Pith. sign in

REVIEW 4 major objections 5 minor 20 references

What do we know about the confinement mechanism?

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read Random gauge copies, restricted to their Gaussian distribution, reproduce the QCD string tension on the lattice.

desk verdict A real but incremental lattice observation wrapped in a review, whose central 'prediction' is weakened by post hoc calibration and missing error bars. read the letter →

arxiv 2412.10767 v1 pith:4OFIP2AO submitted 2024-12-14 hep-lat

classification hep-lat PACS 12.38.-t12.38.Aw12.38.Gc
keywords colorconfinementcentervorticesmaximalgaugeGribovcopiesstringtensionlatticeQCDmonopolesSU(2)theory
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

This paper argues that the confining string tension can be read from center-projected Wilson loops after a modified maximal center gauge (MCG) prescription: instead of searching for the global maximum of the gauge functional, one should restrict maximization to the Gaussian-distributed subset of random gauge copies. On a $24^4$ lattice at coupling strengths $\beta = 2.3$ to $2.6$, the local maxima of the MCG functional over random copies are nearly symmetric Gaussians, and the string tensions extracted from the high-$R_{MCG}$ tail of these distributions agree with the unprojected string tensions in the literature. The same procedure works at $\beta = 2.7$ after excluding copies whose potentials indicate mis-detected vortices. If right, this gives a practical answer to the Gribov-copy problem in vortex detection and strengthens the center vortex model of confinement.

What carries the argument

The maximal center gauge (MCG) functional $R_{MCG}$ of equation (1), which measures how close gauge-transformed link variables are to center elements, together with center projection of links onto the nearest center element $\pm 1$ for SU(2). The new element is the statistical treatment of the ensemble of local maxima: the paper works with the distribution of $R_{MCG}$ over random gauge copies rather than with a single best copy, restricts the maximization to the Gaussian core of that distribution, and uses the near-linearity of $\sigma_{cp}(R_{MCG})$ to select copies that reproduce the unprojected string tension.

What would settle it

Repeat the analysis on a larger lattice or with a different lattice action and check whether the Gaussian-restricted prescription reproduces the unprojected string tension without any symmetrization or copy exclusion; if agreement at $\beta=2.7$ requires removing the non-Gaussian tails, the claim that the random-copy ensemble contains the full string-tension information would be selection-dependent rather than measured.

Watch

Extended reading notes

Core claim

The paper's central claim is that the information needed for the string tension is contained in the ensemble of local maxima of the MCG functional, quantified by $R_{MCG} = \sum_x \sum_\mu |\mathrm{Tr}[{}^g U_\mu(x)]|^2$, and that it must be extracted by maximizing only within the approximately Gaussian distribution of $R_{MCG}$ values over random gauge copies. With 100 random copies per configuration on a $24^4$ lattice, the paper finds Gaussian distributions for $2.3 \le \beta \le 2.6$, with small deviations at $\beta=2.6$ and large ones at $\beta=2.7$. Because the center-projected string tension $\sigma_{cp}$ is nearly linear in $R_{MCG}$, taking the two or three highest-$R_{MCG}$ copies within the Gaussian subset reproduces the unprojected string tensions from the literature, whereas unrestricted maximization underestimates them. For $\beta=2.6$ the ensembles are symmetrized and for $\beta=2.7$ copies with reversed $R$-dependence of the potential are excluded, after which the remaining copies agree with the other couplings.

Load-bearing premise

The load-bearing assumption is that the non-Gaussian deviations seen at $\beta=2.6$ and $\beta=2.7$ are artifacts of incorrect vortex detection rather than physical effects; this is what justifies symmetrizing the $\beta=2.6$ ensembles and dropping the $\beta=2.7$ copies that show an anomalous potential.

Editorial extensions

If this is right

  • Unrestricted MCG maximization systematically lowers the center-projected string tension, so the global maximum of the gauge functional is the wrong target for vortex detection.
  • Taking the two or three highest-$R_{MCG}$ copies inside the Gaussian subset reproduces the unprojected string tension for $2.3 \le \beta \le 2.6$, while taking more than about seven copies overestimates it.
  • The nearly linear relation between $\sigma_{cp}$ and $R_{MCG}$ turns the Gribov-copy ensemble into a controlled statistical sample instead of an ambiguity to be minimized away.
  • At $\beta=2.6$ and $\beta=2.7$, the projected potential itself flags bad copies: when the large-$R$ string tension falls below the small-$R$ value, the copy has mis-detected vortices and can be excluded.

Reading between the lines

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

  • A testable extension would be to apply the same Gaussian-restricted selection rule to SU(3) gauge theory, where center projection uses the three nontrivial center elements and the distribution of local maxima may take a different shape.
  • The non-Gaussian tails at $\beta=2.6$ and $\beta=2.7$ could be cross-checked with an independent vortex detector; if the excluded copies still contain thick vortices, the exclusion criterion would be a selection effect rather than a physical diagnosis.
  • The linear $\sigma_{cp}$--$R_{MCG}$ relation suggests an extrapolation prescription: read the string tension at the Gaussian tip from a linear fit, which would reduce the number of random copies needed per configuration.
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.

Referee Report

4 major / 5 minor

Summary. The paper reviews the dual-superconductor and center-vortex pictures of confinement and presents new lattice SU(2) simulations in maximal center gauge (MCG) on 24^4 lattices at β = 2.3, 2.4, 2.5, 2.6, and 2.7. For each β the authors generate 200 physical configurations, with 100 random gauge copies each, and analyze the distribution of the MCG functional R_MCG at local maxima. They report that these distributions are approximately Gaussian for β ≤ 2.5, that the center-projected string tension σ_cp varies approximately linearly with R_MCG, and that averaging the N = 2 or N = 3 gauge copies with the highest R_MCG values within the Gaussian subset reproduces literature values of the unprojected string tension. The conclusion states that maximization of Eq. (1) restricted to the Gaussian-distributed copies gives very good predictions of the string tension for 2.3 ≤ β ≤ 2.6, with a modified exclusion also working at β = 2.7.

Significance. If the restricted-maximization rule were fixed before comparison and supported by statistical uncertainties, this would be a substantial step toward resolving the Gribov-copy problem in center projection and would strengthen the center-vortex mechanism as a quantitative description of confinement. The paper's main positive contribution is the documentation of the R_MCG distribution across β and the observed nearly linear relation between σ_cp and R_MCG, which is a useful empirical constraint on MCG-based vortex detection. However, the central predictive claim is not yet established: the prescription is calibrated to literature string tensions after inspecting the data, and the reported agreement is not accompanied by error bars or an independent validation. The manuscript is therefore better characterized as a suggestive empirical study than as a demonstrated prediction.

major comments (4)
  1. [Sec. 3, Figs. 3–5] No statistical uncertainties are reported for σ_cp, and the effective independent sample size is much smaller than the 20,000 gauge fields per β, since these arise from only 200 physical configurations (20 starts × 10 separations) with 100 correlated gauge copies each. Without error bars, the apparent agreement between the center-projected and literature string tensions in Fig. 5 cannot be assessed; deviations such as the variations visible for β = 2.6 could be statistically insignificant. The claim of 'very good predictions' requires a quantitative comparison with uncertainties.
  2. [Sec. 3, Fig. 5 and Conclusion] The choice N = 2–3 is made after inspecting the comparison with literature values in Fig. 5, not from an a priori rule. Because σ_cp varies approximately linearly with R_MCG and N selects the average R_MCG of the N highest copies, the procedure can reproduce any target string tension lying between the N = 1 and large-N limits by tuning N. The observed agreement at N = 2–3 is therefore calibration, not a prediction, unless the selection rule is fixed before comparison and tested on independent data.
  3. [Sec. 3, Figs. 2 and 4] The Gaussian-restriction prescription is defined post hoc. For β = 2.6 the ensembles are symmetrized following Ref. [19], and for β = 2.7 gauge copies are excluded using a criterion based on the reversal of the projected potential; both steps are introduced only after non-Gaussian deviations are seen in Fig. 2. The assertion that these deviations are artifacts of incorrect vortex detection is unsupported, and no independent diagnostic distinguishes 'incorrect vortex detection' from physical finite-lattice or weak-coupling effects. Thus the claimed range 2.3 ≤ β ≤ 2.6 does not uniformly satisfy the Gaussian condition, since β = 2.6 already shows systematic deviations before symmetrization.
  4. [Sec. 3, Fig. 3] The literature string tensions used as reference lines [8,16–18] are taken from different lattice actions, volumes, and extrapolation procedures. The manuscript does not discuss how these differences affect the comparison, nor does it compute unprojected string tensions on the same 24^4 ensembles used for the center-projected values. A direct same-ensemble comparison would remove an uncontrolled systematic and make the claimed agreement meaningful.
minor comments (5)
  1. [Fig. 3 caption] The caption contains a typo: 'β =∈ {2.3, 2.4, 2.5}' should read 'β ∈ {2.3, 2.4, 2.5}'.
  2. [Sec. 2] The text says 'colorfull' where 'colorful' is intended, and the phrase 'precautious linearity' is unclear; a more standard wording such as 'approximate linearity' would be preferable.
  3. [Fig. 5 caption] The caption contains a stray period and an incomplete sentence after 'gauge copies.'; it should be cleaned up.
  4. [References] Ref. [19] lists the arXiv number 1908.09711, which is the same as Ref. [2] (GlueX); this is likely a citation error and should be corrected.
  5. [Sec. 3, Fig. 2] The skewness panel would be more informative with error bars or at least a statement of how skewness uncertainties were estimated, since the claim of 'nearly symmetric' for β = 2.3–2.5 is central to the analysis.

Circularity Check

2 steps flagged · score 6.0 of 10

The string-tension 'prediction' is calibrated by post hoc choice of N and by copy-symmetrization/exclusion rules, so the central claim is partly fitted to its target.

  1. fitted input called prediction [Section 3, Fig. 5 discussion (page 6-7); Conclusion (page 8)]
    "For N = 1, the highest RMCG values underestimate the string tension. Since RMCG and σcp are approximately linearly related, an increase in N leads to an increase in string tension. In general, we observe good agreement between center projected and unprojected string tensions with two or three gauge copies. For N > 7, the string tensions are overestimated."

    N is not fixed before comparing to the literature values; it is scanned, and N = 2-3 is reported as good because it agrees with the unprojected string tensions in Fig. 5. Since σcp is approximately linear in RMCG, and selecting the N highest-RMCG copies sets the average RMCG of the sample, N acts as a tuning knob that moves σcp along the measured curve. Any target between the N=1 and large-N limits can be matched by some N. Calling the so-tuned agreement a 'prediction' in the Conclusion is therefore a fitted-input-as-prediction step, not an independent test. No statistical uncertainties are given, so the significance of the agreement is not established.

  2. fitted input called prediction [Section 3, Fig. 4 discussion (page 6)]
    "For β = 2.7, for which we have already found strong deviations from the Gaussian distribution in Fig. 2, we can either accept that this β lies outside the window in which MCG works, or better try to determine the cause of the deviations. ... This reversal of the behavior of the projected string tension is now used as a criterion to exclude gauge copies from the averaging. The remaining copies, which are corrected and symmetrized, result in the last diagram in Fig. 4 and show a behavior that is consistent with the other β values."

    The exclusion criterion is defined after observing that high-RMCG copies reverse the expected ordering of small-R and large-R string tensions, i.e., after inspecting the data that the prescription is supposed to reproduce. The copies are then removed until the remaining set is 'consistent with the other β values' and matches literature. This is a post hoc selection, so the β=2.6/2.7 agreement is partly guaranteed by the selection rule. The β=2.6 symmetrization is likewise justified by the authors' previous Ref. [19] rather than by an independent test.

full rationale

The paper contains an independent and interesting observation: the RMCG local-maximum distribution is approximately Gaussian for β = 2.3-2.5, and σcp depends roughly linearly on RMCG. These measurements do not reduce to definitions or to the target string tensions. However, the paper's central claim that the maximization 'within these random Gaussian distributions' predicts string tensions is not self-contained. The agreement is obtained after (i) varying N and selecting 2-3 because that is where the curves cross the literature values, and (ii) symmetrizing/excluding copies for β = 2.6/2.7 using criteria derived from the same data. Since σcp is monotonically related to RMCG, the N selection is effectively a one-parameter calibration to the literature string tensions; the agreement is therefore partly fitted rather than predicted. The absence of error bars and the non-independence of the 200 copies further weaken the evidential value. These issues concern the predictive step, not the entire paper; the Gaussianity observation and the vortex phenomenology review remain independent content. Hence a score of 6 rather than 8-10.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the inherited vortex-model assumption of center dominance and on several empirically selected recipe choices (N, symmetrization, beta=2.7 exclusion) that are calibrated against literature string tensions. No new theoretical entities are proposed.

free parameters (3)
  • N (number of highest-R_MCG gauge copies averaged) = 2-3
    Figure 5 shows N=1 underestimates and N>7 overestimates the string tension; the intermediate range is recommended because it matches the literature values, so N is selected to fit the benchmark.
  • beta=2.7 exclusion criterion = sigma_largeR < sigma_smallR
    Gauge copies whose projected potential reverses slope are discarded. This rule is introduced ad hoc for beta=2.7 and is not derived from the center vortex model.
  • Gaussian subset and symmetrization prescription = one-third standard deviation bin widths; symmetrized ensembles
    The bin width, the Gaussian subset, and the symmetrization procedure for beta=2.6 and 2.7 are numerical choices needed to define the ensemble that reproduces the literature string tension.
assumptions (4)
  • standard math SU(2) lattice gauge theory with Wilson action and periodic boundary conditions provides the ensemble of gauge fields.
    Section 3, lattice setup: 24^4 lattice, beta=2.3-2.7, Monte Carlo with 20 random starts and 10 configurations.
  • domain assumption Center dominance: center-projected Wilson loops (P-vortices) carry the physical string tension.
    Section 2 and the interpretation of sigma_cp in Section 3; this is the vortex-model hypothesis the paper assumes rather than tests.
  • domain assumption Local maxima of the MCG functional from random gauge copies contain the physical vortex information.
    Section 3, after Ref. [12]; the paper assumes the ensemble of local maxima encodes the string tension even though each individual maximum does not.
  • ad hoc to paper Deviations from a Gaussian R_MCG distribution are numerical artifacts of incorrect vortex detection, so affected copies may be symmetrized or discarded.
    Section 3, beta=2.6 and beta=2.7 paragraphs; this premise makes the final agreement with literature possible.

how reviews work

0 comments
Cite this review

Pith. "Pith review of What do we know about the confinement mechanism?." pith.science (2026). https://pith.science/paper/4OFIP2AO

@misc{pith2026241210767,
  author       = {Pith},
  title        = {Pith review of: What do we know about the confinement mechanism?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4OFIP2AO}},
  note         = {Machine review of arXiv:2412.10767}
}
read the original abstract

Color confinement is a fundamental phenomenon in quantum chromodynamics. In this work, the mechanisms underlying color confinement are investigated in detail, with a particular focus on the role of non-perturbative phenomena such as center vortices and monopoles in the QCD vacuum. By exploring lattice QCD approaches, including the Maximal Center Gauge and center projection methods, we examine how these topological structures contribute to the confining force between color charges. We also address the limitations of conventional methods and suggest improvements to the gauge fixing prescription to enhance the accuracy of string tension predictions. Our findings support the validity of the center vortex model as a key candidate for understanding the dynamics of the confining QCD vacuum.

Figures

Figures reproduced from arXiv: 2412.10767 by the authors.

Figure 1
Figure 1. a) A schematic diagram of a colorful vortex. Observing such a vortex with a [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. For the lattices with β = 2.3 to 2.7, the distributions of 20 000 local gauge maxima of RMCG values are shown on a logarithmic scale. The blue solid lines represent smooth kernels, with the histogram of the data in the background. Red highlights indicate deviations from the normal distribution. The results of distribution tests are provided below the distribution. The last plot illustrates the variation in skewness … view at source ↗
Figure 3
Figure 3. For lattices with β =∈ {2.3, 2.4, 2.5}, the center-projected string tensions, derived from 200 configurations around specified RMCG values, are compared to the unprojected string tensions reported in the literature, represented by horizontal black lines [8,16–18]. The Gaussian distributions of the corresponding local gauge maxima are indicated by dash-dotted lines. The intersection of the red data with the literatur… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: As an example, for the lattice with β = 2.5, the first plot shows the potential Vcp(R) extracted from 200 configurations centered around RMCG = 0.89053. The two other diagrams are analogous to the diagrams in [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: For lattices with β = 2.3 to 2.7, center projected string tensions are ex￾tracted from 200 physical fields from the N gauge copies with the highest values of RMCG within the Gaussian distributed subset of the ensemble of random gauge copies. . The horizontal black line…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 17 canonical work pages

  1. [15]

    Dehghan Z., Golubich R., H¨ ollwieser R., Faber M.A First Analysis of the Ensemble of Local Maxima of Maximal Center Gauge // Universe— 2023.— V. 9, no. 9.— P. 389.— arXiv:2301.05903 [hep-lat]

  2. [19]

    Dehghan Z., Golubich R., H¨ ollwieser R., Faber M.Investigation of the ensemble of maximal center gauge // Phys. Rev. D— 2024.— V. 110— P. 014501, no. 1.— arXiv:1908.09711 [hep-lat]

  3. [1]

    V., Berdnikov V

    Somov S. V., Berdnikov V. V., Tolstukhin I. A., Somov A. S.Study of the Nature of the Confinement in the GlueX Experiment // Phys. Procedia— 2015.— V. 74— P. 86–91. 9

  4. [3]

    Vortices and Quark Confinement in Nonabelian Gauge Theories // Phys

    Mandelstam S. Vortices and Quark Confinement in Nonabelian Gauge Theories // Phys. Rept.— 1976.— V. 23— P. 245–249

  5. [4]

    ’t Hooft GerardOn the Phase Transition Towards Permanent Quark Con- finement // Nucl. Phys. B— 1978.— V. 138— P. 1–25

  6. [5]

    L., Schierholz G., Wiese U

    Kronfeld Andreas S., Laursen M. L., Schierholz G., Wiese U. J.Monopole CondensationandColorConfinement//Phys.Lett.B— 1987.— V.198— P. 516–520.—

  7. [6]

    Del Debbio L., Faber M., Greensite J., Olejnik, S.Center dominance and Z(2) vortices in SU(2) lattice gauge theory // Phys. Rev. D— 1997.— V. 55— P. 2298–2306.— arXiv:9610005 [hep-lat]

  8. [7]

    Faber M., Greensite J., Olejnik S.Casimir scaling from center vortices: Towards an understanding of the adjoint string tension // Phys. Rev. D— 1998.— V. 57— P. 2603–2609.— arXiv:9710039 [hep-lat]

Show all 20 references
  1. [8]

    S., Schilling K., Schlichter C.Observing long color flux tubes in SU(2) lattice gauge theory // Phys

    Bali G. S., Schilling K., Schlichter C.Observing long color flux tubes in SU(2) lattice gauge theory // Phys. Rev. D— 1995.— V. 51— P. 5165– 5198.— arXiv:9409005 [hep-lat]

  2. [9]

    Greensite J.The Confinement problem in lattice gauge theory // Prog. Part. Nucl. Phys.— 2003.— V. 51— P. 1.— arXiv:0301023 [hep-lat]

  3. [10]

    Del Debbio L., Faber M., Giedt J., Greensite J., Olejnik, S.Detection of center vortices in the lattice Yang-Mills vacuum // Phys. Rev. D— 1998.— V. 58— P. 094501.— arXiv:9801027 [hep-lat]

  4. [11]

    Faber M., Jordan G., H¨ ollwieser R.Topology, Center Vortices, Confine- ment and Chiral Symmetry Breaking in SU(2) Lattice-QCD // Interna- tional School-Seminar on New Physics and Quantum Chromodynamics of External Conditions (NPQCD 2007)— 2007.— P. 27–37

  5. [12]

    Center projection vortices in continuum Yang-Mills theory // Nucl.Phys

    Engelhardt M., Reinhardt H. Center projection vortices in continuum Yang-Mills theory // Nucl.Phys. B— 2000.— V. 567— P. 249–292.— arXiv:9907139 [hep-th]

  6. [13]

    Kovacs Tamas G., Tomboulis E. T. Vortices and confinement at weak coupling // Phys. Rev. D— 1998.— V. 57— P. 4054–4062.— arXiv:9711009 [hep-lat]

  7. [14]

    G., Komarov D

    Bornyakov V. G., Komarov D. A., Polikarpov M. I.Pvorticesanddrama of Gribov copies // Phys. Lett. B— 2001.— V. 497— P. 151–158.— arXiv:0009035 [hep-lat]. 10

  8. [16]

    Caselle M., Nada A., Panero M.Hagedorn spectrum and thermodynam- ics of SU(2) and SU(3) Yang-Mills theories // JHEP— 2015.— V. 07— P. 143.— arXiv:1505.01106 [hep-lat]

  9. [17]

    Fingberg J., Heller Urs M., Karsch, F.Scaling and asymptotic scaling in the SU(2) gauge theory // Nucl. Phys. B— 1993.— V. 392— P. 493– 517.— arXiv:9208012 [hep-lat]

  10. [18]

    Michael C., Perantonis S.Potentials and glueballs at large beta // Nucl. Phys. B Proc. Suppl.— 1991.— V. 20— P. 177–180

  11. [20]

    E.4D ensembles of percolating center vortices and monopole defects: The emergence of flux tubes with N -ality and gluon con- finement // Phys

    Oxman L. E.4D ensembles of percolating center vortices and monopole defects: The emergence of flux tubes with N -ality and gluon con- finement // Phys. Rev. D— 2018.— V. 98— P. 036018, no. 3.— arXiv:1805.06354 [hep-th]

  12. [21]

    R., Oxman L

    Junior D. R., Oxman L. E. Ensembles of center vortices and chains: Insights from a natural lattice framework //— 2024.— arXiv:2411.04325 [hep-th]

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.