The Pauli exclusion principle is a theorem

Fermion antisymmetry is not an axiom of quantum mechanics. It is a consequence of a geometric fact about symplectic polar spaces: two states that are non-isotropic with respect to each other cannot simultaneously occupy the same isotropic flat. This is the Excluded Volume Principle — and the periodic table, the nuclear shell model, and the stability of matter all follow from it.

The claim

The Pauli Exclusion Principle (PEP) is a theorem, not a postulate.

The standard textbook treatment introduces PEP as an empirical rule: no two fermions can share the same quantum state. It is postulated, not derived. Spin statistics is then separately postulated (spin-½ particles are antisymmetric under exchange). These two postulates are independent in the standard treatment.

The ISA framework derives both from a single geometric fact.

The Excluded Volume Principle (EVP): In the symplectic polar space $W(2n-1, 2)$ — the phase space of an $n$-qubit system over $\mathrm{GF}(2)$ — two states $u, v$ are called isotropic with respect to each other if $\langle u, v \rangle = 0$ under the symplectic form. Two states that are non-isotropic ($\langle u, v \rangle = 1$) cannot simultaneously occupy the same isotropic flat (maximal isotropic subspace). This is not a rule imposed on the geometry — it is a theorem about the structure of $W(2n-1, 2)$.

The Geometric Pauli Exclusion Principle (GPEP): The nodal surfaces of electron orbitals in 3D — the $s$, $p$, $d$, $f$ orbital shapes that Heusler’s haptic models make tangible — are the isotropic submanifolds of the symplectic structure on the hydrogen-atom configuration space. Two electrons cannot simultaneously occupy the same nodal-surface configuration because:

Different $(n, \ell, m)$ states are non-isotropic with respect to each other.
Non-isotropic pairs are excluded from the same isotropic flat by the EVP.
Therefore no two electrons with the same $(n, \ell, m, \mathrm{spin})$ can coexist — which is exactly the PEP.

The GPEP is not a separate principle from the EVP. It is the EVP instantiated in spherical geometry, with the symplectic structure on $S^2$ replacing the abstract $W(2n-1,2)$ structure. One principle, two languages.

Why this matters

It unifies what was disjointed. Quantum chemistry has accumulated a collection of empirical rules — PEP, Hund’s rules, the Aufbau principle, Taube’s electron-transfer rules, the Woodward-Hoffmann orbital symmetry rules — that are taught as separate facts, sometimes with partial justifications but never from a single foundation. The EVP provides that foundation:

Rule	EVP derivation
Pauli exclusion	Non-isotropic pairs excluded from same isotropic flat (EVP directly)
Hund’s rule (max spin)	Maximum-spin state = maximum number of non-isotropic pairs = largest ORBIT; always preferred by EVP
Aufbau principle	Fill isotropic flats from lowest to highest action integral $\varphi_\ell$; ORBIT closure at each shell
Woodward-Hoffmann	Thermal/photochemical allowed = parity of ORBIT count across HOMO/LUMO crossing; EVP on the reaction ORBIT
Taube electron transfer	Inner-sphere (direct ORBIT overlap) vs outer-sphere (TWIST without ORBIT contact)
4s before 3d anomaly	$\varphi_{4,0} < \varphi_{3,2}$: zero-TWIST-cost s-orbital outcompetes TWIST-penalised d-orbital

None of these is ad hoc once the EVP is the foundation. They are all statements about ORBIT closure and isotropic flat occupation in $W(2n-1, 2)$.

It explains the shape of the periodic table. The block structure of the periodic table — s-block (2 elements/period), p-block (6), d-block (10), f-block (14) — is the count of non-isotropic pairs in the relevant symplectic space at each $H^k$ level:

Block	$\ell$	Orbitals	States	$H^k$ tier	Symplectic dimension
s	0	1	2	$H^0$ (ORBIT only)	$W(1,2)$
p	1	3	6	$H^1$ (TWIST)	$W(3,2)$
d	2	5	10	$H^1$ (TWIST, higher)	$W(5,2)$
f	3	7	14	$H^2$ (BIND / Fano)	$W(7,2) \supset \mathrm{PG}(2,2)$

The f-block has exactly 14 states because $W(7,2)$ contains the Fano plane $\mathrm{PG}(2,2)$ as its maximal isotropic substructure, and the 7 f-orbitals $(m = -3, \ldots, +3)$ are the 7 points of the Fano plane as the 7-dimensional irreducible representation of $G_2$. The f-block is the Fano sector of the periodic table.

It unifies atomic and nuclear shell models. Heusler demonstrates that the same nodal-surface counting that gives atomic quantum numbers $(n, \ell, m)$ also gives the nuclear shell quantum numbers. The ISA explanation: both atoms and nuclei are ORBIT-counting problems in the same $W(2n-1,2)$ symplectic space; only the action integrals $\varphi_\ell$ differ (Coulomb potential for atoms, Woods-Saxon for nuclei). The nuclear magic numbers $(2, 8, 20, 28, 50, 82, 126)$ are the ORBIT closure points for nucleons, shifted relative to atomic magic numbers $(2, 10, 18, 36, \ldots)$ by the spin-orbit TWIST term $\ell \cdot s$, which costs a Berry phase that rearranges the $\varphi_\ell$ ordering above the $\ell = 2$ shell.

The evidence

Paper	What it shows
Paper 375	EVP = PEP: formal theorem; Jordan-Wigner as explicit construction of the EVP isomorphism; fermion antisymmetry as consequence of symplectic non-isotropy
Paper 487	Valence as orbit occupancy: Aufbau, Hund, Taube as ISA theorems (72/73 = 98.6% on benchmark)
Paper 488	G-walk Chemistry: d-electron orbit occupancy predicts spin state; 90% accuracy without DFT
Paper 490	Galois Protein Design: metal-site orbit structure determines enzymatic selectivity
Paper 491	Galois Chemistry = Tropical DFT: spin-state crossing = tropical singularity; derivative discontinuity = TWIST opcode
x344	Geometric PEP verified computationally: non-isotropic pairs in $W(5,2)$ correctly reproduce PEP for 3-qubit states

Key insight from Heusler (IOP 2018): The 3D-printed nodal-surface models demonstrate that quantum numbers are purely topological — they are homotopy invariants of the nodal surface, not properties of the wave-function amplitude. This is the GPEP: the topology of the nodal surface (number of open vs closed nodal planes) determines the orbit label, and the EVP determines which combinations are allowed. Heusler’s models are physical GPEP demonstrators, though he does not use that language.

The connection to Paper 551 (adelic atom)

The Sommerfeld action integrals $\varphi_\ell = 2\pi(\ell + \tfrac{3}{4})$ in the Thomas-Fermi atom are the action integrals of the isotropic orbits in the TF phase space. The $\tfrac{3}{4}$ Maslov correction is the Berry phase (TWIST at $H^1$) accumulated when an isotropic orbit completes one cycle in the symplectic structure. The FS8 aperiodicity condition — that $|\varphi_\ell’’| \ge c_0 > 0$ for all $\ell$ — is the statement that no two isotropic orbits are degenerate: the EVP applied to the continuous TF phase space. The adelic extension (Conjecture 1 of Paper 551) is the EVP extended $p$-adically: no two adelic isotropic orbits are degenerate at any prime. The Riemann Hypothesis (conditional on the spectral correspondence of Paper 551, Problem 2) is thus the adelic EVP for the TF atom.

What would falsify it

PEP violated without EVP violated: a fermion system found to violate the Pauli exclusion principle but whose configuration space geometry satisfies the symplectic non-isotropy condition. This would separate PEP from EVP.
A classification of quantum numbers not derivable from nodal topology: an observable quantum system whose states cannot be labelled by isotropic flat dimension in any symplectic space of reasonable dimension — which would mean the GPEP is not universal.
Hund’s rule exceptions not explained by ORBIT cost: there are known exceptions to Hund’s rule (e.g. Cr, Cu anomalous configurations). If these cannot be explained by modified $\varphi_\ell$ ordering (TWIST cost reordering the isotropic flats) but require genuinely new principles, the EVP derivation is incomplete.

Open questions

Bosons and the EVP: bosons violate PEP (they prefer the same state, not exclude it). In the EVP picture, bosons are isotropic pairs — they can share an isotropic flat. Is there a unified EVP treatment of both bosons (isotropic, bunching) and fermions (non-isotropic, exclusion) as the two sectors of the same symplectic geometry?
Fractional statistics (anyons): anyons in 2+1 dimensions have statistics intermediate between bosons and fermions. Is there a corresponding “fractional isotropy” in a symplectic space over $\mathrm{GF}(p)$ for $p > 2$? The Fano plane is $\mathrm{PG}(2,2)$ — over $\mathrm{GF}(2)$; the analogue over $\mathrm{GF}(3)$ or $\mathrm{GF}(p)$ might encode anyonic statistics.
Why $\mathrm{GF}(2)$? The EVP uses the symplectic space over $\mathrm{GF}(2)$, the simplest finite field. Is there a derivation from first principles — from the binary nature of spin, or from the two-element structure of measurement outcomes — that singles out $\mathrm{GF}(2)$?