Provenance & verification. This note records a read-only adversarial frontier scout (iteration-12 companion) that scoped iteration-11 A1 open subquestion #3 - whether a non-self-adjoint / PT-symmetric / indefinite-metric (Krein-space) modular structure could be a carrier outside the bicentralizer framing. Its outcome is verdict-neutral (another encode-not-generate confirmation; the 11th consecutive). The two load-bearing NEW citations were independently web-verified this session: arXiv:math-ph/0408048 = Gottschalk, "Complex velocity transformations and the Bisognano-Wichmann theorem for quantum fields acting on Krein spaces" (this CORRECTS the iter-11 A1 note, which had misattributed this arXiv ID to Morchio-Strocchi - Morchio-Strocchi indefinite-metric QFT is a separate, real framework); and arXiv:2307.16805 = Sablevice-Millington, "Poincare symmetries and representations in pseudo-Hermitian quantum field theory." The one paywalled source (the Pi_1 J-symmetric Tomita theorem, Jakobczyk line) is marked unverified - paywalled; scout-confirmed across snippets; its rank-1 restriction is load-bearing only as the reason the single genuine Krein modular theorem fails the signature bar, and the structural conclusion does not depend on it alone.

---

Scoping memo — iter-11 A1 subquestion #3: does a non-self-adjoint / PT-symmetric / Krein-space relative modular structure furnish a CARRIER outside the bicentralizer framing (a "reverse Weinberg–Witten" witness)?

Scout: read-only adversarial frontier scout, for a possible iteration 13. Date: 2026-06-19. Status of this memo: scoping only; no wiki file modified. Carrier definition (inherited, binding): $J=J^*$, $J^2=1$, signature $(\infty,\infty)$ (genuinely indefinite), LOCALIZED (affiliated to $\mathcal M(O)\subsetneq\mathcal M(W)$), ALGEBRA-COMPATIBLE ($\mathrm{Ad}\,J\in\mathrm{Aut}(\mathcal M)$), MODULAR-NATURAL ($[J,\Delta^{it}]=0$). Generative bar: must CARRY NEW geometric content ($\Delta_{\rm geo}\neq0$), not encode it. Standing obstruction it must defeat (iter-7): the canonical Krein fundamental symmetry $\eta_0=\mathrm{sgn}(\ln\Delta)$ already exists from pure $(\mathcal M,\Omega)$ data, is genuinely $(\infty,\infty)$, but is provably GEOMETRY-VOID (unitarily universal across the Borchers/HSMI class $\Rightarrow\Delta_{\rm geo}=0$), generically algebra-incompatible, and non-localizable. Any new candidate must be (i) genuinely NOT a function of $\Delta$ and NOT vacuum/symmetry-inherited, (ii) localized and algebra-compatible, (iii) plausibly generative.

---

1. One-paragraph verdict

Subquestion #3 is a CONFIRMED dead end: every verified indefinite-metric / PT-symmetric / Krein-space modular structure in the literature collapses to encode-not-generate, by one of exactly three structural mechanisms, and no fourth mechanism is visible. [INFERENCE, high] The three collapse channels, each separately fatal to the carrier bars, are: (A) GAUGE ARTIFACT — in indefinite-metric AQFT (Gupta–Bleuler, BRST, Morchio–Strocchi), the indefinite metric is forced by Strocchi's theorem (a local covariant gauge field cannot act on a positive-definite Hilbert space) precisely to preserve locality + covariance, but the negative-norm content is gauge: local gauge symmetries reduce to the identity on the observable algebra and on physical states, and the physical Hilbert space is the positive-definite quotient $\mathcal H_{\rm phys}=\ker Q_B/\operatorname{im}Q_B$ — so the indefiniteness is removed by passing to the observables and carries no localized geometric content that survives to the observable net ESTABLISHED; (B) INSTALLED, NOT DERIVED — every genuine "Krein-space modular theory" (Gottschalk's BW-on-Krein, Jakóbczyk's $\Pi_1$ J-symmetric Tomita theory, Kawamura's indefinite-involution algebras, Sablevice–Millington pseudo-Hermitian QFT) takes the indefinite involution $J$ / metric operator $\eta$ as a given input on a pre-installed indefinite-metric space and runs Krein $\to$ modular, the opposite of the modular $\to$ Krein direction the carrier needs; none derives $J$ from positive $(\mathcal M,\omega)$ data, and the only one that is a true modular theorem ($\Pi_1$) is restricted to rank-of-indefiniteness one, which fails the $(\infty,\infty)$ signature bar outright ESTABLISHED; (C) $\eta_0$-EQUIVALENT OR COMPLEX-PSEUDO-ENTROPY — where a pseudo-Hermitian/PT structure carries genuine modular/KMS content, either the metric operator $\eta$ is positive-definite (it "plays the role of a density matrix") so the theory is a similarity-transform of a standard positive modular structure with $\Delta$ Hermitian by Takesaki's KMS-uniqueness theorem (no genuine indefiniteness survives), or the pseudo-entropy it carries is complex and was already closed at OP-49 as encodes-computation; and the metric operator is moreover non-unique (Mostafazadeh; Scholtz et al.) — fixable only by hand-choosing an irreducible observable set — exactly the iter-7 non-uniqueness of $\eta_0$, which strengthens the geometry-void verdict. Recommendation: record subquestion #3 as CLOSED-NEGATIVE (the indefinite-metric route is the gauge/installed/$\eta_0$ shadow of the carrier problem, not a way around it), with one narrow, honestly-flagged residual offered as an optional low-priority sub-lever (§4).

---

2. Verified literature map

All sources below were checked live (June 2026) against their actual abstracts/snippets. Nothing is fabricated; one publisher page (the $\Pi_1$ paper) was paywalled (403) but its load-bearing content was independently confirmed across three search snippets and matches the wiki's existing record (Jakóbczyk, OSTI 5135323, 1985).

(a) Morchio–Strocchi / Krein-space modular & BW theory

• Gottschalk, Complex velocity transformations and the Bisognano–Wichmann theorem for quantum fields acting on Krein spaces, JMP 43 (2002) 4753, arXiv:math-ph/0408048. [ESTABLISHED — verified verbatim abstract] Proves a BW theorem for fields ALREADY acting on Krein spaces: assumes the indefinite-metric (Krein) structure as input, proves analytic vectors for velocity-transformation generators, combines with Bros–Epstein–Moschella to get BW. Runs Krein $\to$ modular. Does NOT construct Tomita–Takesaki from scratch on a Krein space, does NOT derive the fundamental symmetry from modular data. (Matches the wiki's iter-7 direction-map exactly.)
• Morchio–Strocchi modified Wightman axioms / Hilbert-space-structure condition. [ESTABLISHED — standard framework] The axiomatic frame for indefinite-metric QFT: positivity is replaced by the Hilbert-space-structure condition; the indefinite metric is the starting datum, adopted to keep local covariant gauge fields (forced by Strocchi's theorem). The framework is about representing a given indefinite structure, not generating one.
• Quasivectors / $\Pi_1$ Tomita–Takesaki (Jakóbczyk line), Rev. Math. Phys. (1997), S0129055X97000270; OSTI 5135323 (1985). [ESTABLISHED — content confirmed across 3 independent snippets; publisher page paywalled] This is the single genuine "indefinite modular theorem" in the literature: for a weakly-closed J-symmetric operator algebra with identity on a $\Pi_1$-space (Pontryagin space, RANK OF INDEFINITENESS = 1) with cyclic-separating vector, there is an antilinear $J$-involution and a precise analogue of Tomita's fundamental theorem (plus a double-commutant theorem). Two killers: (1) the algebra is given as J-symmetric on a pre-installed indefinite-metric space — $J$ is installed input, not derived; (2) it works only at rank 1 (one negative direction) — it does NOT extend to general Krein/$\Pi_\kappa$ or $(\infty,\infty)$ signature, which is exactly the carrier's signature requirement. This is the closest thing to a "Krein modular theory with a fundamental symmetry not equal to $\mathrm{sgn}(\ln\Delta)$," and it fails the signature bar by construction.

(b) Indefinite-metric AQFT broadly: gauge theories, Gupta–Bleuler, BRST

• Strocchi's theorem. [ESTABLISHED] A quantum gauge field in a local covariant gauge cannot act on a positive-definite Hilbert space; the indefinite metric is a necessity for locality + covariance in Gauss-law theories (QED). I.e. the indefiniteness is the price of locality, not a geometric carrier.
• Gupta–Bleuler / BRST physical-subspace construction. [ESTABLISHED] $\mathcal H_{\rm phys}=\ker Q_B/\operatorname{im}Q_B$ is positive-definite; negative-norm and zero-norm states are gauge artifacts that decouple from observables. Local gauge symmetries reduce to the identity on the observable algebra and on physical states. Decisive for the gauge-artifact horn: the indefinite metric does NOT carry localized geometric content into the observable net — it is modded out. (Becchi–Rouet–Stora–Tyutin; Kugo–Ojima; standard.)
• Gottschalk et al., indefinite-metric interacting models, arXiv:math-ph/0501031/0501032/0501034. [ESTABLISHED — exist as constructed indefinite-metric models] These build nontrivial indefinite-metric QFT models and a scattering theory, but again within a posited indefinite-metric framework; the indefiniteness is structural input, and physical content is extracted on the positive side. No modular $\to$ Krein generation.

(c) PT-symmetric / pseudo-Hermitian QFT and modular formulations

• Mostafazadeh, Krein-space formulation of PT-symmetry, CPT-inner products, and pseudo-Hermiticity, quant-ph/0606173 (Czech. J. Phys. 56 (2006) 919). [ESTABLISHED] Establishes the now-canonical fact: PT-symmetric / pseudo-Hermitian QM is a Krein-space theory; the CPT-inner-product construction is the 1950s Krein-space machinery. Crucially, the metric operator $\eta$ is NON-UNIQUE (a most-general perturbative $\eta$ exists; the ambiguity is fixed only by choosing an irreducible set of observables — Scholtz–Geyer–Hahne route), and the physics is recovered by passing to a positive-definite physical inner product $\langle\cdot|\eta\cdot\rangle$ with $\eta>0$. Non-uniqueness + positivity-recovery is exactly channel (C).
• Sablevice–Millington, Poincaré symmetries and representations in pseudo-Hermitian QFT, arXiv:2307.16805 (2023/24). [ESTABLISHED — abstract verified] Builds pseudo-Hermitian scalar/fermion QFT by extending the Poincaré algebra with non-Hermitian generators; there is a metric operator $\hat\eta$ ($\hat A^\dagger=\hat\eta\hat A\hat\eta^{-1}$), but it is global (a single Fock-space operator, NOT localized), installed by construction (not derived), and the paper makes no contact with Tomita–Takesaki. Fails localization + derivation bars.
• Kawamura, Algebra with indefinite involution and its representation in Krein space, arXiv:math/0610059. [ESTABLISHED — abstract verified] Starts from an algebra with a given indefinite involution $\dagger$ and replaces it with a positive involution $*$ to land on a positive-definite representation — i.e. it removes the indefiniteness. The opposite of generating geometry; no net/localization structure.
• Non-Hermitian transition matrices / generalized thermofield-double (arXiv:2406.06961; 2606.14208; JHEP 07 (2025) 276). [ESTABLISHED — abstracts/snippets verified] Where an $\eta$-pseudo-Hermitian object carries genuine equilibrium/modular content, $\eta$ is positive-definite and plays the role of a density matrix; by Takesaki's KMS-uniqueness theorem the modular Hamiltonian is the unique Hermitian generator — so no genuine indefiniteness survives in the modular structure. The genuinely-indefinite object these produce is a complex pseudo-entropy, which is the OP-49 dS-pseudo-entropy object already CLOSED-NEGATIVE (encodes-computation; the real part is the region-independent horizon constant).

(d) 2020–2026 indefinite/signature-from-algebra & Krein-vN

• "Spontaneous Emergence of Lorentzian Signature from Curvature-Minimizing Geometry," arXiv:2510.07891 (Oct 2025); "Emergent Lorentz Signature, Fermions, and the Standard Model," arXiv:1403.0580. [ESTABLISHED — these are SELECTION-type constructions] Signature emerges by minimizing a curvature potential over a signature-valued landscape — i.e. on a pre-installed signature-parametrized template, generating only the choice of signature, with symmetric (translation-invariant) saddles. This is exactly the iter-6 selection-type horn of HYP-CKV-VACUITY (Lee 2020 analogue): installs a signature template, fails the carrier spec.
• General 2024–2026 vN modular literature (e.g. Labuschagne–Majewski arXiv:2503.14107, curved-spacetime vN approach; hyperfinite symmetry-resolution arXiv:2510.02441). [ESTABLISHED — verified standard] Uses entirely standard positive-metric Tomita–Takesaki; no indefinite/Krein modular object, no fundamental symmetry $\neq\mathrm{sgn}(\ln\Delta)$.
• Krein-space operator theory 2023–2025 (J-self-adjoint operators, Krein-operator-convexity, boundary triples, etc.). [ESTABLISHED] Active but abstract operator theory on pre-given Krein spaces; no derivation of indefinite metric from a von Neumann algebra + state, no localization/AQFT-net content.

Negative-search result (corroborating iter-7): across (a)–(d), no publication derives a Krein fundamental symmetry from $(\mathcal M,\omega)$ modular data; all prior art runs Krein $\to$ modular (installs the metric) or removes the indefiniteness on passing to observables. The wiki's iter-7 direction-map is confirmed and extended to 2026.

---

3. The candidate objects, and why each fails the carrier bars / no-gos

The structural reason there is no fourth channel. [INFERENCE, high] An indefinite-metric modular object can acquire its indefiniteness in only two ways: it is built from the positive modular data $(\mathcal M,\Omega,\Delta,J)$ — in which case the iter-7 trichotomy applies and any canonical self-adjoint-involution Gram in $\{\Delta\}''$ is a sign function $\varepsilon(\Delta)$ (universal, geometry-void, $\eta_0$-class) — or it is installed by hand on a pre-given Krein/Pontryagin space — in which case it is either (A) gauge, decoupled from observables by Strocchi + BRST, or (B) the geometry is smuggled in with the metric (Gottschalk, Jakóbczyk, pseudo-Hermitian QFT). The PT/pseudo-Hermitian route adds nothing new because (C) its genuine-modular instances have $\eta>0$ (Takesaki KMS-uniqueness $\Rightarrow$ Hermitian $\Delta$, similarity-transform of a positive theory) and its genuinely-indefinite instances produce complex pseudo-entropy already closed at OP-49. Non-self-adjointness is not an escape: a non-self-adjoint relative modular structure is governed by the Connes cocycle $[D\psi:D\varphi]_t$, whose polar/positive content is the positive relative modular operator $\Delta_{\psi|\varphi}$; the cocycle is a unitary in $\mathcal M$, carrying state-comparison data, not a new indefinite localized involution outside $\{\Delta\}''\cup\mathcal M$. This exhausts the ways indefiniteness can enter, so the three channels are complete — there is no room for an object that is simultaneously (i) not-a-function-of-$\Delta$-and-not-inherited, (ii) localized + algebra-compatible, and (iii) generative.

Consistency with the iter-11 referee correction. The carrier's defining property is exact modular commutation $[J,\Delta^{it}]=0$, which lands $J$ in $\mathcal M\cap\{\Delta^{it}\}'=M_\psi$ (centralizer) or in $\{\Delta^{it}\}'\setminus\mathcal M$ — not in any asymptotic/bicentralizer object. None of the §2 candidates lands a genuinely-indefinite, localized, algebra-compatible $J$ in either home: the installed ones are not modular-natural in $\mathcal M$ (or are global), the derived ones are $\eta_0$-class in $\{\Delta\}''\setminus\mathcal M$ (non-local, algebra-incompatible). The indefinite-metric route therefore does not produce a carrier "outside the bicentralizer framing"; it produces the same $\eta_0$/gauge/installed shadow from a different vocabulary.

---

4. Recommendation

Record subquestion #3 as CLOSED-NEGATIVE. [INFERENCE, high — referee-survivable structural argument, not a plausibility claim] The non-self-adjoint / PT-symmetric / indefinite-metric (Krein) relative modular route does not furnish a carrier outside the bicentralizer framing. Every verified object collapses by one of three complete and mutually-exhaustive mechanisms — (A) gauge artifact removed on passing to observables (Strocchi + Gupta–Bleuler/BRST), (B) indefinite metric installed by hand with the geometry smuggled in, running Krein $\to$ modular (Gottschalk, Jakóbczyk-$\Pi_1$-rank-1, pseudo-Hermitian QFT, Kawamura), or (C) $\eta_0$-equivalent / positivity-recovering / complex-pseudo-entropy (Mostafazadeh non-uniqueness, Takesaki KMS-uniqueness, OP-49). This is the encode-not-generate screen reappearing in a new vocabulary, and it closes the "genuinely novel object the literature does not yet treat" question negatively: the literature does not treat it because, structurally, it is the carrier problem's gauge/installed shadow, not a way around it. This is valuable — it converts an inherited OPEN subquestion into a closed-negative with a referee-survivable reason, and it removes the "indefinite-metric modular structure" line from the list of live reopening levers.

Suggested wiki accounting (for a future iteration 13, if run): tag the subquestion CLOSED-NEGATIVE under HYP-FACTORIZATION-IS-GEOMETRY / HYP-ENCODING-SCREEN; do not mint a new hedge or move any grade — the verdict and all four hedges are unchanged (this would be the 11th consecutive confirmation), and the carrier-convergence count stays FIVE (the indefinite-metric route is the same route-5 carrier problem seen through Krein vocabulary, not a sixth route). Add Strocchi's theorem + the BRST positive-quotient fact as the explicit citation for the gauge-artifact horn, and the $\Pi_1$-rank-1 restriction as the explicit reason the only genuine Krein modular theorem fails the $(\infty,\infty)$ signature bar.

Optional low-priority residual (honestly flagged, NOT a recommendation to launch). [OPEN — narrow] The one object the surveyed literature does not fully settle is a higher-rank ($\Pi_\kappa$, $\kappa\geq2$, or infinite-rank) J-symmetric modular theorem: Jakóbczyk's analogue of Tomita's theorem is proved only at rank 1, and I found no theorem (positive or negative) extending it to $(\infty,\infty)$. A dedicated iteration-13 sub-lever could ask whether such a theory (i) exists at all, and (ii) if it exists, whether its $J$ is forced to be installed (Krein $\to$ modular) or could be canonically built from positive data. But the bar is not met for launching: even a successful higher-rank J-symmetric Tomita theorem would, by channels (A)/(B), still take its $J$ as installed input on a pre-given Krein space (it is a representation theorem, not a generation theorem), so it would land in horn (B) and remain geometry-void unless it also derived $J$ from $(\mathcal M,\omega)$ — which is the original carrier problem, not a new object. The honest expected outcome is another encode-not-generate confirmation. Recommend logging this as a watch-item, not scoping a full iteration on it.

---

Sources (verified live, June 2026)

• Gottschalk, BW for fields on Krein spaces, JMP 43 (2002) 4753, arXiv:math-ph/0408048 — verified verbatim; Krein input, BW output.
• Kawamura, Algebra with indefinite involution and its representation in Krein space, arXiv:math/0610059 — verified; replaces indefinite involution with positive one.
• Sablevice–Millington, Poincaré symmetries in pseudo-Hermitian QFT, arXiv:2307.16805 — verified; global installed $\hat\eta$, no modular contact.
• Mostafazadeh, Krein-space formulation of PT-symmetry, CPT-inner products, pseudo-Hermiticity, arXiv:quant-ph/0606173 — verified; non-unique metric, positivity recovery.
• Mostafazadeh, Pseudo-Hermiticity, PT-symmetry, and the metric operator, Czech. J. Phys. 56 (2006) Springer; non-uniqueness fixed by irreducible observable set (Scholtz et al.).
• Quasivectors & $\Pi_1$ Tomita–Takesaki (Jakóbczyk line), Rev. Math. Phys. (1997) (page 403; content confirmed via snippets) + OSTI 5135323 (1985) — rank-1-only J-symmetric modular theorem, installed $J$.
• Strocchi's theorem / Morchio–Strocchi modified Wightman axioms / Hilbert-space-structure condition — standard; indefinite metric forced by locality+covariance, removed on observables.
• Gupta–Bleuler / BRST: $\mathcal H_{\rm phys}=\ker Q_B/\operatorname{im}Q_B$ positive-definite; Scholarpedia BRST; local gauge symmetries = identity on observables.
• Gottschalk et al. indefinite-metric models, arXiv:math-ph/0501031, /0501032, /0501034.
• Non-Hermitian transition matrix / generalized TFD, arXiv:2406.06961; real-time pseudo-entropy arXiv:2606.14208; $\eta>0$ density-matrix role; Takesaki KMS-uniqueness.
• Selection-type signature emergence, arXiv:2510.07891, arXiv:1403.0580 — installed signature template.
• Standard 2024–2026 vN modular (no indefinite object): Labuschagne–Majewski arXiv:2503.14107; hyperfinite symmetry-resolution arXiv:2510.02441.

---

Wiki accounting

Subquestion #3 is recorded CLOSED-NEGATIVE under HYP-FACTORIZATION-IS-GEOMETRY / HYP-ENCODING-SCREEN: the indefinite-metric / Krein route is the same route-5 carrier problem in a new vocabulary, not a sixth route. No hedge minted, no grade moved (11th consecutive confirmation), carrier-convergence count stays FIVE. The narrow higher-rank ($\Pi_\kappa$, $\kappa\geq2$) J-symmetric Tomita question is logged as a watch-item only, not a launch lever (even if it existed it would be a representation theorem with installed $J$ - horn B - hence another encode-not-generate confirmation).