a system that was admissible, or eventually connected, were automatically route-closed too. A frozen 30-system pilot corpus, built from five classes of six systems each and carrying 18 TRACEABLE and 12 NON_TRACEABLE labels, was built specifically to test that convenience. It failed.

Admissibility is saturated — and blind to route closure

Every one of the 30 systems, measured by the STRUC-I admissibility observable A_κ, returned the identical maximal value A_κ = 1 — including all 12 systems that are provably route-defective by the rank-one and affine theorems. Because the observable carries no variance in this corpus, the precise statement is not that admissibility and route closure are "uncorrelated" — a constant cannot support that claim — but the stronger fact that route-closure status does not separate STRUC-I admissibility here at all. Losing route closure carried no admissibility consequence.

Percolation: an instrument audit that mattered

The percolative connectivity coordinate initially told a more dramatic story — and then that story was corrected. The first instrument pass, STRUC-PERC-I v2.5.0, returned 25 FULL_PERCOLATION, 4 HARD_FRAGMENTATION, and 1 TAIL_FRAGMENTATION verdicts, suggesting a localized eventual-connectivity failure outside rank one. An internal audit found that reading did not survive scrutiny: v2.5.0's coarser interquartile-range scale and plateau-stopping rule had mistaken a finite-scale threshold effect — not a genuine absence of connectivity — in five higher-dimensional systems.

Two deterministic fixes — a nonterminal-plateau correction for four systems and an effective-zero interquartile-range correction for a fifth — brought every one of those five verdicts to FULL_PERCOLATION, with the other 25 unchanged. The audited instrument, STRUC-PERC-I v2.5.1, is the one this project now cites throughout.

Figure 4 — Route closure is a coordinate, not a proxy. The Common Refinement extension separates three distinct, non-equivalent structural properties: algebraic route closure, perturbative admissibility, and eventual percolative connectivity. A system may remain admissible and fully connected while still failing global route closure.

📚 Where This Sits, Mathematically and Physically

None of this replaces established theory — it sits deliberately alongside it, borrowing machinery and, in places, giving that machinery a structural interpretation and a failure geometry it did not previously have.

📡 Route Coherence Joins Structural Motion

The wider UNNS Substrate synthesis, Nature as Structural Motion, describes how admissible structures move — through representation, admissibility, organization, and transformation. Its structural state S = (L, c, χ) made the third coordinate, χ, deliberately branch-specific: different branches instantiate it as a closure label, orientation coordinate, downstream-regime tag, or other route-relevant structure. Common Refinement supplies one of its strongest exact algebraic realizations.

This is not a claim that bifurcation theory, catastrophe theory, or percolation theory are secretly the same thing as route closure. It is a narrower and more useful claim: UNNS structural motion asks how a represented state moves among admissible regions, margins, routes, charts, and grammars; route closure asks, orthogonally, whether the alternative internal histories of that same state remain mutually reconstructible. The two questions do not answer each other, and that is exactly why keeping them separate is valuable.

Figure 5 — Route coherence realizes an existing UNNS coordinate. Nature as Structural Motion classifies how admissible structures change, using a structural state S = (L, c, χ) in which χ was deliberately left branch-specific. Common Refinement does not add χ — it supplies one of its strongest exact realizations, identifying algebraic route coherence with primal factor traceability. Representation, admissibility, motion, and route coherence remain complementary rather than reducible to one another.

✨ What the Project Has Gained

Individually, the pieces above are exact algebraic facts, a validated computation, a Lean formalization, and a corpus audit. Assembled into one structural program, they support a claim none of the pieces could make alone:

Key findings, condensed

Figure 6 — What the Common Refinement project has gained. The project advances from an exact algebraic law to a geometry of structural obstruction, a formally verified repair mechanism, and a multi-coordinate interpretation of the UNNS Substrate. Its next questions include canonical refinement, domain-independent route invariants, external review of the transfinite realization, and possible predictive uses of route coherence.

Before this investigation, UNNS could be summarized as: physical and mathematical structures occupy constrained admissibility spaces and move through them in constrained ways. That remains true — but it is no longer the whole picture. The relations among routes themselves turn out to form another level of structure, distinct from state, representation, and admissibility alike. Route coherence has become, in its own right, a mathematical object.