Core unresolved structural lemma of the Unified Rigidity Framework (URF).
This file contains no heuristic stability claims. All statements are governed by the Block-Exact standard.
Let:
-
( H ) be a Hilbert space.
-
( \Delta_H : H \to H ) be a self-adjoint operator with compact resolvent.
-
( \mathrm{Per} \subset H ) be a distinguished finite-dimensional subspace.
-
Consider the restricted operator:
[ A := \Delta_H \mid \ker(\mathrm{Per})^\perp ]
Let:
- ( \lambda_k < \lambda_{k+1} ) be two consecutive isolated eigenvalues of ( A ).
- ( E_k, E_{k+1} ) their eigenspaces.
- ( \Pi := \Pi_{E_k \oplus E_{k+1}} ) the combined spectral projection.
Let ( V ) range over the class of URF-admissible perturbations:
- symmetric,
- ( A )-bounded,
- compatible with the combinatorial / geometric structure defining ( H ).
Define the perturbed operator: [ A(\varepsilon) := A + \varepsilon V ]
and the spectral gap: [ \gamma_k(\varepsilon) := \lambda_{k+1}(\varepsilon) - \lambda_k(\varepsilon) ]
The Spectral Gap Rigidity Wall is the decision problem:
For the operator ( A = \Delta_H \mid \ker(\mathrm{Per})^\perp ) and all URF-admissible perturbations ( V ), determine whether
[ \Pi V \Pi = 0 ] holds identically.
This is the entire wall.
No other formulation is valid inside URF.
By the Block-Exact standard:
If: [ \Pi V \Pi = 0 \quad \text{for all admissible } V ]
then: [ |\gamma_k(\varepsilon) - \gamma_k(0)| = O(\varepsilon^2) ]
for all sufficiently small ( \varepsilon ).
If: [ \exists V \text{ admissible such that } \Pi V \Pi \neq 0 ]
then generically: [ |\gamma_k(\varepsilon) - \gamma_k(0)| = \Theta(|\varepsilon|) ]
and no rigidity theorem holds.
The block-exact criterion has been evaluated in the following regimes:
- Trees
- Bounded treewidth graphs
- Finite covers of trees
- Any structure where admissible ( V ) preserves a full block symmetry
- Deterministic high-cycle-overlap graphs
- Cycle expanders
- Hypergraph gadget limits
- Any URF instance with dense local overlap geometry
This wall is:
- not a question about smallness of ( V ),
- not about positivity of ( V ),
- not about smoothness of ( V ),
- not about analytic perturbation theory.
It is only about:
[ \Pi V \Pi \stackrel{?}{=} 0 ]
The wall can be tested mechanically:
Input:
- matrix representation of ( A )
- matrix representation of ( V )
- eigenvalue index ( k )
Compute:
- eigenvectors for ( \lambda_k, \lambda_{k+1} )
- projection ( \Pi )
- norm ( |\Pi V \Pi| )
Output:
- 0 → rigid
-
0 → non-rigid
All URF rigidity results reduce to this invariant.
There is no higher-level obstruction. There is no additional spectral phenomenon. There is no hidden analytic difficulty.
The Spectral Gap Rigidity Wall is exactly:
Does the admissible perturbation class force a nontrivial action on the two-dimensional spectral block or not?
This file is governed by:
standards/URF-Block-Exact/README.mdURF-AUDIT.md
No statement here is allowed to bypass the block-exact criterion.
All remaining theoretical work in URF concerning spectral rigidity is now:
- representation theory,
- symmetry classification,
- combinatorial structure of admissible ( V ),
- geometry of ( H ) inducing unavoidable block mixing.
This is no longer spectral analysis. This is structural mathematics.