UNNS Substrate Research Program · April 2026

A Second Coordinate
of Physical Structure

From Admissibility to Dual Observability
We discovered that physical systems are not one-dimensional.
Every system is defined by two independent structural coordinates: what is allowed to exist — and how it is internally organized.
This breaks a long-standing assumption: that structural validity determines structure itself. Knowing that a system can exist tells you nothing about how it is connected inside. These are different properties. They require different instruments to measure.
Dual Observability · Proved Zero Violations · 5,233 Runs New Structural Layer 14 Domains · Atoms to Galaxies Realizability · Formally Derived Forbidden State · Empirically Confirmed
Dual-layer structural regimes FULL GIANT TAIL HARD
Structural realizability regimes across scales. Connectivity is not binary — systems organize into distinct regimes (FULL, GIANT, TAIL, HARD) under constraint.
What breaks

We expected disconnected systems to violate the structural law.
They do not.

The original formulation predicted: if a physical system's internal gap structure fails to connect across measurement scales, it must violate the Universal Structural Law. This seemed reasonable. It was wrong.

Across nuclear, cosmological, and molecular domains we find systems that are structurally disconnected — yet fully admissible.

Nuclear isotopes ⁴⁸Ca, ¹⁵⁰Nd, ¹⁰⁰Mo, ²³⁸U do not form a connected backbone across scales. Their gap structures contain single transitions with magnitudes 10¹⁸ times larger than the bulk — permanently isolated vertices. Zero admissibility violations. The same pattern appears in molecular ladders and cosmic-web orientation fields.

This single result invalidates the previous equivalence between disconnection and violation. Disconnection and admissibility are not the same thing. They measure different structural properties, require different instruments, and are independent of each other. They are the two coordinates of physical structure.

Sharpest single instance

The most extreme system is also fully connected — at the most extreme delay ever measured.

Zeeman-split atomic spectral ladders — measured from hydrogen to gold — reach the highest structural pressure in the entire corpus. They sit 4.15% from hard violation. At exactly the same time, they are fully connected across scales — but at a connectivity threshold orders of magnitude beyond any other domain.

0.9585
Structural pressure ρ̄ · highest in corpus · 4.15% from violation
2–4 × 10⁵
Connectivity threshold κ_conn · most extreme delay of any domain

Maximum structural pressure and maximum connectivity delay coexist in the same physical system. No formula from the admissibility instrument predicts the connectivity threshold. No formula from the connectivity instrument predicts the pressure. These are not redundant measurements of the same property. They are two genuinely independent observables. This is the operational proof that the second coordinate exists.

The deepest shift

Selection is not Organisation

Admissibility (USL)
Selects what can exist. The constraint that must be satisfied for a system to persist as a stable structure.
Realizability (PRP)
Determines how it exists. The internal connectivity structure of systems that already satisfy the constraint.

These are not the same process. Knowing that a system is admissible — knowing its structural pressure ρ̄ — tells you nothing about whether its internal gap architecture connects immediately or only after a delay of five orders of magnitude. A biological ribozyme and a nuclear isotope can have identical admissibility profiles and be separated by a factor of 10⁵ in connectivity threshold. The admissibility manifold ℳ_adm is not one-dimensional. It has internal geometry, visible only through the second instrument.

What this means

Physical systems can be valid but structurally fractured

A system can satisfy the Universal Structural Law globally — admissible in every test — while containing internal gap structure that is permanently disconnected at any measured scale. Global validity and local coherence are not the same thing. The nuclear TAIL isotopes demonstrate this directly.

Stability does not imply coherence

High structural pressure does not predict connectivity regime. A system under extreme structural loading can be fully connected or tail-fragmented. The pressure alone cannot distinguish them. A second measurement is required.

Different domains follow different structural laws — inside the same framework

Biology achieves immediate internal connectivity (κ_conn ≈ 1). Nuclear spectra connect only after extreme delay (κ_conn ~ 10⁵). Both are admissible, both are physical. The connectivity threshold is a domain fingerprint that admissibility cannot detect. The framework spans both — and distinguishes them.

A single number is not enough to describe structure

The two-component structural state (ρ̄, κ_conn) is the minimum description. Neither coordinate determines the other. Neither can be omitted. This is the content of the Dual Observability Theorem.

This work establishes

A new coordinate of physical structure. Not a refinement of existing theory — an extension of the dimensionality of structure itself.

Admissibility selects what exists — Realizability determines how it is organized.

Open in Fullscreen!

The formal framework: two manuscripts

The theoretical foundations are developed in two companion manuscripts establishing the dual-layer theory: admissibility (USL layer) and realizability (PRP layer) as independent structural coordinates. Together they prove the Dual Observability Theorem, derive the four realizability classes formally, and establish the logical hierarchy USL → Admissibility → Realizability → Dynamics.

Manuscript I · Percolative Realizability Principle (Revised)

Realizability Structure and the Revised Necessary Direction

Defines the vulnerability graph and the four-condition percolation definition. Formally derives the FULL/GIANT/TAIL/HARD partition as an exhaustive, mutually exclusive classification (Appendix A). Retracts the original equivalence — admissibility ⟺ percolation — proved false by TAIL counterexamples. Establishes the surviving theorem: HARD-class fragmentation implies the existence of a deformation producing a USL violation.

Manuscript II · Dual Observability

Structural Realizability and Dual Observability in the Admissibility Manifold

Proves the Dual Observability Theorem in two steps: logical non-reducibility of the two projections and empirical independence via three counterexample sets from the cross-instrument corpus. Introduces the five-state joint structural taxonomy, the Forbidden State Theorem (high pressure + HARD is empirically excluded), and the Layered Structure Theorem (Appendix B: formal derivation of realizability classes).

Logical hierarchy of structural description USL constraint Admissibility selection filter Realizability connectivity · NEW Dynamics physical laws

What we actually measured

We measure how a system's internal gap structure connects across scales. As the threshold widens, more gaps become structurally coupled. We track whether a dominant connected backbone forms and persists across the full scale range — and at what threshold it does so. Formally, this is the vulnerability graph G_κ(L): vertices are the gaps of the ladder, edges connect gaps within exchange distance ε = κ·median(Δ) at each scale κ.

Structural phase plane — admissibility pressure (ρ̄) vs. connectivity threshold (κ_conn) 0.3 0.6 0.9 1.0 Structural pressure ρ̄ 10⁻¹ 1 10³ 10⁵ no κ_conn Connectivity threshold κ_conn (log scale) → FORBIDDEN high ρ̄ + HARD never observed Zeeman ρ̄=0.9585 Biology · κ≈1 Atm. Nuclear FULL κ~10⁵ Atomic normal CMB Nuclear TAIL Mol / CW Relaxed-Connected Moderate-Connected Stressed-Extreme Tail/Fragmented

The horizontal and vertical axes are independent. Neither predicts the other.

FULL

Complete Connectivity

All gaps in one backbone at κ_max. Percolating.

Biology · Zeeman · CMB · Atmosphere
GIANT

Dominant Backbone

≥ 90% span, tiny isolated tail. Formally percolating.

Molecular (CO, N₂) · Cosmic web · Geodesy
TAIL

Extreme Outliers — Admissible

Large backbone + one permanently isolated gap (ratio up to 10¹⁸). Non-percolating. Fully admissible.

⁴⁸Ca · ¹⁰⁰Mo · ¹⁵⁰Nd · ²³⁸U · Cosmic web peak
HARD

Severe Fragmentation → Violation

No backbone at any scale. Only class implying a violating deformation exists.

Adversarial constructions only. TiO₂ raw DOS: representation-dependent
Dual Observability Theorem (Theorem 4.1)
Every admissible configuration S ∈ ℳ_adm admits two independent observable projections: the admissibility coordinate (ρ̄(S), A_κ^min) and the realizability coordinate (𝒞(S), κ_conn(S)). Neither coordinate function determines the other. A complete structural characterization requires both.

The structural state space

Five joint structural states of admissibility manifold
Joint structural states in the admissibility manifold. The structural state of a system is defined by both admissibility (ρ̄) and realizability (κ_conn), forming distinct regions of behavior across domains.

Tested across atoms, nuclei, cosmology, and biology

Domainρ̄Realizabilityκ_connWhat it demonstrates
Zeeman (atomic) 0.9585 FULL 2–4×10⁵ Max pressure + max delay — orthogonal axes
Biology (QT45 ribozyme) 0.19–0.82 FULL 0.42–2.00 Wide ρ̄ range — immediate connectivity (10⁵× below nuclear)
Nuclear FULL (10 isotopes) 0.197 FULL 3.8×10⁴–4.2×10⁵ Same ρ̄ as TAIL below — different realizability
Nuclear TAIL (⁴⁸Ca, ¹⁵⁰Nd…) 0.197 TAIL undefined Equal admissibility — non-percolating, still admissible
Molecular (CO, N₂, HCl) 0.103 GIANT undefined Lowest ρ̄ — percolating backbone, outlier tail
CMB (Planck 2018 TT) FULL 230–2,389 Cosmological scale — same structural law applies
Atmosphere (ERA5) 0.09–0.23 FULL 0.42–2.00 Like biology in connectivity — different domain, same result
STRUC-PERC-I runs
81
14 domains
USL violations
0
across all 81 runs
FULL percolation
48
59% of runs
TAIL — admissible, disconnected
9
refutes the old general claim
Biology vs nuclear κ_conn
10⁵×
at equal pressure — dual observability
Forbidden state (high ρ̄ + HARD)
0
empirically excluded across corpus

Open in Fullscreen!

The instrument: STRUC-PERC-I v2.4.0

A browser-based computational chamber — no backend, no installation — that measures internal connectivity structure from any ladder dataset and returns the four-tier realizability verdict, connectivity threshold, giant ratio profile, and outlier gap analysis.

Open in Fullscreen!

Relation to existing frameworks

This work does not replace existing theories. It extends them by introducing a missing structural dimension — a second coordinate that existing frameworks cannot see, because they each operate within a single descriptive axis.

Statistical Physics

Classical percolation theory describes systems in terms of a binary phase transition: connected versus disconnected, subcritical versus supercritical. This framework introduces a fundamentally richer picture. Connectivity is not binary — it is structured across regimes. FULL, GIANT, TAIL, and HARD replace simple thresholds; connectivity becomes scale-dependent and history-dependent; and percolation is no longer just a phase transition but a structural condition interacting with admissibility. Most importantly: non-percolation is not equivalent to failure. TAIL systems are non-percolating and fully admissible. This breaks a deep assumption embedded in classical interpretations.

Network Theory

Standard network theory focuses on local and mesoscopic descriptors — degree distributions, clustering coefficients, shortest paths. This framework operates at a different level: global structural organization across scales. It introduces two new dimensions invisible to standard graph metrics:

Standard network theory

  • Degree distributions
  • Clustering coefficients
  • Shortest paths
  • Local or mesoscopic descriptors
  • Describes how nodes connect

This framework adds

  • Structural pressure (ρ̄) — constraint dimension
  • Scale-continuous connectivity (κ_conn) — organizational dimension
  • Globally valid but locally isolated structures (TAIL)
  • Delayed connectivity regimes and dominant backbone formation
  • Describes how structure organizes under constraint

Cosmology

Cosmology traditionally describes large-scale structure geometrically — voids, filaments, clusters — through spatial distribution. This framework introduces a structural-spectral perspective: cosmic structure can be analyzed through the connectivity of gaps, not just spatial position. The cosmic web (DESI, SDSS, 2MRS orientation ladders) returns GIANT and TAIL connectivity profiles, revealing persistent structural isolation and non-trivial connectivity delays across scales that geometry alone does not capture. This suggests that cosmic organization is not purely spatial — it is also structurally hierarchical in gap space. It opens a new way to interpret void boundaries, extreme underdense regions, and large-scale coherence transitions.

Quantum and Atomic Systems

Atomic and quantum systems are usually described through energy levels, transitions, and symmetries. This framework shifts the focus to the structure of gaps between levels — how they are organized in magnitude space, not what their absolute values are. Instead of asking what are the energies? it asks how are the gaps organized? This reveals structural regimes entirely independent of absolute energy scales. The Zeeman case is the sharpest example: maximum structural pressure and maximum connectivity delay coexist in the same spectral system, a finding invisible in any energy-only description.

Beyond existing frameworks

The shared assumption that breaks

All existing frameworks share an implicit assumption: structure can be described within a single descriptive dimension. Statistical physics uses the percolation threshold. Network theory uses graph topology. Cosmology uses spatial geometry. Quantum physics uses the energy spectrum. This work shows that structure requires at least two independent coordinates — one for existence (admissibility) and one for organization (realizability). No existing framework provides both.

A unifying structural layer

This framework does not compete with existing theories. It provides a structural layer beneath them — extending percolation into multi-regime behavior, extending networks into constraint-aware organization, extending cosmology into structural connectivity analysis, and extending quantum descriptions into gap-based organization. Every domain tested confirms the same dual-coordinate structure. The admissibility manifold has internal geometry, and that geometry is the same regardless of what physical medium generates the ladder.

What each framework contributes — and what this work adds Statistical Physics binary percolation phase transitions Network Theory graph topology local descriptors Cosmology spatial geometry voids / filaments Quantum / Atomic energy levels symmetries Dual-Layer Structural Framework Admissibility (ρ̄) + Realizability (κ_conn) — two independent coordinates beneath all domain theories extends each framework · does not replace any · adds the missing structural dimension
One-line conclusion
Existing theories describe structure within their domains.
This framework describes the structure of structure itself.
Core discovery

Physical structure is not one-dimensional. It is governed by two independent coordinates — admissibility and realizability — and no measurement of one predicts the other.

Admissibility answers whether a structure can exist.
Realizability answers how it exists.

⬇ Reproduce the Experiments

All data and instruments used in the corpus analysis are publicly available. The struc_perc_i_v2_4_0 chamber runs entirely in the browser — upload a CSV, press Run, and the full admissibility analysis runs in seconds. The corpus packs below are the exact input files used to produce the published results.

Resources

UNNS Substrate Research Program · April 2026 · STRUC-I v1.0.4 × STRUC-PERC-I v2.4.0 · 5,233 + 81 runs · 14 domains · Protocol: falsification-first, preregistered criteria