REVIEW 3 major objections 3 minor 9 references
A Version of Exclusively Perturbative Quantum Field Theory
T0 review · 3 major / 3 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read The paper claims that restricting the generating functional integral to perturbative in/out boundary conditions leaves a theory with no non-perturbative effects, eliminating instantons and solving the strong CP problem without axions.
desk verdict The central claim is assumed rather than derived: the expansion argument after Eq. (5) cannot rule out non-perturbative terms, so the paper does not establish that a purely perturbative QFT exists. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the functional integral (5) with its integration measure restricted by Feynman emission boundary conditions (4): asymptotic incoming fields with positive frequencies and outgoing fields with negative frequencies. This same boundary condition, familiar as the origin of the $+i\varepsilon$ propagator prescription, is imposed on all fields, not just the ones whose propagators define perturbation theory. In Euclidean space these fields vanish at time infinities, making total-derivative terms such as the QCD $\theta$-term vanish. The argument that non-perturbative terms are absent then rests on expanding in the coupling near zero along the special complex directions $-1$, $i$, and $i^{2/x}$ and observing that the coefficients stay finite.
What would settle it
Compute the topological susceptibility of the QCD version defined by (5), for example with a lattice discretization that enforces the emission boundary conditions on all fields; a nonzero result or any dependence on the $\theta$-angle would show that non-perturbative effects survive.
Extended reading notes
Core claim
The central claim is that the generating functional (5), in which all integrated fields obey the perturbative in/out (emission) boundary conditions, contains only perturbative contributions. Since the incoming fields contain only positive frequencies and outgoing fields only negative frequencies, the fields oscillate at time infinities in Minkowski space and decrease in the Euclidean continuation, so total derivatives in the Lagrangian integrate to zero. Expanding the functional integral at the points $g=-0$, $g=i0$, and $g=i^{2/x}0$ gives finite coefficients, which the paper takes as excluding non-perturbative terms of the form $\exp(-1/g)$, $\exp(-1/g^2)$, and $\exp(-1/g^x)$. The paper concludes that instantons are absent from the integration region, the strong CP problem is resolved without axions, and non-perturbative condensates used in QCD sum rules can in principle be generated by summing the full perturbative series.
Load-bearing premise
The load-bearing assumption is that finite coefficients in an expansion at special coupling values rule out hidden exponentially small terms, even though a function like $\exp(-1/g)$ can be nonzero yet have a zero Taylor series.
Editorial extensions
If this is right
- The strong CP problem would be solved without axions, because the $\theta$-term is a total derivative that vanishes for fields that decrease at Euclidean time infinities.
- There would be no instanton contributions and no $\theta$-vacua, so the U(1) problem must be handled by the axial anomaly alone.
- The generating functional of the theory would be uniquely defined by the compact formula (5), with no need to enumerate non-perturbative field configurations.
- The non-perturbative quark and gluon condensates of QCD sum rules would be reinterpreted as effects of summing a complete asymptotic perturbative series with massive propagators.
- All Green functions would be given by ordinary renormalized perturbation theory, leaving precision tests such as the electron anomalous magnetic moment unchanged.
Reading between the lines
- The finiteness of Taylor coefficients does not rule out flat exponential terms: a function like $\exp(-1/g)$ has a zero Taylor expansion at $g=0$ yet is nonzero for real $g$, so the argument is better understood as defining non-perturbative sectors out of the theory than as proving they are absent.
- The same boundary-condition restriction would apply to any gauge theory, eliminating monopoles and sphalerons as well as instantons; that is a much stronger change to the vacuum structure than solving just the strong CP problem.
- A checkable consequence is that a lattice version of the restricted QCD functional integral should show zero topological susceptibility, while lattice QCD with periodic boundary conditions sees a nonzero value, making the two formulations physically distinguishable.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a version of renormalizable quantum field theory in which the generating functional of Green functions is defined by Eq. (5) with Feynman emission boundary conditions imposed on all fields in the functional integral. The author argues that, with this restriction, the generating functional contains only perturbative contributions and is free of non-perturbative effects such as instantons and theta dependence, thereby giving a solution to the strong CP problem without axions. The final section adds an assumption that perturbation theory defines the complete generating functional integral. The manuscript is short and mostly conceptual, with no detailed calculations beyond the boundary-condition argument.
Significance. If the central claim were correct, the paper would provide a radical simplification of quantum field theory: a unique functional integral reproducing only perturbative physics, with no vacuum condensates, instantons, or strong CP problem. It would also imply that QCD sum-rule condensates can be obtained from perturbative series with massive propagators. However, the argument that boundary conditions eliminate non-perturbative effects rests on an invalid inference about formal power series, and the paper explicitly concedes that the conclusion is an assumption rather than a derivation. The paper does not provide a proof that perturbation theory is complete in this sense, and the proposed construction appears to place the absence of non-perturbative configurations into the integration measure by hand. The hypothetical significance is high, but the presented support is not sufficient for publication in a serious journal.
major comments (3)
- [Section 2, after Eq. (5)] The demonstration that only perturbative contributions remain is invalid. The manuscript claims that finite coefficients of expansions at g = -0, g = i0, and g = i2/x0 exclude non-perturbative terms of the form exp(-1/g), exp(-1/g^2), and exp(-1/g^x). This inference confuses formal power series with analytic functions. A function such as f(g) = exp(-1/g^2) has a formal Taylor expansion at g = 0 whose coefficients are all finite (indeed all zero), yet f is a standard non-perturbative term that is perfectly finite for real g and only non-analytic at the origin. Thus the finite-coefficient criterion cannot distinguish a purely perturbative generating functional from one containing flat non-perturbative exponential terms. The specific argument about expansions at complex points g = i0 and g = i2/x0 does not repair this flaw, because non-perturbative terms can be flat along some directions and singular along others; the formal expansion at a single complex point is not a probe of all non-perturbative contributions.
- [Eq. (5) and the surrounding definition] The conclusion that non-perturbative effects are absent is built into the definition of the integration domain rather than derived from the dynamics. Eq. (5) explicitly restricts the functional integral to fields satisfying the perturbative emission boundary conditions (4). Since such boundary conditions exclude configurations like instantons by construction, the statement 'this expression for the generating functional integral contains only perturbative contributions and does not contain non-perturbative ones' is a restatement of the chosen measure, not an independent consequence. The manuscript itself acknowledges this in the abstract and in the conclusions: 'we assume that perturbation theory defines the complete generating functional integral.' The paper therefore does not demonstrate that non-perturbative effects are absent; it postulates a version of the theory in which they are not included.
- [Section 2, final paragraph on condensates] The claim that QCD vacuum condensates such as <qq> and <G^2> 'can be generated within perturbation theory with massive perturbative propagators after the summations of the complete asymptotic perturbative series' is asserted without any calculation or argument. This is not a minor remark: it is the response to the central objection that vacuum condensates are inherently non-perturbative. As stated, the claim is unsupported and is in tension with the standard view that such condensates are non-perturbative order parameters. If the author wishes to maintain this point, a concrete demonstration or at least a reproducible resummation procedure is required.
minor comments (3)
- [Abstract and throughout] There are numerous typographical errors, including 'otai ned' in the abstract, 'interacrions' in Section 2, 'applcations' in the Introduction, and 'sume rule' in the same paragraph. These should be corrected in any revision.
- [Section 2, Eqs. (2)-(3)] The notation for the boundary conditions is confusing: Eq. (2) uses t -> infinity on both lines yet distinguishes incoming and outgoing fields, and the symbolic form in Eq. (4) writes 'Φ out/in' without clearly defining the ordering. Clarifying the time-direction convention would improve readability.
- [Section 2, after Eq. (5)] The phrase 'one can straightforwardly expand the generating functional (5) in the perturbative series in the coupling constant g at the point g = -0' is imprecise, since negative zero is not a distinct real point from positive zero; what is presumably meant is expansion around the origin from the negative real side. Such an expansion still cannot be used to rule out terms that are flat at the origin.
Circularity Check
The central claim is definitional: restricting the functional-integral measure to perturbative in/out fields automatically produces a functional integral with only perturbative contributions, and the paper's expansion argument assumes the flat-exponential terms it purports to exclude.
-
self definitional
[Abstract and Section 2, Eq. (5) and following paragraph]
"We propose that these conditions should be used for all fields integrated in the generating functional integral. It is shown that in this case non-perturbative effects are absent. That is we assume that perturbation theory defines the complete generating functional integral."
The integration measure in Eq. (5) is defined by the restriction Φ(t→±∞)→Φ_out^in, i.e. only fields satisfying the perturbative Feynman emission boundary conditions are integrated. The conclusion that the resulting generating functional 'contains only perturbative contributions and does not contain non-perturbative ones' is therefore not derived from the Lagrangian dynamics; it is a restatement of the chosen integration domain. The abstract's own 'That is we assume...' makes this explicit.
-
other
[Section 2, paragraph beginning 'One can also consider an expansion...']
"One can also consider an expansion of the functional integral (5) at the point g = i0. Again the coefficients of the expansion will be finite on the same reasons as above. This excludes the presence of the non-perturbative terms of the type exp(−1/g2) since infinities would be generated otherwise."
This 'demonstration' relies on the premise that if a function has finite Taylor coefficients at selected points, then terms like exp(−1/g^2) are absent. That premise is false (exp(−1/g^2) is flat at g=0 with all Taylor coefficients zero), so the only way the argument works is if one has already assumed the functional integral is purely perturbative. The finite-coefficient test therefore presupposes the conclusion it is meant to establish.
full rationale
The paper's central assertion—that the generating functional in Eq. (5) contains only perturbative contributions—is built into the definition of the integration measure. The measure admits only fields with the perturbative emission boundary conditions, so the absence of non-perturbative field configurations is an input, not a result. The paper concedes this in the abstract and conclusions: 'we assume that perturbation theory defines the complete generating functional integral.' The attempted proof via expansions at g=−0, i0, and i2/x0 does not provide independent support, because the finite-coefficient criterion cannot exclude flat non-analytic terms such as exp(−1/g^2); applying that criterion to an instanton factor would falsely conclude it is absent. Thus the load-bearing chain is circular at the definitional level. The self-citation to the author's earlier strong-CP solution [3] is not the main source of circularity and is not scored separately. Since the conclusion is effectively forced by the chosen domain rather than established by calculation, a high circularity score is warranted, though the paper is transparent about the assumption.
Assumptions & free parameters
assumptions (5)
- domain assumption All fields in the functional integral must satisfy the perturbative in/out boundary conditions of emission (Eq. 4).
- ad hoc to paper Finite coefficients in expansions at g = -0, g = i0, and g = i2/x0 imply the absence of non-perturbative terms of type exp(-1/g^x).
- domain assumption Wick rotation of fields with emission boundary conditions makes them decrease at time infinity, so total derivatives in the Euclidean Lagrangian integrate to zero.
- ad hoc to paper QCD vacuum condensates such as <qq> and <G^2> can be generated perturbatively using massive propagators after resummation of the complete asymptotic series.
- domain assumption The strong CP solution in the author's previous paper [3] is valid.
Cite this review
Pith. "Pith review of A Version of Exclusively Perturbative Quantum Field Theory." pith.science (2026). https://pith.science/paper/2ZZQXTW3
@misc{pith2026241211246,
author = {Pith},
title = {Pith review of: A Version of Exclusively Perturbative Quantum Field Theory},
year = {2026},
howpublished = {\url{https://pith.science/paper/2ZZQXTW3}},
note = {Machine review of arXiv:2412.11246}
}
read the original abstract
We suggest a version of renormalizable Quantum Field Theory which does not contain non-perturbative effects. This is otained by the proper use of the boundary conditions in the functional integral of the generating functional of Green functions. It is well known which boundary conditions are applied to the fields of the functional integral to get correct perturbation theory. We propose that these conditions should be used for all fields integrated in the generating functional integral. It is shown that in this case non-perturbative effects are absent. That is we assume that perturbation theory defines the complete generating functional integral. It allows, in particular, to formulate the generating functional integral in a unique way as an exact compact mathematical formula.
Reference graph
Works this paper leans on
-
[1]
Particle Data Group, S. Navas et al., Phys. Rev. D 110 (2024) 3, 0 30001
work page 2024
- [2]
-
[3]
A New Solution to the Strong CP Problem
S.A. Larin, Symmetry 15 (2023) 12, 2194. e-Print: 2312.11300 [h ep-ph]
work page Pith review arXiv 2023
-
[4]
A.A. Belavin, A.M. Polyakov, A.S. Schwartz and Yu.S. Tyupkin. Phys . Lett. B 59 (1975) 85
work page 1975
- [5]
- [6]
-
[7]
J.B. Kogut and L. Susskid. Phys. Rev. D 10 (1974) 3468; ibid. D 11 (1975) 3594
work page 1974
-
[8]
L.D. Faddeev and A.A. Slavnov, Gauge fields. Introduction to qua ntum theory, Front. Phys. 83 (1990) 1
work page 1990
Show all 9 references
-
[9]
G.Gabadadze and M. Shifman. QCD Vacuum and Axions: What’s Happening?. Int. J. Mod. Phys. A 17 (2002) 3689-3728. e-Print: hep-ph/0206123 [hep-ph]. 7
2002 arXiv
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.