A counterexample for the polar conjecture of Spencer-Brown

Ben McCarty; Louis H. Kauffman; Scott Baldridge

REVIEW 4 major objections 4 minor 18 references

A counterexample for the polar conjecture of Spencer-Brown

T0 review · 4 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read The parity-pass algorithm can loop forever on a non-polar pentagon of a planar graph, returning to its starting coloring after sixty steps.

desk verdict A plausible and important counterexample to Spencer-Brown's Polar Conjecture, but the decisive 60-step loop is asserted rather than demonstrated, so the paper needs a verification artifact before the claim can be trusted. read the letter →

arxiv 2607.22398 v1 pith:7ITBXLVL submitted 2026-07-24 math.CO math.GT

Scott Baldridge , Louis H. Kauffman , Ben McCarty This is my paper

classification math.COmath.GT MSC 05C1505C10

keywords paritypasspolarconjecturefour-colortheoremTaitcoloringKempechainsplanargraphsedgecounterexample

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 takes aim at the Polar Conjecture, the claim that Spencer-Brown's parity-pass algorithm—a five-step local sequence of color switches around a pentagonal face—always terminates with an extendable three-edge coloring whenever the pentagon is not radially symmetric. The authors construct a concrete trivalent plane graph with a non-polar pentagon and a one-deficient coloring, and report that the parity pass runs through twelve complete cycles (sixty steps) before restoring the original coloring over the entire graph, without ever producing a completable state. If the trace is correct, the conjecture fails, and any proposed proof of the four-color theorem built on the parity pass must either search over all pentagons or find a stronger algorithm. The paper leaves open the wider question of whether some more flexible sequence of color exchanges always works.

What carries the argument

The parity pass is a five-step local operation (Figure 3) on a 1-deficient 3-edge coloring of a trivalent plane graph, in which alternating two-color paths are switched around the pentagonal face. Its companion is the bad configuration C (Figure 5), a pattern where two such paths overlap in at least one edge; the paper uses C as a certificate that no immediate completion exists and that the next switch in the sequence is forced. The counterexample works by making C persist through every step of the loop.

What would settle it

Simulate the parity pass (Figure 3) on the graph of Figure 8 with the displayed 1-deficient coloring. If any step before sixty yields a coloring that extends to the pentagon (by a double switch or direct completion), or if the whole graph's coloring repeats in fewer than twelve cycles, the counterexample collapses.

Watch

Extended reading notes

Core claim

The central discovery is a single graph G (Figure 8) with a pentagonal face that is not polar, together with a 1-deficient 3-edge coloring. The paper argues that at every step of the parity pass a local 'bad configuration' C is present: two alternating two-color paths that share an edge, making a simple completion impossible. The authors state that configuration C persists through all five transformations A–E and through twelve full cycles, until the entire graph's coloring returns to the original one. Consequently the parity pass never terminates in an extendable coloring, contradicting the Polar Conjecture. The graph lacks rotational symmetry about the pentagon because the opposite octagon

Load-bearing premise

The claim that the parity pass completes exactly sixty steps and returns to the original entire-graph coloring is stated but not fully demonstrated: only one full cycle and the first and last steps are shown, and no machine-readable certificate or code is provided.

Editorial extensions

If this is right

If the sixty-step trace is correct, the Polar Conjecture is false: non-polarity of the pentagon does not guarantee termination of the parity pass.
The proposed constructive proof of the four-color theorem via the parity pass loses its foundation, because an arbitrary non-polar pentagon may loop.
A revised algorithm must either test all pentagonal faces for looping behavior or incorporate a more flexible switching rule.
The example also yields a derived graph (Example 4.2) with a non-symmetric loop, and another one (Example 4.3) where the loop terminates after a finite number of cycles, illustrating the subtlety.
The paper shows that one of Spencer-Brown's two 'parity mills' also loops on the same example, while the other terminates, leaving the question of a universal algorithm open.

Reading between the lines

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

A reader could independently verify the counterexample by implementing the parity pass; if the sixty-step trace reproduces exactly, it would serve as a rigorous, human-checkable small certificate for a non-terminating local algorithm.
The construction suggests a template for building further counterexamples by concatenating cycles of configuration C, potentially producing graphs with multiple non-polar looping pentagons.
The persistence of C for a full loop hints that the impasse group—the algebraic structure generated by the local switches—may encode a practical obstruction to local completability, which a broader algorithm would need to circumvent.
One natural next experiment is to test all non-polar pentagons of the Errera graph and other sparse planar graphs to map how common such looping behavior is.

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 one is worth reading carefully. The paper claims to refute a ~40-year-old conjecture that the parity pass on any non-polar pentagon always terminates in an extendable coloring. If true, that kills the 'human-checkable' 4CT proof program as Spencer-Brown framed it. The authors build an explicit trivalent plane graph, give a 1-deficient coloring, and show via configuration C that the algorithm should loop through twelve complete passes and return to the original global coloring. That is a real, checkable finite claim, and the diagrams are clear enough that I was tempted to start checking by hand.

What's good: the construction is genuinely new, not a rehash of the Errera example; the non-polarity argument using coprime face sizes (pentagon and octagon) is clean; and configuration C gives an honest mechanism for why the loop persists. There is no parameter fitting or circular reasoning. Examples 4.2 and 4.3 are nice, showing both a related looping graph and one where the loop eventually breaks, which helps calibrate how fragile the phenomenon is. The authors also candidly flag the omitted proof in Remark 3.1 and admit Theorem 3.6 is proved for one representative step.

The soft spots are real and all in the same place: the central 60-step trace is asserted in a sentence and 'the reader is encouraged to try this.' No code, no certificate, no table of the sixty intermediate colorings. That matters because the counterexample stands or falls on precisely that trace. A single misapplied switch, or a step where configuration C fails to persist, breaks the argument. Theorem 3.6's proof only covers step C and says the others follow similarly, so it does not certify the full run. There is also a subtle issue the paper never addresses: the parity pass assumes a particular alternating path exists at each step, but if multiple such paths exist, different choices could lead to different runs. The authors don't prove the path is unique or specify a deterministic choice rule. Without that, the loop shown might be an artifact of one branch, and the conjecture might still hold for 'the' algorithm under a canonical choice. That's a substantive gap, not just a presentational one.

Still, the paper is worth seriously engaging, especially for people working on constructive 4CT proofs, Kempe-chain reducibility, and the history of Spencer-Brown's program. The referee's job should be to verify the 60-step run—ideally with a machine-readable certificate or a full enumeration—and to pin down the algorithm's path-choice convention. If those come through, this is a solid counterexample. If the path-choice issue persists, the conclusion may need to be weakened to 'there exists a run of the parity pass that loops on a non-polar pentagon,' which is still interesting but not a straight refutation.

My recommendation: send it to peer review. It deserves careful referee time, but the authors should be expected to provide the missing verification before acceptance.

Referee Report

4 major / 4 minor

Summary. The paper claims to disprove Spencer-Brown's Polar Conjecture: the assertion that performing his parity-pass algorithm on a non-polar pentagonal region necessarily terminates in an edge coloring extendable to the whole trivalent plane graph. The authors introduce a particular plane graph G (Figure 8) whose highlighted pentagon is non-polar, together with a 1-deficient 3-edge coloring. They argue that the parity pass is forced through a sequence of steps, that the bad configuration C persists, and that after twelve complete cycles (sixty steps) the entire coloring returns to its starting configuration, so the algorithm never yields an extendable coloring. The paper also gives two derived examples with shorter purported loops and tests two related 'parity mills' on the same graph, finding that one terminates and one does not. The non-polarity of the pentagon is supported by a coprime-face-size argument (a pentagonal and an octagonal face prevent nontrivial rotational symmetry). The main counterexample, however, rests on an unverified finite trace: Figures 9 and 10 show one complete pass and the first/last configurations, but the sixty intermediate colorings are neither listed nor certified.

Significance. If the counterexample is correct, it refutes a conjecture that has been open for roughly forty years and clarifies the status of Spencer-Brown's proposed proof of the four color theorem. The conceptual core is clean: the definition of the bad configuration C, the non-extendability lemma (Theorem 3.3), and the absence of rotational symmetry via coprime face lengths are all transparent and independent of numerical computation. The manuscript also carefully distinguishes the Polar Conjecture from broader claims about existence of some good pentagon, and it proposes meaningful open questions. The main gap is verification of the claimed 60-step loop: the entire refutation depends on this finite computation, and no code, certificate, or step-by-step table is supplied. Because this gap is explicitly acknowledged in the text ('The reader is encouraged to try this'), the paper is not yet at the standard of verifiability expected for a computational counterexample.

major comments (4)

[Section 4, Figures 9–10] The central claim that the parity pass loops for exactly twelve cycles (sixty steps) and returns the entire 1-deficient coloring to its starting state is asserted but not demonstrated. The text shows one pass and the first/last configurations, then directs the reader to reproduce the calculation. Since the counterexample is precisely this loop, an incorrect switch at any step, a missing alternating path, or an early completable configuration would invalidate the claim. Please provide a machine-verified certificate, explicit code, or a full table of all sixty colorings (or a script that generates and checks them). Without this, the central assertion is unverified.
[Section 3.1, Theorem 3.6] Theorem 3.6 is used to justify that, while configuration C persists, the required alternating path for the next parity-pass step exists. The proof only treats the domain of step C and says 'other steps follow similarly.' This is not a proof for the remaining steps, and those steps are exactly what the sixty-step trace needs. A complete case analysis for all five positions, or a symbolic/algorithmic argument covering every step, is required before the looping behavior is established.
[Section 4 and Figure 3] The parity-pass algorithm does not specify a unique choice of alternating path when several paths connect the required endpoints. The claimed 60-step loop could depend on the path chosen at each step; a different selection might reach a completable coloring and terminate. The paper neither proves uniqueness nor fixes a deterministic path-selection rule. Please state the tie-breaking convention, or prove that every allowed choice leads to the same loop.
[Examples 4.2 and 4.3] The derived examples also rely on unshown finite traces (twenty and ten steps, respectively). Although they are not the main counterexample, they are used to support the phenomenon and to illustrate the subtlety. If a computational certificate is provided for the main graph, please extend it to these examples or clearly mark their verification as conditional on the same unprovided computation.

minor comments (4)

[Remark 3.1] The proof that the local coloring may be assumed, up to color permutation, to match Figure 3 is omitted. This does not affect the explicit counterexample, but it should be stated more carefully as an assumption about Spencer-Brown's starting configuration rather than a general theorem, or the omitted proof should be supplied.
[Section 2, Figures 1–2] The transition from formations to 3-face colorings and then to 3-edge colorings is compressed. For a reader unfamiliar with [5,6], a sentence explaining why the edge color in Figure 2 is the third color would improve accessibility.
[Section 5] The discussion of parity mills reports that φ1 terminates and φ2 does not, but again no trace is shown. If this is intended as more than an anecdotal remark, it should be accompanied by the same certificate or explicitly deferred to a supplement.
[Throughout] The phrase 'radially symmetric' is used for 'polar'; the paper correctly notes reflection symmetry is irrelevant, but the formal definition of polar (and non-polar) should be stated explicitly in Section 3 or 4 to avoid ambiguity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the counterexample rests on an explicit graph and algorithmic run, not on a self-referential derivation.

full rationale

The paper's central claim is a concrete counterexample: an explicit graph G (Figure 8) with a specified non-polar pentagonal face and a 1-deficient 3-edge coloring, together with the claim that the parity pass loops through twelve complete cycles (sixty steps) before returning to the original coloring. This claim does not reduce to any fitted parameter, definitional equivalence, or imported uniqueness theorem. The non-polarity of the pentagon is established by an independent symmetry argument (order of the rotation group dividing both 5 and 8), not by citing the authors' own results. The self-citations to [5, 6] and [13] supply notation, terminology, and algorithmic context rather than the target conclusion. The main weakness is that the critical sixty-step loop is asserted rather than exhibited or machine-certified ('The reader is encouraged to try this'), and Theorem 3.6 is proved for one case with 'other steps following similarly.' These are verification/rigor concerns, not circularity: the derivation does not assume the conclusion it claims to establish. No step in the paper's chain of reasoning is equivalent by construction to its own input, so no circular step is present.

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

No numerical parameters are fitted; the result is an explicit combinatorial example. The load-bearing axioms are the representation of Spencer-Brown's algorithm, the unproved normalization of the starting coloring, the unproved generalization in Theorem 3.6, and the interpretation of 'polar' as rotational symmetry.

assumptions (5)

domain assumption The parity-pass algorithm is correctly represented by the five transformations in Figure 3 and the color-path rules of Section 3.
The counterexample's looping behavior is judged against this representation, taken from [13,15]; any distortion of the algorithm would change the claim.
ad hoc to paper A 1-deficient coloring around the pentagon can be taken, up to color permutation, to match the top-left configuration of Figure 3; proof omitted in Remark 3.1.
The authors state 'We omit the proof here but assume this local coloring as the starting point.' This is an admitted unproved premise of the algorithmic framework, though not of the explicit counterexample run.
ad hoc to paper Theorem 3.6: for every step of the parity pass, presence of configuration C implies existence of the alternating path required for the next step; only one case is shown, with 'other steps following similarly.'
The generalized theorem is asserted without a complete proof; it underlies the claim that the loop follows the prescribed algorithm rather than stalling.
standard math Graph G has no nontrivial rotational symmetry because the pentagonal and octagonal faces at opposite poles force any rotation order to divide both 5 and 8.
Used to establish that the chosen pentagon is non-polar under the rotational-symmetry reading.
domain assumption 'Polar' is interpreted as rotational (radial) symmetry; the authors' Remark 4.1 acknowledges reflection symmetry is retained and provides a variant without it.
If 'polar' included reflection symmetry, the main example would not qualify; the authors address this by giving Example 4.2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A counterexample for the polar conjecture of Spencer-Brown." pith.science (2026). https://pith.science/paper/7ITBXLVL

@misc{pith2026260722398,
  author       = {Pith},
  title        = {Pith review of: A counterexample for the polar conjecture of Spencer-Brown},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7ITBXLVL}},
  note         = {Machine review of arXiv:2607.22398}
}

read the original abstract

In 1976, George Spencer-Brown announced a proof of the four color theorem, using operations on Tait colorings for trivalent plane graphs. In subsequent work he formulated these operations in terms of an algorithm that he called a parity-pass and claimed that when the parity pass algorithm is performed on a non-polar pentagon region, it necessarily terminates in an edge coloring that is extendable to the entire graph. We provide here a counterexample to show that this claim is false. We then raise questions related to the existence of this sort of counterexample.

Figures

Figures reproduced from arXiv: 2607.22398 by the authors.

**Figure 1.** The relationship between formations and 3-face colorings of a ribbon graph. The underlying trivalent graph, is shown in grey [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

**Figure 2.** The 3-edge coloring associated to the 3-face coloring in [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** One complete loop of the parity pass algorithm. In each diagram of [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (11 more)

**Figure 4.** Figure 4: A 1-deficient coloring of the Errera graph, which exhibits nonterminating behavior under the parity pass algorithm. We next introduce a configuration that will be of use in analyzing the graph at each stage of the parity pass algorithm. 3.1. A bad configuration [PITH…

**Figure 5.** Figure 5: A bad configuration, C. Theorem 3.3. Let Γ be a plane graph with a pentagonal face and a 1-deficient 3-edge coloring. Suppose that the coloring outside of the pentagon matches configuration C (see [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

**Figure 6.** Figure 6: A double switch allows the coloring to extend. However, in configuration C the red and green highlighted paths, representing alternating two-color paths, share at least one blue edge, as indicated by the crossing in [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

**Figure 7.** Figure 7: Configuration C on the domain of step C of the parity pass. or until the original 1-deficient 3-edge coloring (for the entire graph) is reached. For our counterexample, this latter state is what we will see. 4. The Counterexample In [PITH_FULL_IMAGE:figures/full_fig_p…

**Figure 8.** Figure 8: The graph G for our counterexample. Remark 4.1. The graph G exhibits reflection symmetry, but the Polar Conjecture requires rotational symmetry, not merely reflective symmetry. For a counterexample that lacks reflective symmetry as well, see Example 4.2 [PITH_FULL_IM…

**Figure 9.** Figure 9: One complete parity pass on the counterexample. The ability to construct non-symmetric graphs where configuration C (see [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]

**Figure 10.** Figure 10: The configuration C before and after one complete parity pass [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]

**Figure 11.** Figure 11: A counterexample without a line of symmetry [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]

**Figure 12.** Figure 12: An example where the parity pass loops, but terminates. An even stronger counterexample to the polar conjecture would require the above property for all pentagonal faces; otherwise, the parity pass applied to a pentagon without looping [PITH_FULL_IMAGE:figures/full_f…

**Figure 13.** Figure 13: The parity mill ϕ1. Suppose that one is equipped with the parity pass in [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]

**Figure 14.** Figure 14: The parity mill ϕ2. Question 5.3. Do there exist non-symmetric plane graphs with a non-polar pentagon for which all three algorithms, the parity pass, and the two parity mills, exhibit looping behavior? These algorithms are based on moves from the impasse group of [14…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

18 extracted references · 4 linked inside Pith

[1]

Appel and W

K. Appel and W. Haken,Every Planar Map is Four Colorable Part I. Discharging, Illinois J. Math.21 (1977), 429-490

1977
[2]

Appel, W

K. Appel, W. Haken, and J. Koch,Every Planar Map is Four Colorable Part II. Reducibility, Illinois J. Math.21(1977), 491-567

1977
[3]

Baldridge, L

S. Baldridge, L. Kauffman, and W. Rushworth,On ribbon graphs and virtual links, European Journal of Combinatorics103, June 2022, doi: 10.1016/j.ejc.2022.103520, arXiv: 2010.04238

arXiv 2022
[4]

Baldridge, A

S. Baldridge, A. Lowrance, and B. McCarty,The 2-factor polynomial detects even perfect match- ings, The Electronic Journal of Combinatorics27(2020), no. 2, P2.27, 16 pp. doi: 10.37236/9214, arXiv:1812.10346

arXiv 2020
[5]

Baldridge and B

S. Baldridge and B. McCarty,A topological quantum field theory approach to graph coloring, arXiv:2303.12010

arXiv
[6]

Baldridge and B

S. Baldridge and B. McCarty,A new way to prove configuration reducibility using gauge theory, arXiv:2412.18558

arXiv
[7]

Baldridge and B

S. Baldridge and B. McCarty, New relations for the Penrose polynomial, in preparation
[8]

Baldridge, L

S. Baldridge, L. Kauffman, and B. McCarty,A State Sum for the Total Face Color Polynomial, Journal of Graph Theory, 24 March 2025, https://doi.org/10.1002/jgt.23239

doi:10.1002/jgt.23239 2025

Show all 18 references

[9]

Errera, A., Du Coloriage des Cartes et de Quelques Questions d’Analysis Situs, Thesis, Gauthier-Villars, Paris, 1921, 66 pp

1921
[10]

H.,A state calculus for graph coloring, Illinois Journal of Mathematics, Volume 60, Number 1, Spring 2016, pp 251-271 S 0019-2082 (Special Issue celebrating Wolfgang Haken)

Kauffman, L. H.,A state calculus for graph coloring, Illinois Journal of Mathematics, Volume 60, Number 1, Spring 2016, pp 251-271 S 0019-2082 (Special Issue celebrating Wolfgang Haken)

2016
[11]

H.Map coloring and the vector cross product, J

Kauffman, L. H.Map coloring and the vector cross product, J. Comin. Theory B, 48 (2) (1990) 145-154

1990
[12]

Map Reformulation,

Kauffman, L. H. “Map Reformulation,” Princelet Edition No. 30: Princelet Editions: London (25 Princelet Street) and Zurich (1986). 14 SCOTT BALDRIDGE, LOUIS H. KAUFFMAN, AND BEN MCCARTY

1986
[13]

H.,Reformulating the map color theorem, Discrete Mathematics 302 (2005) 145-172, arXiv:math/0112266

Kauffman, L. H.,Reformulating the map color theorem, Discrete Mathematics 302 (2005) 145-172, arXiv:math/0112266

2005 arXiv
[14]

A group of operations on a partially colored map, Bull

Kittell, I. A group of operations on a partially colored map, Bull. Amer. Math. Soc. 41(6): 407-413 (June 1935)

1935
[15]

Spencer-Brown, G.Laws of Form, Bohmeier Verlag, 6th ed. 2015

2015
[16]

Spencer-Brown, G.Cast and formation properties of maps, MS 734, Royal Society Library, 177 pp, 1980

1980
[17]

Spencer-Brown, G.Some notes on the parity pass, unpublished, 1999

1999
[18]

Spencer-Brown, G.A direct proof of the map theorem, unpublished, 2000. Department of Mathematics, Louisiana State University, Baton Rouge, LA Email address:baldridge@math.lsu.edu Department of Mathematics, Statistics and Computer Science, 851 South Morgan Street, University of...

2000

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [1]

Appel and W

K. Appel and W. Haken,Every Planar Map is Four Colorable Part I. Discharging, Illinois J. Math.21 (1977), 429-490

1977

[2] [2]

Appel, W

K. Appel, W. Haken, and J. Koch,Every Planar Map is Four Colorable Part II. Reducibility, Illinois J. Math.21(1977), 491-567

1977

[3] [3]

Baldridge, L

S. Baldridge, L. Kauffman, and W. Rushworth,On ribbon graphs and virtual links, European Journal of Combinatorics103, June 2022, doi: 10.1016/j.ejc.2022.103520, arXiv: 2010.04238

arXiv 2022

[4] [4]

Baldridge, A

S. Baldridge, A. Lowrance, and B. McCarty,The 2-factor polynomial detects even perfect match- ings, The Electronic Journal of Combinatorics27(2020), no. 2, P2.27, 16 pp. doi: 10.37236/9214, arXiv:1812.10346

arXiv 2020

[5] [5]

Baldridge and B

S. Baldridge and B. McCarty,A topological quantum field theory approach to graph coloring, arXiv:2303.12010

arXiv

[6] [6]

Baldridge and B

S. Baldridge and B. McCarty,A new way to prove configuration reducibility using gauge theory, arXiv:2412.18558

arXiv

[7] [7]

Baldridge and B

S. Baldridge and B. McCarty, New relations for the Penrose polynomial, in preparation

[8] [8]

Baldridge, L

S. Baldridge, L. Kauffman, and B. McCarty,A State Sum for the Total Face Color Polynomial, Journal of Graph Theory, 24 March 2025, https://doi.org/10.1002/jgt.23239

doi:10.1002/jgt.23239 2025