Novel Lattice Formulation of 2D Chiral Gauge Theory via Bosonization

Hiroshi Suzuki; Okuto Morikawa; Soma Onoda

REVIEW 3 major objections 4 minor 17 references

Novel Lattice Formulation of 2D Chiral Gauge Theory via Bosonization

T0 review · 3 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read This paper establishes a lattice regularization of 2D U(1) chiral gauge theory via bosonization, using holes excised from the lattice to represent vector-charged objects, and derives selection rules that match the continuum fermion number…

desk verdict A genuine new construction combining excision with bosonized 2D chiral gauge theory, but the central selection rule rests on an unproved equality of Chern numbers. read the letter →

arxiv 2501.18949 v1 pith:RFYB5RP6 submitted 2025-01-31 hep-lat

Okuto Morikawa , Soma Onoda , Hiroshi Suzuki This is my paper · ORCID

classification hep-lat MSC 81T2581T5081T13

keywords chiralgaugetheorylatticebosonizationadmissibilityconditionexcisionmethodcompactbosonanomalyselectionrules

topics Yang-Mills Mass Gap

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

Chiral gauge theories, the backbone of the Standard Model, lack a fully non-perturbative lattice definition because lattice fermions typically double or break exact gauge symmetry. This paper proposes a lattice formulation of 2D U(1) chiral gauge theory based on Abelian bosonization, in which fermions are replaced by compact bosons and the gauge anomaly is computable at the classical level. Its new ingredient is the excision method: vector-charged objects, ordinarily forbidden by the smoothness condition called admissibility, are realized as holes excised from the lattice. The authors define a gauge-invariant vector charge for each hole and, together with axial vertex operators, derive selection rules of the form sum of axial charges plus or minus half the sum of vector charges equals the appropriate charge times the Chern number. These rules match the continuum fermion number anomaly, so the construction exhibits the same anomaly structure at finite lattice spacing.

What carries the argument

The central object is a 'hole' — a region excised from the lattice, situated around a dual-lattice site, whose boundary carries a non-zero winding number of the compact boson field. The hole realizes a vector-charged object in a way compatible with the admissibility condition, which otherwise forces all field-strength and derivative combinations to be so small that such winding is forbidden. The argument is carried by two Chern numbers: ν, defined in Eq. (4.3) from the total field strength on the original lattice, and ν-tilde, defined in Eq. (4.7) from the dual lattice; the selection rules (4.2) and (4.6) express vector and axial charge quantization through these integers. The equality ν = ν-tilde, assumed under 'sufficiently strict admissibility', is what allows the two rules to be combined into the final left/right selection rules (4.9).

What would settle it

A concrete test would be to search for an explicit lattice configuration that satisfies the stated admissibility bounds (with chosen ε, δ, and any value of δ') for which ν ≠ ν-tilde — for instance, a small hole whose boundary flux on the original lattice differs from the dual-lattice flux by one unit. If such a configuration exists, then the assumption ν = ν-tilde cannot hold under the stated conditions, and the selection rules (4.9) would not follow from the paper's derivation. Alternatively, computing the sharpest upper bound on |ν - ν-tilde| in terms of the admissibility parameters and showing it can vanish only in a limit that excludes all non-trivial holes would settle the issue.

Watch

Extended reading notes

Core claim

The paper claims that a lattice regularization of 2D U(1) chiral gauge theory can be constructed from compact boson variables and U(1) link variables, with an action (Eq. (3.14)) that mirrors the continuum bosonized action. Under gauge transformations, the action shifts by the mixed anomaly term, and when the anomaly cancellation condition sum q_V q_A = 0 holds, the action is exactly invariant at finite lattice spacing. Vector-charged objects are defined by excising a region D from the lattice, a 'hole', and the gauge-invariant vector charge is a line integral of the covariant boson derivative around the hole; this charge can take non-zero values precisely because the admissibility condition is relaxed inside the hole. The paper then derives selection rules: the total vector charge equals -2 q_A times the original-lattice Chern number ν, while the total axial charge equals q_V times the dual-lattice Chern number ν-tilde. Assuming ν = ν-tilde under sufficiently strict admissibility, these combine to give the left/right selection rules of Eq. (4.9), identical to the continuum fermion number anomaly. This establishes that the lattice definitions of charged operators satisfy the expected topological selection rules and thus reproduce the anomaly structure of the continuum theory.

Load-bearing premise

The whole derivation rests on the assumption that, if the lattice fields are required to be sufficiently smooth (the admissibility condition) and the hole flux bound is chosen strictly, the two integers ν and ν-tilde counting total magnetic flux on the original and dual lattices are equal; the paper does not derive this equality from an explicit bound, and the parameter governing the hole flux bound is left unspecified.

Editorial extensions

If this is right

The construction yields a lattice-regularized 2D U(1) chiral gauge theory that is exactly gauge invariant at finite lattice spacing whenever the anomaly cancellation condition holds.
Vector-charged objects can be defined in admissibility-constrained lattice theories through holes, resolving the obstruction that smoothness poses to topological (magnetic) charges.
The selection rules (4.9) guarantee that correlation functions of gauge-invariant charged operators obey the continuum index-theorem constraints, so the anomaly structure is built in at the regularized level.
Because the gauge anomaly is computed classically from the action, no non-perturbative fine-tuning of the fermion measure is needed to recover the continuum anomaly.
The method may extend to non-Abelian chiral gauge theories via non-Abelian bosonization, where the fundamental variable is a compact U(N)-valued field and smoothness is again essential.

Reading between the lines

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

A natural next step is to derive the explicit bound on |ν - ν-tilde| in terms of ε, δ, and δ'; if the bound cannot be made zero for any non-trivial hole, the assumption would need to be modified, possibly by treating ν and ν-tilde as separate inputs to the selection rules.
The hole construction resembles a lattice version of a 't Hooft operator or magnetic monopole; in higher-dimensional bosonization-based constructions, analogous excision defects might serve as the charged objects that admissibility conditions otherwise forbid.
The requirement that q_A and q_V be integers excludes half-integer-charge models such as the 21111 model, so a different non-perturbative treatment may be needed for that class of chiral theories.
A Monte Carlo simulation of the lattice action with a dynamical gauge field and dynamically fluctuating holes could test whether the selection rules hold beyond the classical level and whether the equality ν = ν-tilde is preserved in ensembles of configurations.

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

This is a useful conference contribution, and the main new idea is legitimate: combine the excision method with the bosonized 2D U(1) chiral gauge theory, and represent vector-charged objects as holes in the lattice. The paper then derives selection rules (4.9) that combine vector and axial charges and match the expected continuum anomaly structure. That is genuinely new relative to the authors' earlier work, and the derivation of the lattice anomaly shift (3.15) is explicit and plausible.

The paper is also honest about its weak points. Footnote 3 admits that the operator V_{Q}(D) is not fully defined because boundary conditions around the hole are not fixed, and the text notes that near the hole the dual-lattice link correspondence breaks and the extra links are treated as independent degrees of freedom. These are real gaps, but they are stated openly.

The larger soft spot is the equality of the two Chern numbers. The selection rule (4.9) follows only if nu = nu-tilde, and the justification after (4.7) is a single sentence: the difference is an integer and can be bounded using delta and delta-prime. No bound is exhibited, and delta-prime is never specified. Since the difference is an integer, a bound smaller than 1 would be enough, but it has to follow from the actual ranges in (3.8) and (3.13) while still allowing a hole large enough to carry nonzero charge. Without that, (4.9) is assumed, not derived. This is exactly the load-bearing step, and the stress-test note is correct on this point.

Also worth saying: the selection rules are a consistency check that rewrites the continuum bosonization dictionary in lattice language. That is fine and useful, but it is not an independent derivation of the anomaly.

The math looks coherent and the citations to the prior work of the same group are appropriate. This is a proceedings talk, so the gaps are not disqualifying. The paper deserves a serious referee, and I hope the authors close the nu = nu-tilde gap and the boundary-condition definition in the full version.

Referee Report

3 major / 4 minor

Summary. This proceedings paper proposes a lattice formulation of 2D U(1) chiral gauge theory based on Abelian bosonization and the 'excision method.' Section 3 defines compact boson and U(1) link variables, imposes the admissibility condition (3.8), and represents vector-charged objects as holes D excised from the lattice, with vector charge Q_alpha defined in Eq. (3.11). The lattice action (3.14) is constructed from the bosonized action, and its gauge anomaly shift (3.15) leads to the cancellation condition (3.16). Section 4 derives selection rules for axial and vector charges, culminating in Eq. (4.9), which is claimed to reproduce the continuum fermion-number anomaly selection rules. The central claim is that these lattice definitions reproduce the continuum anomaly structure at finite lattice spacing.

Significance. The topic is significant for nonperturbative lattice chiral gauge theory: the bosonization route has recently produced lattice formulations with exact chiral gauge symmetry at finite spacing, and the excision method is a promising way to accommodate magnetic (vector-charged) objects under the admissibility condition. The paper gives explicit formulas for the anomaly shift, integer-valued Chern numbers, and a concrete dictionary between bosonic operators and chiral fermions. The main limitations are technical rather than conceptual, but they affect the central derivation: the vector-charged operator is not fully defined, and the equality of the two Chern numbers used in Eq. (4.9) is not established quantitatively.

major comments (3)

[Section 4, Eqs. (4.6)–(4.9)] The derivation of the final selection rules requires nu = nu-tilde, but this equality is only asserted. After Eq. (4.7) the authors state that nu - nu-tilde is a finite sum of F/2pi near D, is an integer, and can be bounded using delta and delta-prime, 'hence, assuming sufficiently strict admissibility, we can justify nu = nu-tilde.' No explicit formula for nu - nu-tilde is given, no bound |nu - nu-tilde| < 1 is derived from the ranges in (3.8) and (3.13), and delta-prime in (3.13) is never assigned a range. Since nu - nu-tilde is an integer, a bound smaller than 1 would suffice, but it must be derived from the actual hole geometry and must be compatible with the requirement that the hole perimeter exceed 2pi/epsilon links. Without this step, Eq. (4.9) is an assumption rather than a consequence of the lattice definitions. Please supply the quantitative argument and specify delta-prime.
[Section 3.2, footnote 3, and Section 4, Eq. (4.1)] The object V_{Q_alpha}(D) is used throughout Section 4 as an operator, but footnote 3 explicitly states that simply creating a hole does not fix Q_alpha and that boundary conditions around dD must be fixed to yield specific Q_alpha. These boundary conditions are never constructed. Consequently, Eq. (3.17) describes a gauge transformation of field configurations with definite Q_alpha, and the gauge-invariant dressing (4.1) is not yet an operator definition. Please define V_{Q_alpha}(D) as a genuine operator, for example by a prescribed sum over boundary conditions or a constrained path integral, and prove that its gauge transformation property is the one used in (3.17). This is load-bearing because the selection rules in Section 4 are meant to be statements about correlation functions of this operator.
[Section 3.3, Eq. (3.17)] The shift of the lattice action in the presence of a hole is a key input to the construction, but Eq. (3.17) is presented without intermediate steps. The passage from the anomaly shift (3.15) to the boundary term proportional to -i q_V,alpha Q_alpha Lambda(tilde x*) involves manipulations of the terms at links on dD and uses the definition (3.11) of Q_alpha; as written, the reader cannot verify that no additional boundary terms survive. Please show the calculation at least schematically, or explicitly state that the details are contained in the companion paper [9].

minor comments (4)

[Footnotes 1 and 4] Footnote 1 sets R^2 = 1/2, while footnote 4 says that in the continuum limit R^2 must be tuned to a specific value; please clarify whether the selection rules (4.9) depend on this tuning or hold at finite lattice spacing for any R^2.
[Eqs. (4.2)–(4.3)] The notation involving 'sum over tilde C D tilde C' is hard to parse because the labels for holes, plaquettes, and the complement of the excised regions are not defined precisely. Please state clearly the summation domains and the meaning of each label.
[Section 3.2, after Eq. (3.13)] The claim that Q_alpha can take non-zero values when the number of links forming dD exceeds 2pi/epsilon is stated without proof; please phrase it in terms of the perimeter and give a short derivation or defer to Ref. [1].
[Eq. (4.8)] In the submitted source, the operator dictionary in Eq. (4.8) is not fully legible; please ensure that the assignments of the axial charges B = ±1/2 and vector charges Q = ±1 are displayed unambiguously and explained in the text.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the selection rules follow directly from the lattice definitions; the ν = ν̃ step is an unproven admissibility assumption, not a circular reduction.

full rationale

The paper's derivation chain is self-contained. The lattice action (3.14) is written down from the bosonization dictionary, the gauge variation is computed directly in (3.15)–(3.16), and the selection rules (4.2) and (4.6) are derived from the definition of Q_α in (3.11) and from invariance under constant shifts of φ, respectively. These are algebraic consequences of the stated lattice definitions, not fitted quantities or renamed predictions. The comparison with the continuum anomaly (4.10) is a consistency check, not a first-principles prediction. Refs. [1] and [9] are self-citations supplying the excision method and the dual vertex operator, but the present paper re-derives the charged-object transformation property (3.17) and the selection rules from its own definitions, so the self-citations are not load-bearing in the final anomaly statement. The one genuine weakness is the step after (4.7), where ν = ν̃ is justified only by 'assuming sufficiently strict admissibility' without a quantitative bound; this is an unproven assumption and a correctness risk, but it is not a circular reduction.

Assumptions & free parameters 2 free parameters · 6 assumptions · 1 invented entities

The construction rests on the 2D bosonization dictionary, the admissibility condition, the excision method, and several assumptions left as open points in the paper: fixed boundary conditions at holes, tuning of R^2, and the equality of the two Chern numbers.

free parameters (2)

Compactification radius R^2 = 1/2 at classical level; must be tuned to a non-classical value for the continuum limit (footnote 4)
The lattice action (3.14) sets R^2=1/2 to represent a 2D fermion, but the paper acknowledges the radius should be tuned for the continuum limit, making it a free parameter of the construction.
Admissibility bounds epsilon, delta, delta-prime = 0 < epsilon < pi/2, 0 < delta < min(pi, 2*pi - 4*epsilon), delta-prime unspecified
These cutoffs constrain field smoothness (Eqs. (3.8) and (3.13)). They are regulator parameters chosen by hand; the paper leaves delta-prime unspecified, which matters for the nu = nu-tilde step.

assumptions (6)

domain assumption 2D Abelian bosonization duality between compact bosons and Dirac fermions holds, including the operator dictionary for axial and vector symmetries.
The whole construction relies on Refs. [7,8,12]; the operator identifications in Sections 3 and 4 depend on this duality.
domain assumption The admissibility condition is a valid smoothness constraint that preserves the topological properties of the lattice theory.
Imported from Refs. [2,10,11]; used in Eq. (3.8) to guarantee Eq. (3.10).
domain assumption The excision method from Ref. [1] can be carried over to the bosonized chiral gauge theory, with holes located at dual lattice sites.
The paper assumes holes are the correct lattice representation of vector-charged objects, referencing Ref. [1] without re-deriving the excision construction.
ad hoc to paper Boundary conditions around a hole can be fixed so that the vector charge Q_alpha takes a specified value, without disturbing the rest of the theory.
Footnote 3 states that simply creating a hole does not fix Q_alpha; the operator e^{i Q_alpha phi-tilde} requires fixed boundary conditions, which are not constructed in the paper.
ad hoc to paper Under sufficiently strict admissibility, the two Chern numbers nu and nu-tilde are equal.
Section 4 states this to combine Eqs. (4.2) and (4.6); no quantitative bound is derived, and it depends on delta-prime.
domain assumption The lattice action (3.14) lies in the same universality class as the continuum chiral gauge theory after tuning R^2.
This is the goal of the construction; footnotes 4 and 5 acknowledge that the radius must be tuned to reach the continuum limit, which is not demonstrated.

invented entities (1)

Hole D (excised lattice region) independent evidence
purpose: Lattice defect used to realize vector-charged (magnetic) objects under the admissibility condition.
The hole is a new construction device introduced by the paper (following Ref. [1]); its charges obey selection rules consistent with the known continuum anomaly, providing an external (continuum) check, though no direct experimental handle exists.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Novel Lattice Formulation of 2D Chiral Gauge Theory via Bosonization." pith.science (2026). https://pith.science/paper/RFYB5RP6

@misc{pith2026250118949,
  author       = {Pith},
  title        = {Pith review of: Novel Lattice Formulation of 2D Chiral Gauge Theory via Bosonization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RFYB5RP6}},
  note         = {Machine review of arXiv:2501.18949}
}

read the original abstract

Recently, lattice formulations of 2D Abelian chiral gauge theory have been constructed based on Abelian bosonization. It is remarkable about these 2D lattice formulations that they reproduce the same gauge anomaly structure as the continuum theory, even at a finite lattice spacing. In this talk, we propose yet another lattice formulation based on the ``excision method'' introduced recently in Ref.~\cite{Abe:2023uan}. This approach respects the admissibility condition, which is a constraint on the smoothness of lattice field configurations; it usually prohibits magnetically charged objects, that is, vector-charged objects in fermion theories. We show that such objects can be defined in the excision method as a lattice defect called a ``hole,'' and discuss the selection rules for charged objects.

Figures

Figures reproduced from arXiv: 2501.18949 by the authors.

**Figure 1.** A “hole” D excised from 2, with the corresponding dual lattice (dashed lines). Inside the excised region, a site ˜∗ of the dual lattice is located, as shown in the figure. We note that near the “hole,” the one-to-one correspondence between the links of the original lattice and those of the dual lattice does not hold in general. Thus, in such cases, the relation (3.4) breaks. In such cases, we treat the “extra” link … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

[9]

Morikawa, S

O. Morikawa, S. Onoda and H. Suzuki, Yet Another Lattice Formulation of 2D U(1) Chiral Gauge Theory via Bosonization, PTEP 2024 (2024) 063B01 [24/zero.alt33./zero.alt3342/zero.alt3]

work page 2024
[1]

M. Abe, O. Morikawa, S. Onoda, H. Suzuki and Y . Tanizaki, Magnetic operators in 2D compact scalar ﬁeld theories on the lattice , PTEP 2023 (2023) 073B01 [23/zero.alt34.14815]

work page 2023
[2]

Lüscher, Abelian chiral gauge theories on the lattice with exact gaug e invariance, Nucl

M. Lüscher, Abelian chiral gauge theories on the lattice with exact gaug e invariance, Nucl. Phys. B 549 (1999) 295 [hep-lat/9811/zero.alt332]

work page 1999
[3]

Kikukawa and Y

Y . Kikukawa and Y . Nakayama,Gauge anomaly cancellations in SU(2)(L) x U(1)(Y) electroweak theory on the lattice , Nucl. Phys. B 597 (2001) 519 [hep-lat//zero.alt3/zero.alt3/zero.alt35/zero.alt315]

work page 2001
[4]

Kadoh and Y

D. Kadoh and Y . Kikukawa, A Construction of the Glashow-Weinberg-Salam model on the lattice with exact gauge invariance, JHEP 05 (2008) 095 [/zero.alt37/zero.alt39.3658]

work page 2008
[5]

DeMarco, E

M. DeMarco, E. Lake and X.-G. Wen, A Lattice Chiral Boson Theory in 1 + 1d, 23/zero.alt35./zero.alt33/zero.alt324

work page
[6]

Berkowitz, A

E. Berkowitz, A. Cherman and T. Jacobson, Exact lattice chiral symmetry in 2D gauge theory, Phys. Rev. D 110 (2024) 014510 [231/zero.alt3.17539]

work page 2024
[7]

Coleman, The Quantum Sine-Gordon Equation as the Massive Thirring Mo del, Phys

S.R. Coleman, The Quantum Sine-Gordon Equation as the Massive Thirring Mo del, Phys. Rev. D 11 (1975) 2088

work page 1975

Show all 17 references

[8]

Mandelstam, Soliton Operators for the Quantized Sine-Gordon Equation , Phys

S. Mandelstam, Soliton Operators for the Quantized Sine-Gordon Equation , Phys. Rev. D 11 (1975) 3026

1975
[10]

Lüscher, Topology of Lattice Gauge Fields, Commun

M. Lüscher, Topology of Lattice Gauge Fields, Commun. Math. Phys. 85 (1982) 39

1982
[11]

Hernández, K

P. Hernández, K. Jansen and M. Lüscher, Locality properties of Neuberger’s lattice Dirac operator, Nucl. Phys. B 552 (1999) 363 [hep-lat/98/zero.alt38/zero.alt31/zero.alt3]

1999
[12]

Lectures on Gauge Theory

D. Tong, “Lectures on Gauge Theory.” http://www.damtp.cam.ac.uk/user/tong/gaugetheory.html
[13]

Janke and K

W . Janke and K. Nather, High precision MonteCarlo study of the 2-dimensional XY Vil lain model, Phys. Rev. B 48 (1993) 7419

1993
[14]

Fujikawa, Generalized Pauli-Villars regularization and the covariant form of anomalies , Nucl

K. Fujikawa, Generalized Pauli-Villars regularization and the covariant form of anomalies , Nucl. Phys. B 428 (1994) 169 [hep-th/94/zero.alt35166]

1994
[15]

Shiozaki, A discrete formulation for three-dimensional winding numb er, 24/zero.alt33./zero.alt35291

K. Shiozaki, A discrete formulation for three-dimensional winding numb er, 24/zero.alt33./zero.alt35291
[16]

Witten, Nonabelian Bosonization in Two-Dimensions, Commun

E. Witten, Nonabelian Bosonization in Two-Dimensions, Commun. Math. Phys. 92 (1984) 455

1984
[17]

Chen,Instanton Density Operator in Lattice QCD from Higher Categ ory Theory , 24/zero.alt36./zero.alt36673

J.- Y . Chen,Instanton Density Operator in Lattice QCD from Higher Categ ory Theory , 24/zero.alt36./zero.alt36673. 9

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [9]

Morikawa, S

O. Morikawa, S. Onoda and H. Suzuki, Yet Another Lattice Formulation of 2D U(1) Chiral Gauge Theory via Bosonization, PTEP 2024 (2024) 063B01 [24/zero.alt33./zero.alt3342/zero.alt3]

work page 2024

[2] [1]

M. Abe, O. Morikawa, S. Onoda, H. Suzuki and Y . Tanizaki, Magnetic operators in 2D compact scalar ﬁeld theories on the lattice , PTEP 2023 (2023) 073B01 [23/zero.alt34.14815]

work page 2023

[3] [2]

Lüscher, Abelian chiral gauge theories on the lattice with exact gaug e invariance, Nucl

M. Lüscher, Abelian chiral gauge theories on the lattice with exact gaug e invariance, Nucl. Phys. B 549 (1999) 295 [hep-lat/9811/zero.alt332]

work page 1999

[4] [3]

Kikukawa and Y

Y . Kikukawa and Y . Nakayama,Gauge anomaly cancellations in SU(2)(L) x U(1)(Y) electroweak theory on the lattice , Nucl. Phys. B 597 (2001) 519 [hep-lat//zero.alt3/zero.alt3/zero.alt35/zero.alt315]

work page 2001

[5] [4]

Kadoh and Y

D. Kadoh and Y . Kikukawa, A Construction of the Glashow-Weinberg-Salam model on the lattice with exact gauge invariance, JHEP 05 (2008) 095 [/zero.alt37/zero.alt39.3658]

work page 2008

[6] [5]

DeMarco, E

M. DeMarco, E. Lake and X.-G. Wen, A Lattice Chiral Boson Theory in 1 + 1d, 23/zero.alt35./zero.alt33/zero.alt324

work page

[7] [6]

Berkowitz, A

E. Berkowitz, A. Cherman and T. Jacobson, Exact lattice chiral symmetry in 2D gauge theory, Phys. Rev. D 110 (2024) 014510 [231/zero.alt3.17539]

work page 2024

[8] [7]

Coleman, The Quantum Sine-Gordon Equation as the Massive Thirring Mo del, Phys

S.R. Coleman, The Quantum Sine-Gordon Equation as the Massive Thirring Mo del, Phys. Rev. D 11 (1975) 2088

work page 1975