Lemma 9 — Axiom 6 (Single Hamiltonian Uniqueness) Universal NO
← All English posts · 한국어 · 2026-05-02
The full process lemma, embedded directly. Source: Cycle 1 (initial draft) + Cycle 6 (PNAS 2022 retroactive falsifier test, candidate #11 added).
Status disclaimer
Empirical universal NO. Necessary universal NO is not proven — the project’s S9 (logician) caveat: induction from 165 years of empirical NO to all-future-candidates NO is a leap. ZFC-independence is not ruled out.
Wall mapping: this lemma is the paper-direct quantitative codification of Wall #5 (SELF-ADJOINT-RIGOR).
Statement
Of 11 paper-direct Hilbert-Pólya–style spectral candidates audited, the count satisfying axiom 6 strict is zero.
A single self-adjoint operator $H$ on a fixed Hilbert space, capturing all non-trivial $\zeta$-zeros bijectively, with no fine-tuning parameter — this strict combination is not realized in any paper-direct evidence to date.
Axiom 6 strict — 4-specialist consensus definition
The “strict YES” criterion is the simultaneous agreement of four discipline-specific definitions:
| Discipline | Strict definition | Falsifiability condition |
|---|---|---|
| NCG (S3) | A single self-adjoint $D$ on fixed $H$, all $\zeta$-zeros $\leftrightarrow \mathrm{Sp}(D)$ bijective, no fine-tune | Any $\zeta$-zero missing from $\mathrm{Sp}(D) \implies$ NO |
| Quantum physics (S6) | Single PT-symmetric $H$, unbroken PT phase, biorthogonal complete eigenbasis, eigenvalues bijective $\leftrightarrow \zeta$-zero imaginary parts | Broken PT or fine-tune $\implies$ NO |
| Analytic (S1) | A single mollifier-method transformation (Levinson-style) capturing all zeros | Mollifier family required $\implies$ NO (Pratt-Robles 50% barrier) |
| Logician (S9) | ZFC-provable ”$\exists$ unique $H : \mathrm{Sp}(H) = $ {imag parts of $\zeta$-zeros}” | Existence without uniqueness, or ZFC-independent $\implies$ NO |
Common essence across the four: no fine-tuning + simultaneous capture of all zeros.
Audit table — 11 paper-direct candidates
Each row’s anchor is a direct quotation from the candidate’s source paper.
| # | Candidate | Verdict | Paper-direct anchor |
|---|---|---|---|
| 1 | BBM 2017 (Bender–Brody–Müller, Phys. Rev. Lett.) | NO | Paper itself: “We are not yet able to prove eigenvalues real”. Boundary condition $\psi_z(0) = -\zeta(z, 1) = -\zeta(z)$, so $\psi_z(0) = 0 \iff \zeta(z) = 0$ — spectrum identification is trivially the zero condition (Lemma 1 Step 6, BBM circular). |
| 2 | Sierra §III $xp$ (Berry–Keating-type) | NO | Continuous spectrum on the real line — no point spectrum. Cannot capture discrete $\zeta$-zeros. |
| 3 | Sierra §V $H_I$ | NO | Self-adjoint extension via parameter $\theta \in [0, 2\pi)$ — explicit fine-tune. Sierra 2016 §I: “not able to find a single Hamiltonian encompassing all the zeros at once”. |
| 4 | Sierra 2007 $H_2$ | NO | Deficiency indices $(1,1)$, self-adjoint family parameterized by $1 \times 1$ unitary (Table 2). One operator per choice, not a single canonical operator. |
| 5 | Connes–Consani 2021 $\Theta(\lambda, k)$ | NO | Special $\lambda$ values only; for $\lambda^2 = 10.5, k = 18$, the first 31 zeros match with random-coincidence probability $\sim 10^{-50}$ — but the spectrum agrees only at specific parameter choices, not for all zeros simultaneously. |
| 6 | Connes 1999 §VI/VII | NO | Paper introduction: “unnatural parameter $\delta$” — $\delta$-family of operators, not unique. |
| 7 | Lagarias 2002 §8 (1) hypothetical | NO | Paper §8 hypothetical: $\lambda = s^2 - 1/4$. Substituting $s = 1/2 + i\gamma$ gives $\lambda = -\gamma^2 + i\gamma$, complex — incompatible with self-adjoint operator’s real spectrum. Paper itself frames §8(1) as hypothetical. |
| 8 | Berry–Keating 1999 $H = xp$ | NO | Sierra 2007 §I quote: “no concrete proposal realizing all conditions”. Even Berry–Keating’s own 1999 paper §II frames $xp$ as a heuristic, not a rigorous candidate. |
| 9 | Sierra 2007 §VI $\zeta_H$ Jost | NO | M2 family of $(a, b)$ potentials — many operators, not a single one. |
| 10 | Connes 1999 §III $(\mathcal{H}\chi, D\chi)$ | NO | Paper §III + introduction: ”$\delta > 1$ Sobolev exponent — unnatural”. $\delta$-family. |
| 11 | Connes–Moscovici 2022 ($W_{\mathrm{sa}}$, PNAS) | NO | UV asymptotic only — the paper’s own abstract says “ultraviolet behavior reproduces” (not exact spectrum match). $\lambda \in {1, \sqrt{2}}$ fine-tuning (paper §1). Lemma 2.1: deficiency indices $(4, 4)$ — multiple self-adjoint extensions, no single canonical $H$. |
Result: 11/11 axiom 6 strict NO.
Falsifier search — adjacent fields
Beyond the 11 direct candidates, the lemma searches 5+ adjacent fields for any operator that might satisfy axiom 6 strict:
- Selberg trace formula candidates — Selberg’s $\zeta$ on hyperbolic surfaces $\neq$ Riemann’s $\zeta$ (length spectrum vs prime spectrum). The adelic version is candidate #5 above. Not a falsifier.
- Function field RH (Weil 1948 / Deligne 1974) — function-field side has axiom 6 YES (Frobenius eigenvalues), but the number field side requires Wall #1 cohomological transfer, which is a separate fundamental gap. Not a falsifier on the number field side.
- Berry’s modified $H$, quantum chaos “dressed” Hamiltonians — explicit single-$H$ constructions not found in the literature. Not a falsifier.
- Atiyah 2018 Todd-function approach — see Finding 4 for the explicit gap. Not a falsifier.
- Connes–Moscovici 2022 (PNAS) — the Cycle 6 retroactive test, see candidate #11 above. Not a falsifier (UV asymptotic + fine-tune + multiple extensions).
No falsifier found across 5+ adjacent fields + 11 direct candidates.
What this is not
- Not a proof that no such operator can exist. The lemma is empirical — across 11 candidates audited and 5+ falsifier domains searched, none satisfies axiom 6 strict.
- Not closed under ZFC analysis. RH is $\Pi_1$ (Lagarias 2002). The logical strength of the universal-NO statement is undetermined; ZFC-independence is possible.
- Not RH progress. The lemma is RH’s language change — codification of an obstacle pattern, not a path to resolution. The lemma’s
Caveatsblock explicitly says this.
Falsifier criterion — what would retract this lemma
The lemma is falsifiable. A single paper-direct candidate satisfying all three of the following retracts the lemma:
- Single $H$ on a fixed Hilbert space — paper-direct quote.
- All $\zeta$-zeros $\leftrightarrow \mathrm{Sp}(H)$ bijective — paper-direct verification.
- No fine-tuning parameter — paper-direct quote or explicit parameter-free definition.
If all three are simultaneously paper-direct YES for any single new candidate, the lemma retracts and the cycle that produced it is retroactively corrected.
Why this codification might still be useful
The lemma is a structured checklist future spectral candidates can be tested against systematically. The falsifier criterion is explicit, so applying the lemma to a new paper takes minutes:
- Read the paper’s main spectral candidate definition.
- Check whether the operator is a single operator or a parameterized family.
- Check whether the spectrum is claimed to match all $\zeta$-zeros (not just asymptotically).
- Check whether any parameter was fine-tuned to make the match work.
If any of (2–4) fails, the candidate is axiom 6 strict NO.
Reading order
- For the high-level narrative: see Finding 1: 11/11 axiom-6 ceiling.
- For the falsifier-criterion application that produced candidate #11: see Cycle 6’s reading of Connes–Moscovici 2022 (referenced in Cycles 8–11 update).
- For the parallel codification on Wall #2 (forward heat flow): see Lemma 10.
Refuting / strengthening this lemma
If you have a paper-direct candidate that strict-passes axiom 6 — please email x2ever.han@gmail.com. The lemma is genuinely falsifiable.