Superselected ghost theory: real spectrum

Bob Holdom

REVIEW 2 major objections 4 minor 18 references

Superselected ghost theory: real spectrum

T0 review · 2 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Promoting exact ghost parity to a superselection charge gives ghost quantum field theories a consistent probability rule without modifying the Born rule.

desk verdict A coherent formal proposal for ghost superselection, but the real-spectrum assumption in 3+1D is load-bearing and unproven; still worth a serious referee. read the letter →

arxiv 2608.06605 v1 pith:XHENFUVX submitted 2026-08-06 hep-th

Bob Holdom This is my paper · ORCID

classification hep-th

keywords ghostfieldsindefiniteinnerproductsuperselectionruleexactparitypseudo-HermitianHamiltonianPTsymmetryquantumquadraticgravityKällén–Lehmannrepresentation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Quantum field theories with ghost fields (wrong-sign kinetic terms) are unitary but carry an indefinite inner product, so the usual Born rule can produce negative probabilities. This paper proposes that in the real-spectrum regime, where the interacting Hamiltonian has only real energies, the conserved $\mathbb{Z}_2$ operator $Q$, called exact ghost parity, can be promoted to a superselection charge. Physical states are then required to have definite ghost parity and observables must commute with $Q$, which makes the native Born rule yield only non-negative probabilities and gives the optical theorem a direct probabilistic reading. The propagator splits into two sectorwise spectral representations with fixed-sign spectral functions and no complex-conjugate poles on the physical sheet. The payoff would be a consistent probabilistic interpretation for renormalizable quantum gravity, provided the real-spectrum regime actually occurs in interacting 3+1D theories.

What carries the argument

The load-bearing object is the exact ghost parity operator $Q$, defined through the completeness relation over exact energy eigenstates, $Q=\sum_X \int d\Pi_X\, \psi_X\psi_X^\dagger\eta\,/\,|\psi_X^\dagger\eta\psi_X|$, with eigenvalue $\sigma_X$ matching the sign of the native norm. It commutes with the full Hamiltonian and with the $S$-matrix, and its conservation is tied to the positive inner product $G=\eta Q$. The proof machinery is the spectral decomposition of the identity in the $\eta$- and $G$-inner products: the same completeness relation holds term by term for either metric, so replacing $\eta$ by $G$ flips only signs in negative-norm sectors. The similarity transformation $g=G^{1/2}$ then maps $[Q,\tilde{H}]=0$ to $[\eta,h]=0$ for Hermitian $h$, making free ghost parity an exact symmetry of each order of the transformed perturbation theory.

What would settle it

Calculate or approximate the spectrum of an interacting 3+1D ghost theory such as quantum quadratic gravity and find a complex-conjugate pair of energy eigenvalues, or a null-norm eigenstate, at any coupling or in the continuum limit; that would remove the real-spectrum regime on which the paper's construction rests.

Watch

Extended reading notes

Core claim

The paper claims that the obstruction to probabilities in ghost theories is not the indefinite metric itself but the failure to respect a hidden $\mathbb{Z}_2$ symmetry. In the real-spectrum regime the full Hamiltonian has an exact ghost parity operator $Q$, built from the spectral decomposition, whose eigenvalue on each energy eigenstate is exactly the sign of that state's native norm and which reduces to free ghost parity at zero coupling. Treating $Q$ as a superselection charge defines a new physical theory: only states of definite ghost parity exist, only $Q$-commuting observables are physical, and transitions between sectors vanish. Under this rule the native Born rule produces non-negative probabilities, the optical theorem becomes probabilistic term by term, and each propagator decomposes into independent $Q$-sector Källén–Lehmann representations whose fixed-sign spectral functions exclude complex-conjugate poles on the physical sheet. A similarity transformation by $g=G^{1/2}$, with $G=\eta Q$ the positive inner product, turns the Hamiltonian into a Hermitian operator commuting with free ghost parity, giving a perturbation theory whose terms preserve the superselection sectors order by order.

Load-bearing premise

The construction assumes that an interacting ghost theory of interest, such as quantum quadratic gravity in 1+3 dimensions, has a diagonalizable Hamiltonian with an entirely real spectrum and no zero-norm eigenstates; if complex-conjugate energy pairs or null states appear, the superselection rule and its probabilistic interpretation do not apply.

Editorial extensions

If this is right

Transition amplitudes between different ghost-parity sectors vanish, so the negative terms that once spoiled the probability interpretation disappear from the optical theorem.
The full propagator splits into a $Q=+1$ and a $Q=-1$ Källén–Lehmann part, each with fixed-sign spectral density, so complex-conjugate poles cannot appear on the physical sheet.
The similarity-transformed perturbation theory is Hermitian and preserves free ghost parity at every order, giving a calculable framework for the superselected theory.
The lightest $Q=-1$ gravitational excitation is kinematically stable and is identified as a candidate dark-matter particle, with its phenomenology left for future work.
Standard Model fields commute with $Q$, so their local observables survive unchanged, while the superselected theory has no conventional classical limit.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the real-spectrum regime holds for quantum quadratic gravity, one testable extension is to compute the sign of the residue of the would-be ghost pole in the full $Q$-sector propagator; a negative residue confined to the $Q=-1$ sector would confirm the two-sector picture nonperturbatively.
The parallel to color confinement suggests that inclusive hard-scattering calculations in ordinary ghost perturbation theory could remain reliable even when exact superselection governs asymptotic states, turning the paper's high-energy discussion into a calculable approximation scheme for ultra-Planckian scattering.
The stable $Q=-1$ excitation could be a dark-matter candidate whose abundance depends on same-sector pair-production processes, a phenomenological route the paper does not develop.
Because $Q$ is nonlocal and built from the exact spectrum, any $Q$-even projection of a local operator is generally nonlocal, which pushes physical observables in a gravitational superselected theory toward asymptotic data and reinforces an $S$-matrix description of quantum quadratic gravity.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The new thing here is the move from treating exact ghost parity Q as a curiosity to imposing it as a superselection charge, plus the sector-preserving similarity-transformed perturbation theory that goes with it. On a first read, the formal structure is internally consistent: Q is defined explicitly from the spectral decomposition, Q^2=1, [Q,H]=0, and Q's eigenvalue equals the norm sign by construction. The optical theorem is then just the standard unitarity identity written with the native inner product, and the superselection rule kills the cross-sector terms that would have produced negative probabilities. The spectral representation decomposition into Q-sectors with fixed-sign spectral functions is also clean, and the claim that complex-conjugate poles cannot appear on the physical sheet follows naturally from the sector separation.

The similarity transformation is the most interesting part. Defining g=G^{1/2} and h=g H g^{-1} gives a Hermitian h with [η,h]=0, and the first-order result h1 = (H1 + η H1 η)/2 is a neat projection onto sector-preserving couplings. That is a genuinely useful perturbative tool, even if only first order is shown here and the higher-order nonlocal vertices are deferred to companion papers.

The soft spots are real but the paper names them honestly. The principal one is the existence of the real-spectrum regime in interacting 1+3D ghost theories, especially quantum quadratic gravity. Section II.B simply assumes diagonalizability, real spectrum, no null eigenstates, and a well-defined continuum limit. If that regime fails, Q's eigenvalue no longer tracks the norm sign and the probabilistic interpretation collapses. The conclusion explicitly calls this the principal open question, so the paper is not overclaiming—but that means the framework is conditional, not demonstrated. A second soft spot is that the superselection rule is a new postulate defining a new theory, not a resolution of the original ghost theory's interpretation. That distinction is stated clearly, but it is a conceptual limitation worth keeping in view. The finite-volume regularization and the treatment of the continuum limit are also somewhat formal.

Who should read this: people working on ghost QFT, quantum quadratic gravity, or the interface with PT-symmetric quantum mechanics. It is a formalism paper, not a phenomenological one, and the dark-matter speculation is appropriately flagged as tentative. The citation pattern looks fine; the relevant pseudo-Hermitian and PT literature is cited, and the accompanying papers are marked as to-appear rather than being hidden.

My recommendation: send it to a serious referee. A well-chosen referee can check the formal consistency, push on the real-spectrum assumption, and judge whether the superselection postulate is physically motivated enough. The paper deserves discussion even if the eventual verdict is that the dynamical question remains open.

Referee Report

2 major / 4 minor

Summary. The paper considers quantum field theories with an indefinite ('native') inner product, so-called ghost theories, and proposes a resolution to their lack of a probabilistic interpretation in the case where the interacting Hamiltonian has a purely real spectrum. Under this real-spectrum assumption, the author constructs an exact Z2 symmetry Q, called 'exact ghost parity', whose eigenvalue on each energy eigenstate equals the sign of the native norm (Eqs. (6) and (10)). Promoting Q to a superselection charge defines a new theory in which physical superpositions and observables respect the Q sectors. The manuscript argues that this makes the native Born rule non-negative, gives the optical theorem a direct probabilistic interpretation, splits the propagator into sectorwise Kallen-Lehmann representations with fixed-sign spectral functions, and permits a similarity-transformed, sector-preserving perturbation theory. The central caveat, acknowledged in the conclusion, is that the existence of the real-spectrum regime in interacting 3+1D theories (notably quantum quadratic gravity) remains an open dynamical question.

Significance. If the real-spectrum regime exists in a physically relevant interacting ghost theory, this construction would provide a concrete probabilistic interpretation for unitary but indefinite-metric theories, with potential implications for renormalizable quantum gravity. The paper's strengths include an explicit operator construction (Eqs. (6) and (10)), a clear derivation of the optical theorem and spectral decompositions under the stated axioms, and a concrete similarity transformation (Section V) that yields order-by-order ghost-parity conservation. The author is also transparent about the central unproven input. However, the framework is conditional: the existence of the real-spectrum regime is the load-bearing assumption, and the paper provides no nontrivial interacting 3+1D realization. The additional superselection postulate is imposed by hand, with no dynamical or observational motivation beyond analogy. As a formal conditional construction the paper is coherent, but its physical applicability rests on assumptions that are not established.

major comments (2)

[Section II.B, Eq. (10); Section VI (Conclusion)] The entire construction presumes the real-spectrum regime: Eq. (10) defines Q only if the interacting Hamiltonian is diagonalizable with a real spectrum, no null eigenstates, and a well-defined continuum limit, as explicitly assumed in Section II.B. The paper supplies no evidence that this regime occurs in any interacting 3+1D ghost theory, and the conclusion states that this is the principal open question. Since the superselection rule, the positivity of Born probabilities, the optical theorem interpretation, and the sectorwise spectral representations all depend on this assumption, the manuscript is a conditional construction rather than a demonstrated theory. To make the contribution more than a formal 'if-then' statement, the author should either exhibit a nontrivial interacting model (even in lower dimensions or in a solvable limit) where the real-spectrum regime is proven or numerically established, or explicitly frame the paper as a formalism awaiting a dynamical input and temper the claims about quantum quadratic gravity and dark-matter candidates.
[Section I and Section VI] Promoting Q to a superselection charge is stated in the conclusion to be 'an additional physical postulate that defines a new theory'. Unlike familiar superselection rules that emerge from gauge symmetries or environmental decoherence, this rule is imposed by hand. The paper does not discuss possible mechanisms that could enforce the superselection rule, nor does it examine whether the acknowledged nonlocality of Q (Section VI) is compatible with the locality of physical observables and microcausality. This is load-bearing because the probabilistic interpretation rests entirely on the postulate; without an argument for why a given ghost theory should select this rule, the applicability to quantum quadratic gravity remains speculative rather than derived.

minor comments (4)

[Section IV (typo)] The word 'complex-congutate' before the references in Section IV should read 'complex-conjugate'.
[General notation] The paper alternates between '3+1D' and '1+3D'; a consistent convention would improve readability.
[References [10] and [11]] Companion papers [10] and [11] are cited with placeholder arXiv numbers and 'to appear'. If they are not yet publicly available, the dependence of Section V on [11] should be clarified or made self-contained.
[Eq. (19) discussion] The sentence defining T# could be more precise: T# = K^{-1} T† K is the K-adjoint, and the statement that it is obtained by replacing H† with H is only obvious if K is diagonal in the energy basis; a brief explanation would avoid confusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the construction is an explicitly postulated superselection rule conditional on the unproven real-spectrum regime.

full rationale

The paper's derivation chain is conditional but not circular. Equation (6) constructs Q from the exact energy eigenstates with Qψ_n = σ_n ψ_n and σ_n = sgn(ψ_n† η ψ_n), and Section VI states that "Treating Q as a superselection charge is an additional physical postulate that defines a new theory." The non-negative Born probabilities, the sign-definite optical theorem right-hand side, and the sectorwise fixed-sign spectral functions are consequences of this postulate together with [Q, H̃] = 0; they are not disguised fits or results imported from the companion self-citations. The paper explicitly flags the load-bearing assumption: in Section II.B it assumes a diagonalizable Hamiltonian with a real spectrum and no null eigenstates, and the conclusion states "the principal open question is dynamical: whether the real-spectrum regime occurs in interacting 1+3D theories of practical interest, especially quantum quadratic gravity." The similarity transformation in Section V is derived from G = ηQ and g = G^{1/2}; its higher-order completion is deferred to companion paper [11], which is a missing-support limitation rather than a circular step. Self-citations [9]–[11] and [17] provide notation, 0+1D motivation, and companion developments but do not carry the central conditional argument, which rests on the stated assumptions and on external pseudo-Hermitian QM results [3,7,8]. Therefore no enumerated circular pattern is present, and the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central results rest on the listed axioms. The key dynamical assumption is the existence of the real-spectrum regime (Axiom 1); the construction of Q (Axiom 2) is a definition chosen to enforce the desired properties. No free parameters are fitted to data.

assumptions (5)

domain assumption The Hamiltonian is diagonalizable with a real spectrum and no null eigenstates, and the continuum limit is well defined.
Section II.B states these as assumptions defining the real-spectrum regime; the conclusion admits that whether this regime occurs in 3+1D is the principal open question.
ad hoc to paper Exact ghost parity Q is defined by the spectral decomposition of the Hamiltonian, with eigenvalue equal to the sign of the native norm on each energy eigenstate.
Eq. (6) constructs Q in this way; the paper notes an infinite family of Z2 symmetries exists and this one is chosen so that eigenvalues match norm signs.
domain assumption The exact vacuum is unique and continuously connected to the free Fock vacuum, so it has Q=+1 and positive norm.
Stated in Section I; needed to fix the vacuum sector and the zero-coupling limit of Q.
domain assumption A conserved positive-definite inner product G=ηQ exists, equivalent to the reality of the spectrum.
Section II.A invokes this to define the similarity transformation g=G^{1/2} in Section V; existence of G is the pseudo-Hermitian condition for a real spectrum.
domain assumption The asymptotic Fock space is built from exact one-particle states, and Q acts multiplicatively on multiparticle states.
Section II.B and Eq. (10); needed to prove the S-matrix commutes with Q and that cross-sector transitions vanish in Eq. (15).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Superselected ghost theory: real spectrum." pith.science (2026). https://pith.science/paper/XHENFUVX

@misc{pith2026260806605,
  author       = {Pith},
  title        = {Pith review of: Superselected ghost theory: real spectrum},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XHENFUVX}},
  note         = {Machine review of arXiv:2608.06605}
}

abstract

Quantum field theories with ghosts can be unitary and perturbatively stable, yet the negative-norm states of their conserved indefinite inner product obstruct a probabilistic interpretation. This problem is especially relevant to renormalizable quantum gravity. The focus of this paper is the real-spectrum regime, in which the interacting Hamiltonian has an exact $\mathbb{Z}_2$ symmetry $Q$, called exact ghost parity. Its eigenvalue on each energy eigenstate equals the sign of the norm, and it reduces to free ghost parity at zero coupling. Imposing $Q$ as a superselection charge defines a new theory in which the physical states have definite ghost parity and the observables commute with $Q$. The native Born rule then yields non-negative probabilities, while the optical theorem acquires a direct probabilistic interpretation. The propagator decomposes into $Q$-sector spectral representations with fixed-sign spectral functions and no complex poles on the physical sheet. Finally, a similarity transformation yields a Hermitian perturbation theory that preserves free ghost parity order by order, making the exact superselection structure perturbatively manifest.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 5 canonical work pages

[1]

Renormalization of Higher-Derivative Quantum Gravity,

K. S. Stelle, “Renormalization of Higher-Derivative Quantum Gravity,” Phys. Rev. D16, 953–969 (1977)

work page 1977
[2]

Quadratic Gravity,

A. Salvio, “Quadratic Gravity,” Front. in Phys.6, 77 (2018) [arXiv:1804.09944 [hep-th]]

arXiv 2018
[3]

Pseudo-Hermiticity versus PT symmetry: The necessary condition for the reality of the spectrum,

A. Mostafazadeh, “Pseudo-Hermiticity versus PT symmetry: The necessary condition for the reality of the spectrum,” J. Math. Phys.43, 205 (2002) [arXiv:math-ph/0107001 [math-ph]]

arXiv 2002
[4]

Complex extension of quantum mechanics,

C. M. Bender, D. C. Brody and H. F. Jones, “Complex extension of quantum mechanics,” Phys. Rev. Lett.89, 270401 (2002) [Erratum: Phys. Rev. Lett.92, 119902 (2004)] [arXiv:quant- ph/0208076 [quant-ph]]

work page arXiv 2002
[5]

Exact PT symmetry is equivalent to Hermiticity,

A. Mostafazadeh, “Exact PT symmetry is equivalent to Hermiticity,” J. Phys. A36, 7081-7092 (2003) [arXiv:quant-ph/0304080 [quant-ph]]

arXiv 2003
[6]

Making sense of non-Hermitian Hamiltonians,

C. M. Bender, “Making sense of non-Hermitian Hamiltonians,” Rept. Prog. Phys.70, 947 (2007) [arXiv:hep-th/0703096 [hep-th]]

arXiv 2007
[7]

Pseudo-Hermiticity versus PT symmetry II: A complete characterization of non-Hermitian Hamiltonians with a real spectrum,

A. Mostafazadeh, “Pseudo-Hermiticity versus PT symmetry II: A complete characterization of non-Hermitian Hamiltonians with a real spectrum,” J. Math. Phys.43, 2814–2816 (2002) [arXiv:math-ph/0110016 [math-ph]]

arXiv 2002
[8]

PseudoHermiticity versus PT symmetry III: Equivalence of pseudoHer- miticity and the presence of antilinear symmetries,

A. Mostafazadeh, “PseudoHermiticity versus PT symmetry III: Equivalence of pseudoHer- miticity and the presence of antilinear symmetries,” J. Math. Phys.43, 3944-3951 (2002) 14 [arXiv:math-ph/0203005 [math-ph]]

arXiv 2002

Show all 18 references

[9]

Making sense of ghosts,

B. Holdom, “Making sense of ghosts,” Nucl. Phys. B1008, 116696 (2024) [arXiv:2408.04089 [hep-th]]

2024 arXiv
[10]

Superselected ghost theory: entangled pairs,

B. Holdom, “Superselected ghost theory: entangled pairs,” to appear, arXiv:2608.XXXXX
[11]

ghosts violate unitarity

develops this construction further and shows how the nonlocality ofQis handled in perturbation theory. II. FROM 0+1D TO 3+1D QFT A. 0+1D: Pseudo-Hermitian Quantum Mechanics We begin with 0+1D QFT, where field-theoretic constructions such as the full Feynman propagator already ...
[12]

Superselected ghost theory: perturbation theory,

B. Holdom, “Superselected ghost theory: perturbation theory,” to appear, arXiv:2608.XXXXX
[13]

The Lee-Wick Standard Model,

B. Grinstein, D. O’Connell and M. B. Wise, “The Lee-Wick Standard Model,” Phys. Rev. D 77, 025012 (2008) [arXiv:0704.1845 [hep-ph]]

2008 arXiv
[14]

Fakeons, microcausality and the classical limit of quantum gravity,

D. Anselmi, “Fakeons, microcausality and the classical limit of quantum gravity,” Class. Quant. Grav.36, 065010 (2019) [arXiv:1809.05037 [hep-th]]

2019 arXiv
[15]

Unitarity, stability and loops of unstable ghosts,

J. F. Donoghue and G. Menezes, “Unitarity, stability and loops of unstable ghosts,” Phys. Rev. D100, 105006 (2019) [arXiv:1908.02416 [hep-th]]

2019 arXiv
[16]

Anti-Instability of Complex Ghost,

J. Kubo and T. Kugo, “Anti-Instability of Complex Ghost,” PTEP2024, no. 5, 053B01 (2024) [arXiv:2402.15956 [hep-th]]

2024 arXiv
[17]

Remarks on ghost resonances,

L. Buoninfante, “Remarks on ghost resonances,” JHEP02, 186 (2025) [arXiv:2501.04097 [hep- th]]

2025 arXiv
[18]

Ultra-Planckian scattering from a QFT for gravity,

B. Holdom, “Ultra-Planckian scattering from a QFT for gravity,” Phys. Rev. D105, no. 4, 046008 (2022) [arXiv:2107.01727 [hep-th]]

2022 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Renormalization of Higher-Derivative Quantum Gravity,

K. S. Stelle, “Renormalization of Higher-Derivative Quantum Gravity,” Phys. Rev. D16, 953–969 (1977)

work page 1977

[2] [2]

Quadratic Gravity,

A. Salvio, “Quadratic Gravity,” Front. in Phys.6, 77 (2018) [arXiv:1804.09944 [hep-th]]

arXiv 2018

[3] [3]

Pseudo-Hermiticity versus PT symmetry: The necessary condition for the reality of the spectrum,

A. Mostafazadeh, “Pseudo-Hermiticity versus PT symmetry: The necessary condition for the reality of the spectrum,” J. Math. Phys.43, 205 (2002) [arXiv:math-ph/0107001 [math-ph]]

arXiv 2002

[4] [4]

Complex extension of quantum mechanics,

C. M. Bender, D. C. Brody and H. F. Jones, “Complex extension of quantum mechanics,” Phys. Rev. Lett.89, 270401 (2002) [Erratum: Phys. Rev. Lett.92, 119902 (2004)] [arXiv:quant- ph/0208076 [quant-ph]]

work page arXiv 2002

[5] [5]

Exact PT symmetry is equivalent to Hermiticity,

A. Mostafazadeh, “Exact PT symmetry is equivalent to Hermiticity,” J. Phys. A36, 7081-7092 (2003) [arXiv:quant-ph/0304080 [quant-ph]]

arXiv 2003

[6] [6]

Making sense of non-Hermitian Hamiltonians,

C. M. Bender, “Making sense of non-Hermitian Hamiltonians,” Rept. Prog. Phys.70, 947 (2007) [arXiv:hep-th/0703096 [hep-th]]

arXiv 2007

[7] [7]

Pseudo-Hermiticity versus PT symmetry II: A complete characterization of non-Hermitian Hamiltonians with a real spectrum,

A. Mostafazadeh, “Pseudo-Hermiticity versus PT symmetry II: A complete characterization of non-Hermitian Hamiltonians with a real spectrum,” J. Math. Phys.43, 2814–2816 (2002) [arXiv:math-ph/0110016 [math-ph]]

arXiv 2002

[8] [8]

PseudoHermiticity versus PT symmetry III: Equivalence of pseudoHer- miticity and the presence of antilinear symmetries,

A. Mostafazadeh, “PseudoHermiticity versus PT symmetry III: Equivalence of pseudoHer- miticity and the presence of antilinear symmetries,” J. Math. Phys.43, 3944-3951 (2002) 14 [arXiv:math-ph/0203005 [math-ph]]

arXiv 2002